FACTORS INFLUENCING RESEARCH PRODUCTIVITY AMONG ACADEMIC STAFF IN SELECTED UNIVERSITIES IN KENYA.

By

Joash A. Migosi

A Thesis Submitted to the Faculty of Education in Partial Fulfillment for the Requirements for Degree of Doctor of Philosophy Department of Postgraduate Studies in Education

Faculty of Education

The Catholic University of Eastern Africa

September, 2009

DECLARATION

I, Joash Abere Migosi, the undersigned, declare that this thesis in its form and nature, organization and content is a fruit of my personal effort. To the best of my knowledge, it has never been submitted for academic credit in any other University.

________________________ Sept, 2009 Joash Abere Migosi ED/PhD/037/06/07

Date:

20th

This thesis has been submitted with our approval as University supervisors;

_________________________ _________________________ Prof. P. Ogula Professor of Education, Catholic University of East Africa

Date

_________________________ _________________________ Rev. Dr. M. Kawasonga Senior Lecturer, Faculty of Education, Catholic University of East Africa Dedication

Date

I dedicate this work to my parents, Mzee Zedekiah Migosi Isoe and Mama Josephine Moraa, who, in spite of their limitations, found it wise to take me to school.

A bstract The current academic climate in higher education in Kenya threatens the Kenyan universities'

ability to sustain the conditions support that research

productivity, teaching and service to

community. Increased demands government private funding, on and a

deteriorating physical infrastructure,

increased pressure on undergraduate programs, expansion university strategies

and general economic climate in the country have raised concerns about the continued capacity universities maintain of to teaching,

research productivity and service to the community. situation This dictates

deliberate evaluation to be made in the scientific research

ii

arena at all times. This study examine sought to the

following; factors that influence productivity, research the

attitude of academic staff to research and publishing academic possible enhance productivity. among staff ways and to

research Survey

research method and document analysis

were employed in this study. questionnaire The was

iii

used

to

collect

information from 277 (70.2% 29.8% university male and

female) academic

staff and 17 heads of departments drawn

from 11 public and private universities. 15) do

SPSS (Version was used to

descriptive statistics, cross tabulations. testing out

Hypothesis was through carried

ANOVA.

Factor analysis was used for data

iv

reduction, identification description major influencing of and the

factors research

productivity as noted by respondents. analysis

Document

dwelt on the analysis of the sampled

research policies of the universities. selected The

results obtained from this study indicate that the research

productivity index for the universities in

Kenya is low. Most interesting conclusion indicated Research that content

knowledge and self motivation were the key factors that on

influenced individual researchers productivity. were

Others for

resources

research, equipments and availability of

technology. There was found to be significant relationship age group, between highest

vi

degree individual and

obtained, university degree

highest

obtained and attitude towards research and publishing. The main recommendation made by this study is for the development and application of and

national

institutional research policies to guide and manage research in this country.

vii

Acknowledgements This work

will not be complete without the mentioning special

contributions made by numerous persons

towards its successful completion. First and foremost I must the gesture

acknowledge generous

extended to me by my employer, Ministry Education of the Higher Science

and Technology for having awarded me a

viii

full

government to

scholarship

undertake this study. I am indeed grateful to the Permanent Prof.

Secretary

Chrispus Kiamba for this generous gesture. Were it not for this contribution, I am not sure whether this

work could have been completed successfully. I appreciate guidance also the and

supervision extended

ix

to me by my two supervisors, Prof. P. Ogula and Rev. Dr. M. Kawasonga. They have, in various ways helped shape this

work to its present stage. appreciate I also other

members of staff from the faculty at of the

education

Catholic University of Eastern Africa. They helped in guiding this study from the onset; these Akala, include; Drs.

Ammanuel,

Githui, Prof.

Kanga Onsongo.

and I the

acknowledge

assistance I received particularly in

research methodology from Mr. Makinda, Mr. Kiarie, and Prof OConnor Department Psychology University of British Columbia Okanagan who assisted me with materials interpretations and on of

Principal Component Analysis. My

xi

classmates supported especially Kangethe,

also me Liz, Ogoti,

Jared and others. Mr. P. Nyaswa willingly agreed to go through the final draft of this work and gave

constructive criticisms. Am

grateful to Rev. Prof. Majawa who

introduced me to the truths cosmotheandrian education for of

transformative living

xii

and destiny

meaningful during the

earlier part of this course. Indeed this

has had a great impact in my life. Mr. and Mrs. Elkana Osinde, my neighbors, gave me all support. I forget will Mr. not

Moturi

Wisley who has done most of the typing of this work and doubled up as a research

assistant in this work. I am indeed indebted to hundreds of

xiii

academic staff from 11 universities across the country who

generously

accepted

to give part of their time to give responses to the questionnaires, HODs, chancellors selected of Vice the

universities

and all my research assistants. To them all I say thank you. My immediate and current Director, Dr. Mokabi and Dr. E. Mwangi respectively gave me easy time

xiv

and encouraged me to finish thanks workmates Directorate Research Management Development, particularly officemate Omoto. My provided family invaluable Dr. my W. and up. to at Many my the of

support for the three years I was involved in this study. Regular travels to and from Eldoret,

xv

notwithstanding

the

condition of the road then, was indeed a challenge to all of us. I appreciate particularly the

continuous urge from my wife Zipporah,

daughters Joy, Lewin; sons Allan and

Emmanuel to finish up. I sincerely thank them for their voices

supporting

which cheered me on. Last but not least, I appreciate the many graces from the

xvi

everlasting one. For, were it not for Him, all could have been in vain.

xvii

Table of Contents
Table of Contents .............................xviii CHAPTER ONE. .xlix INTRODUCTION ..............................xlix

1.1 Background to the Problem ...........xlix 1.1.2 Backgrou nd of Universit y Education in Kenya .............liv 1.1.3 Challenge s Facing Universit y Education

xviii

in Kenya ...........lviii 1.2 The Statement of the Problem ..........lxvii 1.3 Research Questions ...........lxix 1.4 Hypotheses ............lxx 1.5 Significance of Study ..........lxxii 1.6 Scope and delimitati on of the study .........lxxiii 1.7 Theoretical Framewo rk.....lxxiv 1.7.1 Expectan cy Theory of Motivatio n - Victor Vroom .........lxxiv

xix

1.7.2 Rationale for Motivatio n Theories .........lxxvi 1.8 Conceptual Framewo rk....lxxxv 1.9 Operational Definition of Terms .....lxxxviii 1.10 Organizat ion of the study....xc
CHAPTER TWO...xci LITERATURE REVIEW................xci

2. Introduction ............xci 2.1 Research Productiv ity and Theories ...........xcii 2.2 Research Productiv ity and

xx

Publicatio ns............c 2.3 Theories of Motivatio n..........cxi 2.4 Determin ants of Research Productiv ity ...cxxiii 2.4.1 Research Productiv ity Measure ment cxxvi 2.4.2 Webomet rics...cxxx 2.4.3 Research Productiv ity and Type of Institutio n.....cxxxii 2.5 Factors that Influence Research Productiv ity .....cxlii

xxi

2.5.1 Factors associated with Personal Career Developm ent....clxiv 2.6 Research Productiv ity and Academic Discipline s........clxvi 2.7 Research Productiv ity and Technolog y Transfer/ Patents .......clxxxi 2.8 Research Productiv ity and Teaching Effectiven ess. clxxxv 2.9 Summary of Literatur

xxii

e Review .....clxxxvi
CHAPTER THREE ......................clxxxviii RESEARCH DESIGN AND METHODOLOGY ......................clxxxviii

3 Introduction ...clxxxviii 3.1 Research Design ...clxxxviii 3.2 Target Populatio n...clxxxix 3.3 Sample and Sampling Procedure s....clxxxix 3.4 Description of Research Instrume nts.......cxc 3.4.1 The Questionn aire.....cxc 3.4.2 Questionn aire for Lecturers

xxiii

(Appendi x 1)....cxci 3.4.3 Questionn aire for Heads of Departme nts (Appendi x I1)..cxcii 3.5 Piloting/P re-Testing .........cxciv 3.6 Validity and Reliabilit y of Instrume nts. . .cxciv 3.6.1 Validity .........cxciv 3.6.2 Reliability ..........cxcv 3.7 Description of Data Collection Procedure s......cxcvii 3.8 Description of Data Analysis

xxiv

Procedure s.....cxcviii
CHAPTER FOUR. cci PRESENTATION AND DISCUSSION OF THE FINDINGS ................................cci

4.1 Introducti on........cci 4.2 Demogra phic characteri stics of participa nts.......cci 4.2.1 Academic staffs demograp hic informati on........cci 4.2.2 Head of departme nts demograp hic informati on....ccviii 4.3 Nature of Research

xxv

Productiv ity among public and private universiti es in Kenya between 2004-2008 .........ccxii 4.3.1 Average publicatio n output by university between 2004 2008 ccxiii 4.3.2 Test of Hypothes es....ccxxx 4.4 Lecturers views on individual and Institutio nal factors influencin g research

xxvi

productiv ity ...ccxlii 4.4.1 Lecturers views on individual factors influencin g research productiv ity ...ccxlii 4.4.2 Lecturers views on Institutio nal factors influencin g research productiv ity . .ccxliv 4.4.3 Heads of Departme nts views on Individua l factors influencin g research productiv ity . ccxlvi 4.4.4 Heads of Departme

xxvii

nts views on Institutio nal factors influencin g research productiv ity . ccxlvii 4.4.5 Principal Compone nt Analysis ...........ccli 4.4.5.1 Selecting the Variables for PCA .........ccliii 4.4.5.2 Further Interpret ation of the PCA ...........cclx 4.5 Attitudes of academic staff on Research and Publicatio

xxviii

ns in their institution s ......cclxv 4.5.1 Lecturers total attitude scores on research and publicatio ns....cclxvi 4.5.2 University Academic Staffs Mean Attitude Score towards Research and Publishin g.....cclxix 4.5.2.1 Hypothes es testing .........cclxx 4.6 Lecturers perceived factors hindering research

xxix

productiv ity . . .cclxxxiv 4.6.1 Enhance ment of research productiv ity.. .ccxcii 4.6.1.1 Academic staff views on possible solutions to problems hindering research productiv ity. .ccxciii 4.6.1.2 Heads of Departme nts views on possible solutions to problems hindering research

xxx

productiv ity.......ccc 4.7 Document Analysis. ..........cccv 4.7.1 Summary of Document Analysis .......cccxix
CHAPTER FIVE ...........................cccxx

SUMMARY, CONCLU SIONS AND RECOM MENDAT IONS ........cccxx 5.1 Summary ........cccxx 5.2 Conclusions .....cccxxiii 5.3 Recomme ndations .....cccxxvi 5.4 Suggestions for further

xxxi

studies .....cccxxix
REFERENCES ........................cccxxxi APPENDIX I . ccclxvi APPENDIX II ccclxxv APPENDIX III ....................ccclxxxiii APPENDIX IV ......................ccclxxxv .....................ccclxxxvi

xxxii

Table1.1: Summary of related studies on research productivity employing the expectancy model. ............................lxxix Table 2.1: Illustration of Lotkas Law . .xcvii Table 4.1: Frequency Distribution of academic staff by Gender...................ccii Table 4.2: Distribution of academic staff by type of university.........cciii Table 4.3: Distribution of academic staff by Age group....................cciv Table 4.4: Distribution of academic staff by Position Held in Their respective Universities..........ccvi Table 4.5: Distribution of academic staff by years since last highest degree was obtained ............ccviii Table 4.6: Distribution of Heads of departments by gender ..................ccix

xxxiii

Table 4.7: Distribution of Heads of departments by type of university ccix Table 4.9: Distribution of heads of department by highest degree obtained ...............ccxi Table 4.10: Distribution of heads of departments by years since last highest degree was obtained ..............ccxii The information presented below is a calculation of the average publication output of the academic staff in Kenya. The calculation was done by dividing the total publications from self reported data in this study by the number of academic staff in the sample. This was done for all the universities under study in each of the five years under study...................ccxiii Table 4.11: Average publication output By

xxxiv

University 2004-2008 ...........................ccxiii Table 4.12: Frequency distribution of those who Published by rank..................ccxviii Table 4.13 Frequency distribution of those who did not publish by rank...............ccxix Table 4.14: Average publication output by gender 2004-2008 ............................ccxxi Table 4.15: Average publication output By Rank 2004-2008 .........................ccxxiii Table 4.16: Average publication output by Age group 2004-2008 ..........................ccxxvi Table 4.17: Average publication output By University Type 2004-2008.......ccxxvii Table 4.18: Average publication output by highest degree obtained 2004-2008 ..........................ccxxix Table 4.19: One way ANOVA test of difference in the mean score of academic staffs type of

xxxv

university and research productivity. ..........................ccxxxi Table 4.20: One way ANOVA test of difference in the mean score of academic staffs age group and research productivity. .......................ccxxxiii Table 4.21: One way ANOVA test of difference in the mean score of academic staffs gender and research productivity. ........................ccxxxiv Table 4.22: One way ANOVA test of difference in the mean score of academic staffs rank and research productivity. .........................ccxxxv Table 4.23: One way ANOVA test of difference in the mean score of academic staffs highest degree obtained and research productivity. ccxxxviii Table 4.25: One way ANOVA test of difference in the mean score of academic staffs university and

xxxvi

research productivity ..............................ccxl Table 4.26: Individual Factors influencing research productivity ..........................ccxliii Table 4.27: Institutional Factors Influencing Research Productivity......ccxliv Table 4.28: Individual factors Influencing research productivity as perceived by Heads of Departments ccxlvi Table 4.29: Heads of Departments attitudes on institutional factors affecting academic staff research productivity....ccxlviii Table 4.30: Rotated Component Matrix (a) ..............................cclv Table 4.31: Correlation Matrix and Rotation.......cclvii Table 4.32: KMO and Bartlett's Test ...........................cclvii Table 4.33: Frequency distribution of lecturers attitude scores on Research and Publications in

xxxvii

their institutions .........................cclxvii Table 4.34: Academic Staff gender mean attitude score ....cclxix Table 4.35: ANOVA test of difference in the gender mean attitude scores towards research and publishing ........cclxxi Table 4.36: Academic Staffs types of university mean attitude score ...cclxxii Table 4.37: ANOVA test of difference in academic staffs type of university mean attitude scores towards research and publishing ......cclxxiii Table 4.38: Academic Staffs age group mean attitude score ........................cclxxiv Table 4.39: ANOVA test of difference in academic staffs age group mean attitude scores towards research and publishing ......cclxxvi Table 4.40: Academic Staffs rank mean attitude score .cclxxvii

xxxviii

Table 4.41: ANOVA test of difference in academic staffs rank mean attitude scores towards research and publishing .....cclxxvii Table 4.42: Academic Staffs highest degree mean attitude score ........................cclxxix Table 4.43: ANOVA test of difference in academic staffs highest degree mean attitude scores towards research and publishing .......cclxxx Table 4.44: Academic Staffs years since highest degree was obtained mean attitude score ..cclxxxi Table 4.45: ANOVA test of difference in academic staffs years since highest degree was obtained mean attitude scores towards research and publishing .....cclxxxii Table 4.46: Academic Staffs University mean attitude score .......................cclxxxii Table 4.47: ANOVA test of difference in

xxxix

academic staffs university mean attitude scores towards research and publishing ....cclxxxiii Table 4.48: Academic staff views on problems hindering research productivity ........................cclxxxv Table 4.49: Heads of Department views on problems hindering research productivity ............................ccxci Table 4.50: Academic staff views on possible solutions to problems hindering research productivity ..........................ccxciv Table 4.51: Heads of Department views on possible solutions for problems hindering research productivity ...............................ccc Table 4.52: Research Career Development Framework.......cccxvi

xl

List of Figures

Figure 1.1: Factors Responsible for Faculty Research Productivity......lxxxvi Figure 2.1: Factors Responsible for Faculty Research Productivity............cx Figure 4.1: Distribution of academic staff by Highest Degree Obtained .............ccvii Figure 4.2: Distribution of heads of departments by

xli

Position held in university...............ccx Figure 4.3: Scree plot for principal component analysis ..........................cclviii

xlii

List of Abbreviations/Acron yms

AAU Association of African Universities AAUP American Association of University Professors ANOVA Analysis of Variance ARIPOAfrica Regional Intellectual Property Organization

xliii

ASPA American Society for Public Administration AUB American University of Beirut CFA Confirmatory Factor Analysis CLS Clinical Laboratory Science CUEA Catholic University of Eastern Africa

xliv

FRP Faculty Research Productivity HODs of Department IADR International Association for Dental Research IPR Intellectual Property Rights JIF Impact Factor KEPSA Private Sector Alliance Kenya Journal Heads

xlv

KIPI

Kenya

Industrial Property Institute KMO Mayer-Olkin NCST National Council for Science and Technology PAA Population Association of America PAR Administration Review Public Kaiser-

xlvi

PCA Principal Component Analysis PhD of Philosophy POM Production and Operations Management PPPs Private Doctor

Public Partnerships PSDS Private

Sector Development Strategy R&D Research & Development

xlvii

SEM Structural Equation Modeling SPSS Statistical Package for Social Sciences UASU University Academic Staff Union UNESCO United

Nations Education Scientific and Cultural Organization UON University of Nairobi

xlviii

WIPO

World

Intellectual Property Organization

xlix

CHAPTER ONE INTRODUCTION 1.1 Background to the Problem Governments expect universities to become more efficient and effective in teaching, research and community service. However, there appears to be many obstructions to research productivity that in turn cause low levels of research outcomes (Lertputtarak, 2008). In Kenya, for example, the ranking of local Universities has nosedived. This has been due to the recent innovation of module II in higher education, massification of higher education and aggressive expansion strategies employed by various universities. This has resulted in possibilities of imbalance between available time for teaching and research roles of the academic staff in Universities. Brewer, Douglas, Facer, and O'Toole (1999) describes the fears that Frederick C. Mosher (a public administration scholar) had in the mid 1950s. Mosher had complained that scholars of public administration were not doing enough to advance knowledge in the field. He maintained that too little research had been performed, the stimulus for research effort was inadequate, and research output was not meeting the needs of society. According to Mosher, the problem was more than academic (pun intended). He felt that research was the first step in improved practice, and that the real tragedy of poor research performance was the human suffering that could be alleviated if public administrators were provided with better knowledge.

Mosher argued that the close relationship between scholarship and practice was a strength of public administration, and he encouraged scholars and practitioners to work together to solve the "research problem". Furthermore, he recognized that the American Society for Public Administration (ASPA) was one of the few professional societies in any field with close connections to scholarship and practice, and he urged ASPA to assert a leadership role and make the advancement of knowledge one of its primary objectives. As a result, the Public Administration Review (PAR) began providing a forum for the debate over research, and this forum has been sustained by lively commentary for more than forty (40) years. Unfortunately, the research problem has not been resolved. Knowledge production by public administration scholars continues to be criticized as insufficient to meet the field's needs (Brewer et al, 1999). Clearly this is a strong support for research development in areas that are perceived not to be research oriented. Due to the world-wide economic and social imperatives, universities in all countries are engaged in a significant reconceptualisation of their public roles. Geiger (1986) notes that the higher education sector in the twenty-first century is very different from that of the late nineteenth and twentieth centurys. Universities now perform important roles as the guardians of public knowledge. They are an important part of the modern capitalist engine and are recognized as generators of public scientific and technological knowledge. Clearly

li

this shows the critical role universities are playing in national development through research and efforts must be made not to deviate from this noble responsibility. African Governments are committed to the development of university education on the premise that higher education is a most sensitive and productive area of investment. It is politically and socially sensitive in that governments need both highly-trained people and top-quality research to formulate policies, plan programmes, and implement projects that are essential to national development (AAU, 1987). This is a reaffirmation of the urgent need to develop university education in Africa so as to cater for the divergent needs of society. The roles and responsibilities of college and university faculty members are closely tied to the central functions of higher education. One primary formal description of these functions was contained in the 1915 "Declaration of Principles" formulated by a representative committee of faculty members including members of the American Association of University Professors (AAUP). According to the Declaration, the functions of colleges and universities are to promote inquiry and advance the sum of human knowledge, to provide general instruction to the students, and to develop experts for various branches of the public service (Joughin, 1969). These roles can be summed up as teaching, research and community service.

lii

The teaching, research, and community service roles of faculty members overlap conceptually and practically. For example, instruction in a particular discipline or skill yields a community service in the form of educated or appropriately trained persons, and outreach to a farmer or small business owner may lead to an applied research project undertaken by the faculty member. Some attempts have been made to validate the various forms of faculty work and unify them conceptually. Perhaps the most famous recent model has been the American educator and government official Ernest Boyer's (1990) stipulation of discovery, application, integration, and teaching as separate but related forms of scholarship. Among other outcomes, these models address concerns regarding the implicit hierarchy that grants the most prestige to research and the least to community service. (Education Encyclopedia, 2008) In almost all African countries, public universities receive financial assistance mainly from the state (Psacharopoulos, 1982). The result is that the level of higher education activities in a country has for long depended on the soundness of national economic performance. From the 1980s, most African countries experienced financial constraints due to poor economic performance and rapid population growth, added to the need to provide other basic services like primary education, food, health and shelter. University education, therefore, has faced severe competition from other sectors for limited

liii

government funds (Psacharopoulos, 1985; World Bank, 1988). In recent years governments and international donors have challenged universities in Africa to justify their existence and their claim on the massive public funds allocated to them. (Fuller, 1992; Psacharopoulos; 1985; World Bank, 1988) The Government of South Africa (2003) for example, through the Department of Education came up with a document entitled Policy for Measurement of Research Output of Public Higher Education Institutions The development of this policy was driven by the imperatives for transformation of the higher education system as part of the strategic objective envisioned by the National Plan for Higher Education. This policy aims to sustain research strengths and promote research and other knowledge outputs required to meet national development needs. (National Plan for Higher Education, 2003; 70) The purpose of this policy was to encourage research productivity by rewarding quality research output at public higher education institutions. However, the policy was not intended to measure all output, but to enhance productivity by recognising the major types of research output produced by higher education institutions and further use appropriate proxies to determine the quality of such output. It is important to note that governments are realising the crucial role being played by research. Kenya can clearly borrow a leaf from the Government of South Africa by determining research productivity of its academic

liv

researchers and establish elaborate reward scheme for its researchers both at the university and at the research institutes.

1.1.2 Background of University Education in Kenya The idea of an institution for higher learning in Kenya goes back to 1947 when the Kenya Government drew up a plan for the establishment of a technical and commercial institute in Nairobi. By 1949, this plan had grown into an East African concept aimed at providing higher technical education for the region. In September 1951, a Royal Charter was issued to the Royal Technical College of East Africa and the foundation stone of the college was laid in April 1952 (UON, 2009) Soon after the arrival of students at the college, the pattern of higher education in East Africa came under scrutiny. Through the recommendation of a working party formed in 1958, chaired by the Vice-Chancellor of the University of London, Sir John Lockwood, the Royal Technical College of East Africa was transformed. On 25th June 1961, the College became the second University College in East Africa, under the name "Royal College Nairobi." (Nafukho, 1999; UON, 2009) The Royal College Nairobi was renamed "University College, Nairobi" on 20th May 1964. On the attainment of "University College" status, the institution prepared

lv

students for bachelor's degrees awarded by the University of London, while also continuing to offer college diploma programmes. The University College Nairobi, provided educational opportunities in this capacity until 1966 when it began preparing students exclusively for degrees of the University of East Africa, with the exception of the Department of Domestic Science (Abagi, 1999). With effect from July 1970, the University of East Africa was dissolved and the three East African countries set up their national Universities. This development saw the birth of the University of Nairobi set up by an Act of Parliament. Others were Makerere and Dar Es Salaam Universities for the break-up of the University of East Africa was partly due to ideological differences between the member states Uganda and Tanzania respectively. Kenyatta College, a teacher-training institution situated on the outskirts of Nairobi, became a constituent college of the University of Nairobi in 1972 and was elevated into a full -fledged university in 1985 (Abagi,1999; Nafukho,1999; UON, 2009). As years went by, the number of Kenyans seeking university education exceeded the capacity of the University of Nairobi. This led to the establishment of Moi University in 1984 as the second university in Kenya following the recommendations of the Mackay Report of the Presidential Working Party on the Second University in Kenya (Government of Kenya, 1981) which collected views from many people and found an overwhelming

lvi

support by Kenyans for the establishment of a second and technologically oriented university in the country. From then, university education in Kenya has expanded with a rise in student enrolments, expansion of universities, diversity of programmes and setting up of new universities and campuses. A previous agricultural college also gave way to Egerton University in 1988 (Nafukho, 1999). The most salient feature of university education expansion in Kenya has been rapid growth in the number of institutions and enrolments. The number of public universities increased from one in 1970 to seven in 2009 while the university colleges have grown to thirteen (13). Like its public counterpart, the private university sector in Kenya has also grown tremendously. The private and accredited universities stand at 21 though most of them concentrate in theological studies. The numbers of university students in Public and accredited universities stood at 118,300 (GOK, 2008).The University of Nairobi intends to hit the 50,000 mark in terms of enrolment by September 2009 (UON, 2009). These figures indicate an increase in enrolment from 91,541 students in 2004/2005 academic year to about 170,000 students in 2009 (CHE, 2009). With the Kenyan economy experiencing negative growth for most of the 1980s and 1990s, the Government of Kenya found itself no longer able to sustain its previous levels of financial support to the public universities. This situation arose at the same time as the growth in student numbers that resulted from both

lvii

the pressure exerted from the expanded lower levels of education as well as the fact that possession of higher education qualifications was becoming more highly regarded as a ticket to formal sector employment (Mwiria, 2007). Chege (2006) notes that the watershed of higher education in Africa was experienced at the establishment of colleges that were affiliates of the University of London for example, Makerere was giving external degrees of the University of London. Other institutions that were appendages of the University of London included University of Ibadan in Nigeria, University College of Ghana at Legon and Fourah Bay College at Freetown, Sierra Leone. The implication of this was that higher education institutions in Africa should maintain the same parity in instruction, teaching and culture as the University of London. With the exception of universities established in Cairo (970 in Egypt), Fourah Bay (1827 in Sierra Leone), Liberia (1862), and Omdurman (1912 in Sudan), most African universities were founded around independence in the mid-1960s and 70s, this is where the universities in East Africa fall. The new universities became a constituent part of the new institutions that were created to satisfy the aspirations of the new nations. Identified among the tasks of the new universities were the training of future leaders of thought in the professions, commerce and industry. Additionally, they were to train a highly educated

lviii

cadre of persons who would give leadership by passing on their education through formal and informal means. Simultaneous with the training of high-level human resources, was the need to prepare a number of social and economic research studies to serve as guidelines to the political leaders who were spearheading the reconstruction of institutions inherited from both the traditional and colonial eras.

1.1.3 Challenges Facing University Education in Kenya Kenya as a member of the international community is also having a good share of her challenges in the higher education sector. The rapid expansion of university education in Kenya has led to a number of challenges. According to UNESCO World Conference on Higher Education (1998), low funding from the exchequer, increased enrolment, limited access compared to the population level, increased enrolment without commensurate improvement in available facilities, gender inequality, and a low research capacity, are some of the problems facing universities in the region. These problems have led to fears that quality of education is in a downward trend in most of these universities.

Nyaigotti-Chacha (2004; 6) made the following observation about research in Kenya;

lix

Research is one of the core pillars of any university system; Publication of research findings in reputable journals is one of the ways in which these findings are widely disseminated to stakeholders. Studies show that research and publishing by faculty has sharply dropped over the last few years. Due to heavy teaching responsibilities, brought about by the rising student numbers, plus the need to moonlight so as to make some extra money to supplement the meager pay faculties are not keen on undertaking meaningful research and publishing their work.

This observation, by one of the scholars in Kenya, is an indication that all is not well in the research system at the University level in Kenya. There is need therefore to find the extent to which this has gone. Teferra, Altbachb (2004; 12) made the following observation on African scholarship, African higher education, at the beginning of the new millennium, faces unprecedented challenges. Not only is the demand for access unstoppable, especially in the context of Africa's traditionally low postsecondary attendance levels but higher education is recognized as a key

lx

force for modernization and development. Africa's academic institutions face obstacles in providing the education, research, and service needed if the continent is to advance.

This is indeed a clear observation that research function is also facing a crisis in Kenya and Africa at large and possible solutions should be sought to bring research back to its former status. Atwoli (2008) observed that expanding university education and taking higher education closer to the people in need of it is an important prerequisite for any country aspiring for industrial development. However, he notes that this should be done with the same degree of preparation. He notes that it is time our universities sat back and rethought this whole expansion strategy. There is need therefore to strike a balance between provision of higher education and quality issues. Muchie (2008) noted that research universities are critical levers, along with government and industry needed to shape a knowledge economy in any part of the world. The key question for Africa is how universities can be aligned to support economic development, the eradication of poverty and sustainable use of natural resources. Here research and knowledge become critical to making poverty history and preparing countries

lxi

to cope with disasters. Africa needs a strong pan-continental community of researchers to discover resourceful timely ways to deal with povertys many causes. This requires the development of strong research universities, institutions with a strong emphasis on graduate research, as opposed to undergraduate teaching, and where graduates are taught by lecturers who themselves are expanding the frontiers of knowledge. There is no doubt that the developing countries need research more than the developed world. Kelchtermans and Veugelers (2005) wondered on what makes someone a top researcher, why a substantial part of academics hardly ever publish anything, what factors explain differences in research productivity and the nature of the research system. He questioned why some top performers managed to sustain their high productivity level while others peak in scientific output only sporadically or never. This calls for considerable effort in understanding insights and dynamics in the factors that drive differences in research performance. This studys concentration on the factors that determine research productivity may illuminate some of the questions posed above. Some scholars (Gibbons 1998; Kennedy, 1997; Trow, 1996) argue that Universities are being challenged by other knowledge producers. Universities are no longer the sole producers of knowledge. Knowledge is now being produced by a variety of organizations like non governmental organizations, research organizations, business firms, and

lxii

government laboratories and even individual researchers. Whichever way one looks at this, there is a case that has been brought forward and puts the universities on the defensive. These concerns clearly demonstrate that Universities are at the crossroads and must do more to remain relevant in the present society. In Kenya just like other parts of the world, various research institutes, firms and even individuals are taking part in the generation of knowledge, so universities are being challenged. The universities, however, will not respond appropriately without data in the area of research management. Information has to be availed on the research output and this is what this study has tried to do. Kenya has been doing well in terms of research and publishing. Ngome (2003) observes that in the 1970s and early 1980s, the volume of research carried out at the University of Nairobi, the oldest and largest public university in the country was one of the highest in Africa. One of the key factors that stunted the growth of research in the Kenyan university system was lack of adequate research funds. The large portion of support (although inadequate) for postgraduate and staff training and research work came from donors and international organizations. Lack of adequate qualified researchers constituted the second major constraint to research expansion. The Government of Kenya recognizes that research and development plays a crucial role in wealth creation and enhancement of human development in the socio economic

lxiii

development of the country. The importance placed upon research by the Government of Kenya is stated in Sessional Paper Number 1 of 2005 p.85:

Research and development (R&D) is a means of creating wealth and enhancing human development and is a critical component of higher education and training. It also plays a vital role in industrial transformation, economic growth and poverty reduction. However, quality research requires sufficient funding, availability of highly trained research staff, adequate and appropriate facilities and equipment. For Kenya to meet her needs in R&D, there is need to give R&D priority in national development.

The strategies recorded in the above quoted Sessional Paper seek to strengthen research and development through: Increased investment in Research & Development (R & D), creation of a strong linkage between national goals, aspirations, linkages and research and wide dissemination of research findings for operational activities. Despite this, the government acknowledges that researchers are faced with various challenges which must be overcome (GoK, 2005). One major highlight in Kenyas National Strategy for University Education reform process is the emphasis on the creation

lxiv

of a culture of innovation through acquisition, creation and application of knowledge. In the strategy report, the strategic goal for quality and relevance of University is stated as: To improve quality and relevance of learning through research for socio-economic transformation of society (Kenya). (GoK, 2007). Kenyas current Mid Term Plan (MTP) of Vision 2030 P.102 states thus; The rapid increase in enrolments at all levels of education without commensurate increase in infrastructure and personnel has led to overstretched facilities, overcrowding in learning institutions and high student staff ratios. All these challenges have had a negative effect on the quality of education. In addition, the different curriculum has not kept pace with the demands of globalization. For instance, rapid expansion in the demand for University education has strained the existing facilities and adversely affected the teaching and learning, research productivity and the intellectual climate of universities as a whole. These challenges need resolution and elimination in order for universities to perform to the expected standards. Currently, this is a critical issue facing higher education institutions in Kenya, and the purpose of this research is to focus on the factors that have an influence on the research productivity of academic lecturers in universities in Kenya. Ouma

lxv

(2008) brings in a new concept of marketisation. He argues that universities in Kenya started enrolling full fee-paying students at a time when they were strained in terms of institutional capacity. There were no enough physical facilities, and most of those available were suffering decay following many years of neglect. They did not have enough teaching staff, a problem, which the marketisation agenda has made worse. These views suggest that rapid university expansion in one way or the other affect the core mandates of the university in Kenya. Insights in the factors that drive differences in research performance and its dynamics have important policy implications. Policy makers in Kenya have started assessing research performance (NCST, 2009; GOK 2009). The use of publications as instruments for evaluation of individual scientists within research institutes as well as for funding decisions for universities as a whole is becoming more widespread and gaining acceptance. Furthermore, the allocation of research funding is increasingly being driven by criteria of scientific excellence, resulting in a concentration of more funds in fewer hands. Yet, there are few academic studies done in Kenya on what drives research productivity. Publications are the major output of scientific research (Rennie, Yank, & Emanuel 1997). They are the most commonly used vehicles through which new scientific discoveries are conveyed to the rest of the world (Nelkin, 1998). They are also the principal

lxvi

currency for academic recognition and promotion for researchers in most westernized countries (Horton, 1998). Traditionally, the United States (US) has been leading the world in publication output (Stossel & Stossel 1990). However, with increased globalization of over the past two decades, other nations are gaining ground (Stossel, 1990; Nahrwold, Pereira, Dupuis, 1995). Despite this trend towards internationalization of research, there remain large variations in publication output among nations (Tompkins, Ko & Donovan 2001). The exact reason(s) for this variation are largely unknown. This study has employ books and journal publications as a surrogate for research productivity. It has centred on those factors that are influencing research output among Kenyan scholars. This study has shown that there are a number of factors that influence the variation in research output among scholars in Kenya. Kelchtermans et al (2005) notes that many empirical researches have recently emerged in the west that attempts to pin down the determinants of scientific productivity, both at the level of the individual researcher and institutional level. However few studies have combined both the researches publications output and factors that contribute to that output. This study explores factors that may explain why some faculty members are more productive than others. Some of these factors explain or predict increased productivity,

lxvii

educational administrators in Kenya might be able to implement policies to encourage and support higher levels of research productivity.

1.2 The Statement of the Problem The current academic climate in higher education in Kenya threatens the Kenyan universities ability to sustain the conditions that support research achievements. Increased demands on government and private funding, a deteriorating physical infrastructure, increased pressure on undergraduate programs, module II, university expansion strategies and general economic climate in the country have raised concerns about the continued capacity of universities to maintain teaching, research productivity and service to the state (Lertputtarak, 2008; Atwoli, 2008). UNESCO, (2006) has raised serious concerns over the nature of university education in the developing countries. It says that most universities are under immense pressure to increase their enrolment in order to meet the human resource development targets of their respective countries. This has led to teaching becoming their first priority and often their only pursuit. Also, because of scarce financial resources, they are unable to adequately equip and maintain their research facilities or replenish their libraries. In addition, they are unable to recruit or retain well-qualified faculty with strong research credentials who, for

lxviii

various reasons, prefer to move to developed countries (brain drain). Other pertinent issues include (a) how much of the research carried out in universities in developing countries is directly or indirectly relevant to the development needs of the country, and (b) how much of the findings gets effectively transmitted to the relevant users. These concerns need to be addressed urgently if the universities in the third world countries have to make an impact in society. In most developing countries universities are the main and often the only institutions to undertake research, and if these falter, knowledge production for the country as a whole will be seriously affected. Statistics show the very poor state of research output of many developing countries, and the most disadvantaged region is Sub-Saharan Africa (UNESCO, 2006). The slogan publish or perish is commonly used in universities in the west in quest for promotion (Mwamwenda, 1994). The Universities in developing world are also taking into consideration publications by lecturers as a requirement for upward mobility. Many of the academic staff in universities are not involved in productive research work while a few of them are. This is what (Creswell 1985) referred to as a puzzle why some faculties produce research year after year while others do not conduct any research at all.

lxix

In spite of all these concerns, and the demand by universities for academic staff to publish, there has never been any concern or understanding of the circumstances under which the academic staff operate. There have never been deliberate efforts to understand the problems that academic staff faces in their quest to publish. In this regard, there are very few studies done in Kenya to analyze the factors that influence research productivity in institutions of higher learning. The published literature in Kenya to date on the factors influencing research productivity among university academicians in Kenya is limited. Considering that many studies have been done in various countries of the world, particularly in the developed countries, it is therefore important that a study is conducted in a developing country to compare notes with those studies from other countries of the world. Therefore, this study sought to establish the factors that influence research productivity among academic staff in selected public and private universities in Kenya. 1.3 Research Questions The following research questions guided study; 1. What is the status of research productivity in selected public and private universities in Kenya?

lxx

2. What are the individual/institutional factors that hinder/encourage research productivity among academic staff in selected public and private universities in Kenya? 3. What is the attitude of university academic staff in Kenya towards research and publishing? 4. How can research productivity among academic staff in selected public and private universities in Kenya be enhanced? 1.4 Hypotheses
1. There is a significant relationship between academic staffs type of university and

research productivity.
2. There is a significant relationship between academic staffs age groups and research

productivity.
3. There is a significant relationship between academic staffs gender and research

productivity.
4. There is a significant relationship between academic staffs rank and research

productivity.

lxxi

5. There is a significant relationship between academic staffs highest degree obtained

and research productivity.

6. There is a significant relationship between years since last highest degree was

obtained and research productivity.

7. There is a significant relationship between academic staffs university and research

productivity. 8. There is a significant relationship between academic staff s gender and attitudes towards research and publishing. 9. There is a significant relationship between academic staff s type of university and attitudes towards research and publishing. 10. There is a significant relationship between academic staff s age groups and attitudes towards research and publishing. 11. There is a significant relationship between academic staff s rank and attitudes towards research and publishing. 12. There is a significant relationship between academic staff s highest degree obtained and attitudes towards research and publishing.

lxxii

13. There is a significant relationship between academic staff s years since last highest degree was obtained and attitudes towards research and publishing. 14. There is a significant relationship between academic staff s university and attitudes towards research and publishing. 1.5 Significance of Study The purpose of this study was intended to stimulate thought, and to recommend specific actions about management of research output in Kenya. The general aim of the study is to provide information that can assist in the design, development and formulation of institutional research policies in the changing global situation, and in particular to highlight those factors that should be emphasized in order to further encourage academic lecturers to increase their research productivity. Reward differentials, promotions and rankings among men and women in higher education are outstanding issues that have not been fully addressed and results of this study can be useful in addressing this problem. The results of this study may provide important information about how universities, research funding organisations and even government can implement policies or develop strategies to foster lecturers creative research productivity and other scholarly activities.

lxxiii

According to Henry and Burch (1974), most decision makers continue to use published research as the primary indicator of academic quality. Similar to corporations who measure "success" by bottom line profits or market share, academic institutions use research productivity as the index to their overall reputation and as a means to strengthen their national and international stature. Published research forms the best available criterion for evaluating the quality and quantity of individual faculty members and of their departments and institutions. It is therefore prudent to carry out some form of audit like what the current study has done so that the authorities know the true picture of the research system in the country.

1.6 Scope and delimitation of the study For the purpose of this study, research productivity of university academic staff in Kenya has been limited to research output in journals, conference papers and books authored between 2004 2008. The study focused in the academic staff in both private and public universities in Kenya. There were 5 and 6 private and public universities respectively. Academic staffs demographic information and their attitude towards research and publication were studied.

lxxiv

Seventeen Heads of department also participated in the study. 1.7 Theoretical Framework This study has been grounded on a model developed by Victor Vroom (1964). The model is known as expectancy motivation theory. This is due to the conviction that to determine factors influencing someone to do something, they must have a motivation behind it. This theory has been modified by various scholars over the years. The most recent scholar to modify this model was Lertputtarak (2008) who used it for her doctoral dissertation on determinants of research productivity in Thailand. Business dictionary (2009) has defined motivation as internal and external factors that stimulate desire and energy in people to be continually interested in and committed to a job, role, or subject, and to exert persistent effort in attaining a goal. Motivation is the energizer of behavior and mother of all action. It results from the interactions among conscious and unconscious factors such as the (1) intensity of desire or need, (2) incentive or reward value of the goal, and (3) expectations of the individual and of his or her significant others. 1.7.1 Expectancy Theory of Motivation - Victor Vroom

lxxv

Expectancy theory relates choices to outcomes. Individuals assess the probability of success if a certain behavior is performed and choose to act based on the probability of certain outcomes, which may be intrinsic or extrinsic (Nadler & Lawler, 1977). Self determination theories are developed through the choices individuals make and the behavior toward the interaction with and mastery of ones environment. Self determination is the capacity and need to choose and to have these choices be determinant of ones actions (Deci and Ryan, 1991). According to this theory, effort, performance, and outcomes determine motivation. High motivation develops when individuals believe a realistic amount of effort will result in successful performance leading to desired outcomes. When an individual does not expect performance to bring a desired result, then effort will not be worth expending and motivation will be low. Thus, the key to motivation is to understand those elements that will enhance the linkages between effort and performance and outcomes (Nahavandi, 1997). As a cognitive choice theory, expectancy theory focuses on the manner in which decisions are made regarding allocation of effort, highlighting key components of the motivation process, such as effort-performance and desired outcomes, and how they work together intrinsically and extrinsically as the basis for such decisions (Mowday & Nam, 1997).

lxxvi

An individual is motivated to behave in a certain manner because (a) he or she has a strong desire for a certain task outcome and a reasonable expectation of achieving that outcome and (b) because he or she also expects that the achievement of the task outcome will result in reward in terms of pay, promotion, job security, or satisfaction of individual needs - physiological, safety, esteem and so on. Therefore, Vroom would maintain that we do things in our jobs in order to achieve second level rewards: Vroom (1964) indicated that if a worker sees high productivity as a path leading to the attainment of one or more of his or her personal goals, he or she will tend to be a high producer. Conversely, if he or she sees low productivity as path to the achievement of his or her goals, he or she will tend to be a low producer.

1.7.2 Rationale for Motivation Theories This motivation theory was selected for this study because the motivated environment drives staff to produce more research outcomes. Kuh and Whitt (1988) stated that academic environments and cultures or climates generally provide both socializing and reinforcing organizational norms, values and expectations concerning research. Therefore the environment under which researchers do their work is very critical in realizing increased research productivity.

lxxvii

Starbuck, 2006; 84) recognizes the critical role of researchers and notes; Since research is an occupation that involves prestige and salaries, one should expect to see career-oriented behaviour, and one does. Social scientists seem to be more concerned with producing papers than with producing knowledge.

This quote draws attention to the fact that university research, which is Starbucks focus in his book, is not just knowledge production, but is also, more mundanely, paid work with reputation and career opportunities attached to it. The image of the brilliant yet isolated mind, locked up in an ivory tower and striving to enhance knowledge just for the sake of enhanced knowledge has been subject to erosion (Barry et al., 2001). A clear indication here is that academic staff in Kenya and elsewhere is also concerned about developing their careers and reaping maximum output out of it. They are therefore motivated to work just like any other employees in an organization. Nadler and Lawler (1977) summarized the four assumptions of expectancy theory: 1. Behavior is determined by forces that exist within the individual and their work environment.

lxxviii

2. Individuals make decision about work behavior based on examining whether they are part of the group (membership) plus their effort to perform the task for how hard to work, how much to produce, and at what quality. 3. People have different needs, desires and goals. 4. People make decisions among a variety of choices based on their expectations that a particular behaviour will lead to desired output. In conclusion, expectancy theory appears suitable for this study as it views motivation and performance as critical aspects to concepts such as research productivity. The following Table summarizes studies that have been done focusing on research productivity and employing the expectancy model in their studies.

Table1.1: Summary of related studies on research productivity employing the expectancy model.

lxxix

Name(s) of researcher(s) Butler and Cantrell (1989)

Topic of study Extrinsic reward valence and productivity of business faculty: A within and between subjects decision modeling

Theories Employed Expectancy theory

Tein and Blackburn (1996)

experiment Faculty rank systems, research motivation and faculty research productivity measure refinement and theory

Reinforcement theory, Cognitive evaluation, Expectancy theory

Blackburn and Lawrance (1995)

testing. Faculty at Work: Motivation, Expectation, Satisfaction.

Reinforcement theory, Personality and career development theories, Dispositional theories Expectancy theories, Attribution Theories, Efficacy theories. Information-processing

lxxx

theories Williams (2003) A mediated hierarchical regression analysis of factors related to research productivity of human resource education a workforce development Chen, Gupta, and Hoshower (2006) postsecondary faculty Factors that motivate business faculty to conduct research: An expectancy theory analysis. Journal of Lertputtarak S. (2008) Education for Business. An Investigation of Factors Related to Research Productivity in a Public University in Thailand: A Case Study Expectancy theory Expectancy theory Expectancy theory, Efficacy theory.

Mowday, R., and Nam, S.

Expectancy Theory

Expectancy theory

lxxxi

(1997), Tien F.F. (2000),

Approaches to Faculty motivation. To What Degree Does the Desire for Promotion Motivate Faculty to Perform Research? Testing the Expectancy Theory.
Expectancy theory

Modified from Lertputtarak (2008) The above Table has given a summary of various studies that have been done since 1989 to as recent as 2008 and all have employed the expectancy theory in their work. This is a justification for this study to employ this theory in this study.

Efficacy Theory Perhaps this is the most recent modification to the expectancy model. Besides expectancy theory, efficacy theory is important to this thesis. Although efficacy theory is not included in process or content motivation theories, efficacy theory was mentioned in the research by Blackburn and Lawrence (1995), and William (2003) that studied research productivity. In regards to expectancy and value, efficacy theory is closely related to

lxxxii

expectancy theory (Bandura and Locke 2003). Gist and Mittchell (1992) suggested that the significance of self-efficacy for motivation and performance in work settings has been well demonstrated and also used in the technical repertoire of human resource management professionals. In his social cognitive theory, Bandura (1997:3) introduced the construct of self-efficacy. He describes self-efficacy as;

Confidence in ones capabilities to organize and execute the courses of action required to produce given attainments

As a consequence, he suggests that efficacy theory plays an important role in a persons self-regulation processes (Bandura 1991). In this theory, a persons behaviour is motivated and regulated by self-evaluation reactions to their own actions, and therefore, self-directedness partially determines the course of ones behaviour. People will participate in and try to deal with situations that they have ability to handle, but avoid situations that they perceive as being beyond their capabilities. Self-efficacy theory helps us to demonstrate how much effort people will expend and how long they will persist in the face of difficulties (Bandura 1977), and helps us to

lxxxiii

predict how a persons level of effort and persistence on a task will vary in relation to their level of goal commitment. This suggests that the higher a persons perceived self-efficacy, the greater is the potential for performance related accomplishments (Bandura, Reese & Adams 1982). Selfefficacy is different from self-esteem and self-concept, which tend to be more global assessments of the self across several situations. Self-efficacy is task-specific and varies in relation to experience, learning, and performance feedback (Bandura1982) Bandura (1977) indicated that efficacy is derived from four major sources: performance accomplishment, vicarious experiences, verbal persuasion and physiological arousal. Furthermore, expectations of personal efficacy appear to determine coping behaviour, that is, initiation, effort expended and sustained effort. In this regard, Bandura (1977:191) postulated that: Cognitive processes mediate change but that cognitive events are induced and altered most readily by experience of mastery arising from effective performance...psychological changes can be produced through other means than performance accomplishments.

lxxxiv

He also stated that behaviour patterns are formed through observation of others and that these observations later serve as a guide for action. These research findings indicate that people who view themselves as highly efficacious act, think and generally feel differently than people who perceive themselves as inefficacious (Bandura 1986), suggesting that personal accomplishments require both skills and belief in what they can do or the ability to use their skills and knowledge. Phillips and Russell (1994) found a statistically significant correlation between research self-efficacy and research productivity (r=0.45) and between self-efficacy and the research training environment (r=0.39). A study by Taylor, Locke and Gist (1984) demonstrated that self-efficacy is directly linked to performance of academic research productivity. This accorded with the work of Landino and Owen (1998), who found that facultys research productivity was positively correlated with self-efficacy (r=0.17), and Vasil (1992), who found that when self-efficacy perception increased, academic research productivity also increased. Another related study by Blackburn et al. (1991), who conducted a study of 3,930 faculty members from all institution types across the United States, found that self-efficacy accounted for a significant proportion of explained variance in research productivity (r=0.44). 1.8 Conceptual Framework

lxxxv

The following figure illustrates how this study has been conceptualized by the researcher. This figure has been modified to suit the current area of study.

Figure 1.1: Factors Responsible for Faculty Research Productivity

Institutional Characteristics

High Research Productivity

Behaviour

Individual Characteristics

Low Research Productivity

lxxxvi

This conceptual framework states that the Individual and institutional characteristics work together to influence ones behaviour towards research productivity. Both institutional and individual characteristics work together in influencing the behaviour of an academic staff towards research productivity. In this case, ones behaviour can lead to high or low research productivity, high research productivity in this case means that an academic staff will produce more books, articles, patents, conference papers etc. Low research productivity on the other hand means that an academic staff, who is not affected by institutional and individual characteristics, will produce less and hence have low research productivity. It shows the interrelationship at play between individual and institutional factors to form a basis for a productive, research oriented institute. In this case a university. This

lxxxvii

model has been modified from Bland, Seaquist, Pacula, Center and Finstad (2002) who synthesized literature on faculty research productivity and finally developed a model that asserted high research productivity to be strongly associated with both institutional, individual and leadership characteristics. In the Bland et al (2002) model, faculty research productivity is highest when a faculty member has specific individual qualities works in an institution that is highly conducive to research and supported by able institutional leadership. The model went further to make a hierarchical order of the factors and how they influence research productivity. That is the individual characteristics are essential but they have more or less power in assuming faculty research productivity depending on how research productive the faculty members institution is. In this model, the final output has been provided as articles, books, conference papers, Patents etc. These are products of a research endeavour. When all these products are divided by the number of the academic staff, then research productivity is realized. It is important to note that a higher value realized means that the research productivity is higher and vice versa. 1.9 Operational Definition of Terms Research Productivity

lxxxviii

For the purpose of this study, research productivity has been limited to all forms of output from a research endeavor. This is in the form of research papers, conference presentations, research publications in journals, and books authored. This can be expanded to mean any scholarly research produced by academic faculty members that contributes to the knowledge base of a discipline. For example; a research publication in refereed journal, academic book or book chapter. (Creswell, 1986; UNESCO, 2006) Institutional factors Institutional factors are those factors that directly emerge from the institutions structure, such as the type of institution, institution policy for promotion, research policy, and faculty collaboration toward a community of scholars, workload, salary, mentorship, recruitment and selection of staff, resources, and material support. Individual Factors Demographic factors used in this study were derived from socio-demographic factors of Blackburn and Lawrence (1995). These demographic factors include age, gender and marital status, and these were included in individual factors. They have been used to find out whether they interfere with an academic staff members ability to carry out research. Others are attitude toward conducting research, academic qualifications, advanced degree earned, research experience, skills and training, and rank status.

lxxxix

Measuring Research Productivity Research productivity is conventionally measured as the ratio of total publications to number of lecturers. Publication analysis of journal articles and books is clearly the most common measure of such research performance (Olson, 1994). The principal dependent variable for research productivity in the current study is the number of total articles/books published per average faculty member. Publication data gathered from self-reported information was found to be a reliable indicator. Allison and Stewart (1974) found that self-reported response from chemists was correlated with publication counts obtained from Chemical Abstracts (r = .94).

1.10 Organization of the study This thesis has been organized into five chapters, Chapter one concentrates on introduction which forms the preliminaries to the whole thesis. The background to the study, research problem, the research questions and hypothesis are presented here. Definitions of terms used in this study are also given in this chapter. Chapter two presents the Literature review which is a highlight of previous work done by other scholars in areas related to the present study. This chapter mainly

xc

concentrates on work done on research productivity across the globe. The most relevant models to this study are also reviewed. A Summary of the literature is given. Chapter three explains the research design and methodology used in this study. Chapter four concentrates on presentation, interpretation and discussion of the findings. Chapter five makes summary, conclusions and recommendations of the study.

CHAPTER TWO LITERATURE REVIEW 2. Introduction The following literature review has mainly been done on the area of research productivity at various levels from diverse parts of the world but mainly from the west. The keywords that were used to search for the related literature were publication productivity, and research productivity. Some issues have been highlighted that have been borrowed

xci

by the present study and finally a summary of the same has been done. This literature review has been grouped into the following sub headings;

Theories Related to Research Productivity Research Productivity and Publications Theories of Motivation Determinants of Research Productivity Research Productivity Measurement Research Productivity and Type of Institution Factors that Influence Research Productivity Factors associated with Personal Career Development Research Productivity and Academic Disciplines Research Productivity and Technology Transfer/Patents Research Productivity and Teaching Effectiveness Summary of Literature Review

xcii

2.1 Research Productivity and Theories Based on the rationale of expectancy theory, Tien (2000) examined the degree to which the desire for promotion motivates faculty to do research. Using Taiwanese faculty survey data, the study found that faculty members, who showed higher motivation for promotion, also displayed better research performance than their colleagues who showed lower motivation for promotion. The study also found that different kinds of rewards had different motivating effects on various types of faculty research performance. After controlling for the effects of demographic, educational, and institutional variables, the results of logistic regressions showed that faculty who thought that promotion and the satisfaction of curiosity were important, tended to publish articles; faculty who wanted to demonstrate their mastery of their disciplines tended to publish books; and faculty who cared about personal income were more likely to seek and receive the National Science Council Research Outcome Grant. Bean (1982) developed a model known as a causal model of faculty research productivity. This model proposes two general types of variables which are assumed to affect individual research productivity: institutional variables and individual variables. Institutional variables are as follows: level of research emphasis at the institution, granting of advanced degrees at the institution, institutional reputation, and size of institution,

xciii

degree of affluence, degree of centralization, and degree of autonomy of the institution. Individual variables are as follows: level of research goals, number of research colleagues, degree of undergraduate teaching responsibilities, level of research resources, level of perceived equity of rewards, level of alienation, perceived level of legitimacy in one's research, level of expectancies, level of need for personal growth, level of publication in graduate school, period of time as a faculty member after beginning as a productive academic in research, academic rank at an institution with a research emphasis, and level of individual autonomy for individuals with high levels of research goals. Several variables that have received mixed support in empirical research, and which are not included in this model are individual ability, sex, field of study, career stage, and prestige. The model indicates the pathways through which these variables are expected to produce variations in the dependent variable. These variables act through a set of individual variables in influencing individual productivity which is the dependent variable. These variables are expected to have multiplicative/interactive effects on the dependent variable. The model also indicates that organizational factors affect individual behaviour which results in a faculty member being more or less productive in the area of research.

xciv

A number of studies have proposed models to account for researcher productivity. Finkelstein (1984) proposed seven variables to predict the publication rates of academics. However, this model did not include the institutional factors that impact on researcher productivity. The current study has brought both institutional and individual factors on board. Creswells (1985) model recognized the importance of the institution and its research culture in influencing an individual academics productivity. Dundar and Lewis (1998) developed a statistical model from US multi-disciplinary data that found research productivity to be associated with both individual and environmental attributes. Bland, Center, Finstad, Risbey and Staples (2005), empirically tested the model of Bland, et al (2002) using data from a medical school and quantitative methods. They concluded that a combination of individual and institutional factors, facilitated by effective leadership, influenced research productivity in the school. Two models linked to research productivity were identified that have been applied to a business context. The first tailored a theoretical model developed to assess the scholarly output of economists (Burke, Fender and Taylor 2007) to academic accountants, using multivariate statistics. In a second model, Chen, Gupta and Hoshower (2006) applied Vrooms expectancy theory model with an aim to better understand the motivation of business related academics from eight main disciplines, including accounting.

xcv

Lotka's law, named after Alfred J. Lotka (1926) is one of a variety of special applications of Zipf's law. It describes the frequency of publication by authors in any given field. It states that the number of authors making n contributions is about 1 / na of those making one contribution, where a nearly always equals two. More plainly, the number of authors publishing a certain number of articles is a fixed ratio to the number of authors publishing a single article. As the number of articles published increases, authors producing that n publications become less frequent. There are 1/4 as many authors publishing two articles within a specified time period as there are single-publication authors, 1/9 as many publishing three articles, 1/16 as many publishing four articles, etc. Though the law itself covers many disciplines, the actual ratios involved (as a function of 'a') are very disciplinespecific.

The general formula says: XnY = C Or Y = C / Xn,

xcvi

Where X is the number of publications, Y the relative frequency of authors with X publications, and n and C are constants depending on the specific field (Lotka, 1926). This law is believed to have applications in other fields for example in the military for fighter pilot skills. Illustration For 100 authors, who on average write one article each over a specific period, we also have those making one contribution, i.e. a power law, where a is often nearly 2. It is an empirical observation rather than a necessary result. This form of the law is as originally published and is sometimes referred to as the "discrete Lotka power function Table 2.1: Illustration of Lotkas Law Number of articles written Number of authors writing that number of articles 10 9 8 7 6 5 4 3 2 1 Source (Lotka 1926) 100/102 = 1 100/92 1 (1.23) 100/82 2 (1.56) 100/72 2 (2.04) 100/62 3 (2.77) 100/52 = 4 100/42 6 (6.25) 100/32 11 (11.111...) 100/22 = 25 100

xcvii

There is need for a study on research productivity in Kenya to work out and see whether the output in Kenya confirms the Lotka power function. This law has been

applied elsewhere, this was done in Nigeria. Gupta (1987) generated a bibliography of entomological research in Nigeria, 19001973 in total, 1720 publications were analysed to study the author productivity patterns and to test the applicability of Lotka's law for the obtained distributions. Four different files were generated, one for the publications of all the authors, second for the publications by first authors, third for single authors and fourth for coauthors. Lotka's law, in its original form as inverse square law, was found not applicable to any of the four data sets. However, it was found to apply in its generalized form with the calculated values of characteristic exponent . The values of were found to be 1.9, 1.8,

2.2 and 2.4 for the four different data sets. K-S statistical test was applied to test the applicability of generalized form of Lotka's law. The maximum difference in the observed and estimated values of the proportions of authors was found to be highly insignificant at 0.01 level of significance in each of the four cases Nwagwu (2006) carried out a study on Bibliographic data on biomedical literature of Nigeria drawn from articles listed in Medline Journal covering the period 19672002, and numbering 6820. These articles were analysed to study the pattern of

xcviii

productivity of various author categories using Lotkas law. The total of 2184 authors who wrote the papers was divided into four different files, namely all authors, first authors, noncollaborative authors and co-authors. The hypothesis of this study was that the productivity patterns of each of the categories of authors differed from Lotkas inverse power law. The results showed that only the co-author category differed from the inverse power version of the law, while the other categories did not, although they yielded various exponents. Both Nwagwu (2006) and Gupta (1987) are in agreement with Lotkas law. This is attributed to the fact that their exponential ranges within the limits of 1.8 and 2.6.

Bland et al (2002) synthesized the literature on faculty research productivity into a model that asserts high research productivity is strongly associated with eight individual characteristics, fifteen institutional characteristics, and four leadership characteristics. This model has evolved through its application in several studies, as noted earlier. In the Bland et al. (2002) model, faculty research productivity is highest when a faculty member has specific individual qualities, works in an institution that is highly conducive to research, and is led by someone who possesses essential leadership qualities and uses an assertive participatory management approach.

xcix

Figure 2.1 displays the model and briefly describes the individual, institutional, and leadership characteristics in the model. Further, the Bland et al. (2002) model suggests a hierarchical order to these three sets of qualities. That is, the individual characteristics are essential, but they have more or less power in assuring faculty research productivity depending on how research-conducive the faculty members institution is. Finally, the impact of the institution is mediated by the qualities and style of the leader. This study has picked on the interaction of the individual and institutional factors on the influence the overall research productivity of the university academic staff in Kenya.

2.2 Research Productivity and Publications Price (1963) who studied the growth of scientific literature went on to generalize that 50% of scientific publications was produced by 6% of the scientific community and that the average scientist published about three papers in his lifetime. In another study, Bottle, et al (1994) compared publication counts produced by chemical professors, readers and senior lecturers in the United Kingdom and those in the United States (1981-1991) and found no significant difference in their publication productivity. Reskin (1977) studied a random sample of 238 academic chemists between 1955 and 1961 and found that 7.5% published nothing in the first decade following the receipt of

their degree and 11% published 1 article. Although the average rate of publications achieved was low, the variations of publication productivity between the scientists were high (Blume & Sinclair, 1973). Lotka (1926) analysed papers published in physics journal and found the distribution of publication was highly skewed. This indicated that a small minority of scientists produced the bulk of the papers. Kyvik (1990) in a study noted that productivity differences were the least in natural science (women published 20% fewer articles than men) whilst women in medicine, social science and humanities were 30-35% less productive than men. Academic rank was found to be important in relation to productivity. Professors were more productive than associate professors, and since there were fewer women in senior positions, the difference in productivity between ranks had consequences for average productivity between male and female researchers. Tower, Desai, Carson and Cheng (2006) reached the same conclusions in a large scale study of Australian accounting academics. Interestingly, Kyvik (1990) observed that women published less than men in the same positions but that they were more productive than men in lower positions. Thus female associate professors published more than male associate professors, and female associate professors published more than male assistant professors.

ci

There was a relationship between age and productivity and this connection was upheld for both men and women. High age was negatively related to productivity for both men and women. Women were more productive in the age group 50-54, while men were more productive in the age group 45-49. Considering all researchers, productivity was highest in the 45-49 age groups. For both men and women, married and divorced persons were more productive than single persons. Women with children were more productive than women without children. (Tower, G; Plummer, J; and Ridgewell, B 2007) This study has used publications in refereed research journals as a surrogate for research productivity. This approach is supported by the literature. Radhakrishna and Jackson (1993) reported that publishing in refereed journals was ranked as the most important factor when agricultural and extension education department heads were asked to rank the importance of 13 factors in the evaluation of faculty. In a related study, Radhakrishna, Yoder and Scanlon (1994) concluded Publications (refereed articles in journals and paper presentations in conferences) are considered to be a very important component of faculty productivity (p. 17). In Kelly and Warmbrods (1986) study, most of the variance (84.1%) in their research productivity score was explained by publications in refereed journals, with the remaining variance explained by seven other variables. The

cii

decision to use refereed journal articles as a surrogate for research productivity was based on the studies cited here. Wagner, Hornsby, Talbert, Hobbs, Brown, & Kenrick (1994) undertook a study whose purpose was to quantify research publication productivity of family medicine departments in selected family medicine and interdisciplinary journals. A 5-year journal search was conducted to identify original research articles published by family medicine department faculty. Publication productivity of all departments was ranked, and regression analysis was used to identify predictors of publication productivity. The departments leading in publication productivity published more than 25 articles over the 5-year period. The number of faculty and mean dollar value of family medicine department establishment grants were the strongest predictors of publication productivity. When adjusted for departmental size, some departments were found to have high publication productivity per faculty member, even though total numbers of departmental publications were low in comparison to other departments. This study identified one way of comparing departments in terms of publication productivity. Large departments, and those with more developmental grant support, had the highest publication productivity. Quantification of research output is not enough as this study attempted to do. The present study has gone beyond quantifying the research productivity and found out reasons that propel higher

ciii

productivity of research output among university lecturers in selected public and private universities in Kenya. The counting of total or average publications achieved is therefore a common and popular method used to assess research productivity; it is also easier to obtain such bibliographic data (Martin, 1996). This study has used publication data gathered from selfreported information from University lecturers in Kenya. It was easier to obtain this information from the lecturers themselves than to the journal publications. This is because most of the journals are published out of the country. Generally, the concept of productivity is considered as "units of output per units of time" (Waworuntu & Holsinger, 1989). When applied to research, Print and Hattie (1997) stated that research productivity is the totality of research performed by academics in universities and related contexts within a given time period. Then, research performance indicators can be devised by measuring that productivity in order to provide a basis for making judgments about research quality. Zainab (1999) states that the outputs of research comprised of intangible and tangible outcomes. The intangible outcomes are more complex, which include new scientific knowledge and awareness of new methodologies, and theories. On the other hand, the tangible outputs of research are published research findings such as research

civ

report or publication in refereed journals which has achieved national or international recognition, or communicated at conferences. Researchers grant different forms of recognition based on their contribution to the field, which include citations, positive ratings and rankings by peers, award of honors and prizes.

Publication counts became units for measuring output. Publication count is an indicator of research productivity, which may also include patents, inventions and awards. Publication counts are used to rank faculties and academics. Institutions can be ranked based on the total of publications and the ratio of publications to full-time faculty (Toutkoushian, Porter, Danielson and Hollis, 2003). Lange (2001) indicates that quantitative science indicators are essential indicators for evaluation purposes. They are used for the allocation of funds, scholarships, and tenures. Apart from publication lists, the most frequently used quantitative indicators for scientific performance, are the citations which scientists, journals, or scientific institutions receive. Author productivity, together with the type of publication and the rank of author, can be used to assess the output of a researcher (Tsay, 2004). The number of papers published by a group, institution or nation is a partial indicator of its size and productivity, which give an indication of the research activity in a

cv

particular discipline. Therefore, the publication produced in a particular discipline need to be determined in order to assess its productivity (Gu and Zainab, 2001).Research performance and publication productivity by faculty members of an institute could be used as indicators for ranking institutions. In their study, Academic Ranking of World Universities Methodologies and Problems, Liu and Cheng (2005) have ranked more than 1000 universities worldwide by several indicators of academic or research performance, including alumni and staff winning Nobel Prizes and Fields Medals, highly cited researchers in twenty-one broad subject categories, articles published in Nature and Science, articles indexed in Science Citation Index-Expanded(SCIE) and Social Science Citation Index (SSCI), and academic performance with respect to the size of an institution. It would be impossible to rank the quality of university education worldwide due to the huge differences of universities in different countries and the technical difficulties in obtaining internationally comparable data. Therefore, Liu and Cheng ranked research universities in the world by their academic or research performance based on internationally comparable data. Generally, research publication is used to assess the qualifications for promotion and tenure. Therefore, scientists do research in order to get promoted to higher rank among their colleagues. Although they preferred teaching as one of the criteria used for evaluation

cvi

process for tenure and promotion, but the emphasis was placed on research (Ali, Young and Ali, 1996). Thus, scientists prefer to collaborate with other researchers in order to be more productive and to produce better quality research. Published literature has reported a number of studies that used the quantity of publication to assess research productivity. Blackburn, Behymer and Hall (1978) used total articles published over two years, total career publication and total book published from self-reported data to assess the productivity of 1,216 academic staff members from 4-year colleges and 7,484 staff from universities in the United States. The instrument used was the questionnaire. Publication data gathered from selfreported information was found to be a reliable indicator. Allison and Stewart (1974) found that self-reported response from chemists was correlated with publication counts obtained from Chemical Abstracts (r = .94). Publication counts have not only been used to provide productivity counts but also used to assess research trends in certain disciplines. David, Piip and Haly (1981) used total number of publication counts in textile research to identify trends in specific areas of research and found a decline in basic research at the expense of applied textile research. Though this study was done many years ago, the present study has borrowed the use of self reported data and the questionnaire in its study format.

cvii

Onsongo (2005) published an article entitled, The Role of Research and Publication in the Promotion of Academics in Kenyan Universities based on a study that was carried out in one of the Kenyan public universities between January and April 2000 to investigate the role of research and publication in the promotion of academics in Kenyan universities. Data were collected from 14 academic men and 13 academic women through questionnaires/interview guides and document analysis. Documents such as promotion criteria and academic staff lists were used to analyse the promotion criteria and determine the ranks occupied by the academic staff. Findings from the study revealed that academic promotions were strongly linked to research and publications. Academic staff had limited access to research and publication as shown by their low involvement in research and low publication rates. Academics were found to face a number of obstacles such as inadequate funding, lack of information on available resources, absence of support systems to enhance research and publications, and inadequate time. Perhaps this is the only study carried out in Kenya on the area of research productivity. The sample population used here was not representative of the university academic staff at the time. The present study increased the sample population so as to realize generalizable findings. Maske, Durden and Gaynor (2003) examined the factors that cause disparity between male and female publications. They found 41.3% of the difference between male

cviii

and female article production was explained by experience, number of courses taught, type of university orientation, and other control factors. They argued that the unexplained difference may be related to discriminatory practices in the publication process. Other contributory factors showed that women were more involved in community service activities at the expense of research. Their statistical regression results showed that females had 12.2 years experience whereas males had 17.2 years experience; the marginal year of experience was associated with an increase of 0.99 papers for males and 0.45 for females. Other significant predictive factors included a negative relationship with time devoted to administration, teaching or working in a teaching-focused institution. This study dwelt on the factors causing disparity in research performance between male and female researchers. This is not enough; it should have proceeded to find out those factors that promote or hinder research productivity of all sexes. That is what the present study has strived to accomplish. Oppenheim and Ellerslie (2008) carried out an investigation whether a relationship existed between motivation and publication productivity of UK academic Information Scientists. A motivational questionnaire survey was performed, and citation analyses undertaken to determine the publication and citation count of the 45 respondents. Findings of this study demonstrated significant differences in motivational levels and publication

cix

counts by age, gender, caring responsibilities and hours spent on research. The paper concluded that those likely to produce more publications were older males without responsibilities who did 6-15 hours research per week. The conclusions of this study cannot be so useful in academic circles. The present study came up with tangible conclusions on the way forward to motivate university lecturers to work even harder in their academic endeavours. Figure 2.1: Factors Responsible for Faculty Research Productivity

cx

Source: Bland et al (2002) This model has been borrowed and modified to fit the conceptual framework for this studyit forms a perfect interaction of the various factors at play in the research productivity efforts of the academic staff. 2.3 Theories of Motivation There are a number of different views as to what motivates workers. The most commonly held views or theories are discussed below and have been developed over the last 100 years or so. Unfortunately these theories do not all reach the same conclusions! (Wikipedia) Taylor Frederick Winslow Taylor (1856 1917) put forward the idea that workers are motivated mainly by pay. His Theory of Scientific Management argued the following: Workers do not naturally enjoy work and so need close supervision and control Therefore managers should break down production into a series of small tasks Workers should then be given appropriate training and tools so they can work as efficiently as possible on one set task.

cxi

Workers are then paid according to the number of items they produce in a set period of timepiece- rate pay.

As a result workers are encouraged to work hard and maximize their productivity. Taylors methods were widely adopted as businesses saw the benefits of increased

productivity levels and lower unit costs. The most notably advocate was Henry Ford who used them to design the first ever Production line, making Ford cars. This was the start of the era of mass production. Taylors approach has close links with the concept of an autocratic management style (managers take all the decisions and simply give orders to those below them) and Macgregors Theory X approach to workers (workers are viewed as lazy and wish to avoid responsibility). However workers soon came to dislike Taylors approach as they were only given boring, repetitive tasks to carry out and were being treated little better than human machines. Firms could also afford to lay off workers as productivity levels increased. This led to an increase in strikes and other forms of industrial action by dis-satisfied workers. Mayo Elton

cxii

Elton Mayo (1880 1949) believed that workers are not just concerned with money but could be better motivated by having their social needs met whilst at work (something that Taylor ignored). He introduced the Human Relation School of thought, which focused on managers taking more of an interest in the workers, treating them as people who have worthwhile opinions and realizing that workers enjoy interacting together. Mayo conducted a series of experiments at the Hawthorne factory of the Western Electric Company in Chicago He isolated two groups of women workers and studied the effect on their productivity levels of changing factors such as lighting and working conditions. He expected to see productivity levels decline as lighting or other conditions became progressively worse. What he actually discovered surprised him: whatever the change in lighting or working conditions, the productivity levels of the workers improved or remained the same. From this Mayo concluded that workers are best motivated by: Better communication between managers and workers (Hawthorne workers were consulted over the experiments and also had the opportunity to give feedback) Greater manager involvement in employees working lives (Hawthorne workers responded to the increased level of attention they were receiving)

cxiii

Working in groups or teams. (Hawthorne workers did not previously regularly work in teams) In practice therefore businesses should re-organize production to encourage greater

use of team working and introduce personnel departments to encourage greater manager involvement in looking after employees interests. His theory most closely fits in with a paternalistic style of management. Maslow Abraham Maslow (1908 1970) along with Frederick Herzberg (1923- ) introduced the Neo-Human Relations School in the 1950s, which focused on the psychological needs of employees. Maslow put forward a theory that there are five levels of human needs which employees need to have fulfilled at work. All of the needs are structured into a hierarchy and only once a lower level of need has been fully met, would a worker be motivated by the opportunity of having the next need up in the hierarchy satisfied. For example a person who is dying of hunger will be motivated to achieve a basic wage in order to buy food before worrying about having a secure job contract or the respect of others. A business should therefore offer different incentives to workers in order to help them fulfill each need in turn and progress up the hierarchy (see below). Managers should

cxiv

also recognize that workers are not all motivated in the same way and do not all move up the hierarchy at the same pace. They may therefore have to offer a slightly different set of incentives from worker to worker. Herzberg Frederick Herzberg (1923- ) had close links with Maslow and believed in a twofactor theory of motivation. He argued that there were certain factors that a business could introduce that would directly motivate employees to work harder (Motivators). However there were also factors that would de-motivate an employee if not present but would not in themselves actually motivate employees to work harder (Hygiene factors). Motivators are more concerned with the actual job itself. For instance how interesting the work is and how much opportunity it gives for extra responsibility, recognition and promotion. Hygiene factors are factors which surround the job rather than the job itself. For example a worker will only turn up to work if a business has provided a reasonable level of pay and safe working conditions but these factors will not make him work harder at his job once he is there. Importantly Herzberg viewed pay as a hygiene factor which is in direct contrast to Taylor who viewed pay, and piece-rate in particular Herzberg believed that businesses should motivate employees by adopting a democratic

cxv

approach to management and by improving the nature and content of the actual job through certain methods. Some of the methods managers could use to achieve this are: Job enlargement workers being given a greater variety of tasks to perform (not necessarily more challenging) which should make the work more interesting. Job enrichment - involves workers being given a wider range of more complex, interesting and challenging tasks surrounding a complete unit of work. This should give a greater sense of achievement. Empowerment means delegating more power to employees to make their own decisions over areas of their working life. Reinforcement theory Reinforcement theory was developed by Skinner (1953). In reinforcement theory, behavior can be explained by environmental conditions. The theory relies on the concept of the law of effect, which demonstrates that positive or pleasant behaviors are more likely to be repeated (Thorndike 1911). There are four types of reinforcement: positive reinforcement, negative reinforcement, extinction and punishment. Positive and negative reinforcement intend to

cxvi

increase behavior, while extinction and punishment aim to decrease behavior. Due to reinforcement theory, people learn several things during the process of reinforcement. Although the reinforcement theory is a powerful influence tool, the theory contains some limitations (West Virginia University 1996); (1) it is difficult to identify rewards and punishment. Finding good rewards and punishments requires a great deal of experience and insight. (2) It requires control all sources of reinforcement, (3) internal changes can be difficult to create. It works best with the heuristic thinker, not requiring systematic thinking. It needs to maintain steady reinforcement cues to maintain the desire actions. (4) Punishing is difficult to do well. Adam's Equity theory Adam's Equity theory is a motivation theory that points out the fact that the managers should seek a fair balance between the employees' inputs (effort, loyalty, hard work, sacrifice, etc) and their outputs (recognition, status, salary, status etc), in order to motivate employees (Adams, 1965). He also stated that it is very vital to make the employee feel that he is treated fairly if the managers are to achieve positive outcomes and motivate the employees effectively. McClellands Need for Achievement

cxvii

The one single motivating factor which has received the most attention in terms of research is the need for achievement (n-ach). Much of this knowledge is due the work of David McClelland of Harvard. Individuals with a high n-ach have a number of distinctive characteristics which separate them from their peers. First of all, they like situations where they can take personal responsibility for finding solutions to problems. This allows them to gain personal satisfaction from their achievements. A second characteristic of high n-ach people is that they like to set moderately high goals for themselves. These goals are neither so low that they can be achieved with little challenge, nor so high that they are impossible. A third distinctive characteristic of high achievers is that they want concrete feedback on their performance. Only certain types of jobs provide this kind of feedback, however, and so some kinds of jobs are unattractive to high achievers.

Expectancy Theory of Motivation - Victor Vroom Victor Vroom (1964), of Carnegie-Mellon in Pittsburgh, has challenged the assertion of the human relationists that job satisfaction leads to increased productivity.

cxviii

(This theory has been called the contented cow approach to management.) The assumption is that if management keeps employees happy, they will respond by increasing productivity. Vroom defines motivation as: "A process governing choices, made by persons or lower organisms, among alternative forms of voluntary behaviour." In organizational terms, this concept of motivation pictures an individual, occupying a role, faced with a set of alternative voluntary behaviours, all of which have some associated outcomes attached to them. If the individual chooses behaviour 1, outcome A results; if 2 then B results and so on. Expectancy theory relates choices to outcomes. Individuals assess the probability of success if a certain behavior is performed and choose to act based on the probability of certain outcomes, which may be intrinsic or extrinsic (Nadler & Lawler, 1977). Self determination theories are developed through the choices individuals make and the behavior toward the interaction with and mastery of ones environment. Self determination is the capacity and need to choose and to have these choices be determinant of ones actions (Deci and Ryan, 1991). According to expectancy motivation theory, effort, performance, and outcomes determine motivation. High motivation develops when individuals believe a realistic amount of effort will result in successful performance leading to desired outcomes. When an individual does

cxix

not expect performance to bring a desired result, then effort will not be worth expending and motivation will be low. Thus, the key to motivation is to understand those elements that will enhance the linkages between effort and performance and outcomes (Nahavandi, 1997). As a cognitive choice theory, expectancy theory focuses on the manner in which decisions are made regarding allocation of effort, highlighting key components of the motivation process, such as effort-performance and desired outcomes, and how they work together intrinsically and extrinsically as the basis for such decisions (Mowday & Nam, 1997). Knowing that individuals choose behaviours in order to obtain certain outcomes is nothing new. The question is why they choose one outcome over another. The answer provided by the other motivational theories (by Maslow, Herzberg, McClelland) is that the choice reflects the strength of the individual's desire or need for a specific outcome at a certain time. However, Vroom makes the point that task goals (productivity, quality standards or similar goals attached to jobs) are often means to an end, rather than the end in itself. There is a second level of outcomes which reflect the real goals of individuals and these may be attained, in varying degrees, through task behaviour.

cxx

An individual is motivated to behave in a certain manner because (a) he or she has a strong desire for a certain task outcome and a reasonable expectation of achieving that outcome and (b) because he or she also expects that the achievement of the task outcome will result in reward in terms of pay, promotion, job security, or satisfaction of individual needs - physiological, safety, esteem and so on. Vroom would maintain that we do things in our jobs in order to achieve second level rewards: "If a worker sees high productivity as a path leading to the attainment of one or more of his or her personal goals, he or she will tend to be a high producer. Conversely, if he or she sees low productivity as path to the achievement of his or her goals, he or she will tend to be a low producer" Vroom (1964) Certainly Vroom has hit on an important aspect of motivation. We do not attempt simply to satisfy a need or even a set of needs in a straightforward, "If I do this, then I will achieve that" manner. We work with a chain of goals and rewards, where goals in one area are only a means of achieving goals in another. In conclusion, the tasks of the managers to motivate the employees are indeed not that easy. This is because each and every employee has got their very own needs that tend to motivate them. However, the managers need to have some sort of acknowledge that will

cxxi

help them to understand the employees well and think of better ways of motivating them. This is where the motivational theories come into consideration. It is these theories that provide an explanation of how to motivate them based on what motivates them. The Hierarchy of needs theory and Hertzberg's two factor theory deals with explaining how the employees are to be motivated by way of looking into their needs. On the other hand, Adams theory also helps the managers to understand that a fair balance between inputs and outputs of employees are important. The Expectancy theory too shows that rewards tend to motivate the employees. However, overall all motivation theories do state that rewarding and recognizing employees are important in order to motivate employees thus acting as the foundation to motivate employees.

Choice of Expectancy Model This study is going to be guided by the expectancy model that has been outlined above. This justification is due to the fact that several studies have been conducted in the area of research productivity and most of them have based their studies on expectancy model. For example; Williams (2003) made very significant studies and in each case, he used the expectancy model effectively. These studies were; research productivity of

cxxii

nursing faculty and a mediated hierarchical regression analysis of factors related to research productivity of human resource education workforce development of post secondary faculty. Expectancy theory appears suitable for this study as it views motivation and performance as critical aspects to concepts such as research productivity. Nadler and Lawler (1977) summarized the four assumptions of expectancy theory:

Behavior is determined by forces that exist within the individual and their work environment.

Individuals make decision about work behavior based on examining whether they are part of the group (membership) plus their effort to perform the task for how hard to work, how much to produce, and at what quality.

People have different needs, desires and goals. People make decisions among a variety of choices based on their expectations that a particular behaviour will lead to desired.

cxxiii

One well-established research productivity theory, Life-Cycle theory (Hu and Gill, 2000), suggests that, in general, the research productivity of a researcher rises sharply in the initial stages of a career, peeks at the time of tenure review, and then begins a decline.

2.4 Determinants of Research Productivity Taking a slightly wider view, research productivity can include research publication in professional journals and in conference proceedings, writing a book or chapter, gathering and analyzing original evidence, working with post-graduate students on dissertations and class projects, obtaining research grants, carrying out editorial duties, obtaining patents and licenses, writing monographs, developing experimental designs, producing works of an artistic or creative nature, engaging in public debates and commentaries (Creswell 1986). However, research is typically a private and self-mastered activity, and it can be difficult for university staff members to balance an effective project agenda with the demands of teaching, service and life in general. According to Boice (1987), productivity should emerge from hard work, and a fair schedule for research activity should utilize a benchmark that encourages a struggling researcher to relate to their current level of activity. For example, Boice (1987) found that a new faculty member who could find only one hour per weekday to work on their research, generally managed to submit about 1.5 manuscripts

cxxiv

per year, which is then consistent with the expectations for a pay rise and higher tenure status. Furthermore, faculty members who adopt a regimen of brief daily periods for research projects typically experience less stress in managing their time and their lives (Boice 1987). The development of clear measures for research productivity will be a significant influence in the nature of the service sector. Research productivity has been defined as the relationship between the outputs generated by a system and the inputs provided to create those outputs. It may also include the term efficiency and more importantly effectiveness, which measures the total output or results of performance (Turnage 1990). However, in combining the two words as research productivity, a simple definition becomes more difficult in a research environment because different people have very different attitudes about its meaning. Whilst productivity is very important in industrial circles, public concern over competitiveness and productivity in universities enters virtually every policy discussion, whether the subject is education, the budget deficit or national politics (Krugman 1991). It is important to define the term research productivity. For the purposes of this study, it is important that the notion of research productivity be carefully defined, since it is a key element in the development of the research questions. To begin, Research means

cxxv

the careful study or investigation, especially in order to discover new facts or information (Oxford University 1995). Productivity means the total production compared with inputs or consumption over the same period of time, which serves as a measure of whether the producers production processes are working efficiently (Witzel, 1999). Print and Hattie (1997) define research productivity as the totality of research performed by academics in universities and related contents within a given time period (p.454), and research efficiency has been defined as the productivity of research per unit of input resource (Kostoff 1995). Research productivity is an outcome measurement of scholarly effort (Jacobs, Hartgraves & Beard 1986; Kurz et al. 1989), and has two components that are; (i) knowledge creation (research) and (ii) knowledge

distribution(productivity) (Gaston 1970). For the most part, the product of academic lecturers research is scholarly publication (Carnegie Foundation 1991). The importance of this definition of research productivity is that it enables faculty members to share insights, demonstrate academic scholarship, gain recognition for creative thinking, and finally to develop a reputation for expertise in a specialty area (Rhodman 2002). 2.4.1 Research Productivity Measurement The most frequently used measure of the quantity or amount of research productivity is a numerical publication count or the journal article count over a certain time

cxxvi

period. The activities included in measuring productivity range from a narrow perspective of number of research articles published to a broad interpretation which consists of presentations, both formal and informal, number of graduate students that a staff member is advising, publications of any type and proposals submitted for funding. Moreover, it also includes counts of the number of editorial duties, conference deliveries, licenses, patents, monographs, books, experimental designs, and works of an artistic or creative nature, public debates and commentaries (Creswell 1986). Rotten (1990) stated that a common approach to measuring research productivity was to count the number of books, articles, technical reports, bulletins, and book reviews published, as well as presentations given and grants received through reviewing curriculum vitae or other print materials. Price (1963) who studied the growth of scientific literature went on to generalize that 50% of scientific publications was produced by 6% of the scientific community and that the average scientists published about three papers in his lifetime. In another study, Bottle, et al (1994) compared publication counts produced by chemical professors, readers and senior lecturers in the United Kingdom and those in the United States (1981-1991) and found no significant difference in their publication productivity. The counting of total or average publications achieved is therefore a common and popular method used to assess research productivity since it is easier to obtain such bibliographic data (Martin, 1996).

cxxvii

Allison and Stewart (1974) found that self-reported response from chemists was correlated with publication counts obtained from Chemical Abstracts (r = .94). Braun, Glanzeland and Schubert (1990) used publication data from the corporate index files of the Science Citation Index (SCI) database for the period 1981-1985, to assess the publication productivity of authors from 10 major OECD countries. Budd (1995) addressed the level of publishing productivity of academic staff members from a number of universities for the years 1991 and 1993 who were also members of the Association of Research Libraries. The publication data was collected from the three citation indexes of the SCI. The universities were ranked by the number of publication and per capita publication achieved (total publication by number of academic staff). Publication counts have not only been used to provide productivity counts but also used to assess research trends in certain disciplines. David, Piip and Haly (1981) used total number of publication counts in textile research to identify trends in specific areas of research and found a decline in basic research at the expense of applied textile research. Reskin (1977) studied a random sample of 238 academic chemists between 1955 and 1961 and found that 7.5% published nothing in the first decade following the receipt of their degree and 11% published 1 article. Although the average rate of publications achieved was low, the variations of publication productivity between the scientists were high (Blume and

cxxviii

Sinclair, 1973). Lotka (1926) analysed papers published in physics journal and found the Distribution of publication was highly skewed. This indicated that a small minority of scientists produced the bulk of the papers. Fielden and Gibbons (1991) pointed out that within the business faculty, many lecturers emphasize articles published in refereed journals and trivialize all other measures of productivity. Clement and Stevens (1989) found that management administrators put greater weight on scholarly research and less on trade and newspapers articles than their non-management business peers. Radhakrishma and Jackson (1993) reported that publishing in refereed journals was ranked as the most important factor in research productivity, and Radhakrishma, Yoder and Scanlon (1994, p.17) noted that publication (in refereed articles in journals and paper presentations at a conferences) are considered to be very important component of faculty productivity. This statement was supported by Kotrlik, Bartlett, Higgins, & Williams (2002) in reference to Personal Communication from William J Cooper, former Dean of the Louisiana State University Graduate School. Kotrlik et al. (2002) quoted William Cooper as stating that the only magic number is zero; if you havent published in refereed journals, then publications in research conference proceedings, books and other publications are meaningless (p.3). This statement clearly gives credence to publishing in refereed journals. Going by the state of

cxxix

poor journal development in developing countries, it now becomes quite challenging for researchers from the developing countries to publish their works. Most universities do not have the financial ability to maintain these journals for long. The quantity of research productivity can be measured as number of published pages in journals (Malhotra & Kher, 1996; Hoverstad 1991, cited by Babber et al., 2000), as number of articles published (Stahl, Leap & We 1988; Hadjinicola and Soteriou, 2006); or through a combination of both methods (Grover et al., 1992; Babber et al., 2000). Such studies that have used a combination of both methods have aimed to rank institutions as well as individuals based on their research output. This study has employed such a combination was beyond the capacity of this study, so it focused on the number of articles published. Published literatures have reported a number of studies that used the quantity of publication to assess research productivity. Blackburn, Behymer and Hall (1978)used total articles published over two years, total career publication and total book published from self-reported data to assess the productivity of 1,216 academic staff members from 4-year colleges and 7,484 staff from universities in the United States. The instrument used was the questionnaire. Publication data gathered from self-reported information was found to be a reliable indicator.

cxxx

2.4.2 Webometrics Webometrics or cybermetrics is the study of the quantitative aspects of the construction and use of information resources, structures and technologies on the Web drawing on bibliometric and informetric approaches (Almind and Ingwersen, 1997) or the study of web-based content with primarily quantitative methods for social science research goals using techniques that are not specific to one field of study (Wikipedia). Generally webometrics are concerned with the uploaded materials to the internet. Since 2004 the Webometrics ranking of world universities has been offering information about more than 6,000 universities ranked according to indicators measuring Web presence and impact (link visibility).The Webometrics Ranking is a new measurement of research productivity. Webometrics is devoted to the quantitative analysis of the Internet and Web contents specially those related to the processes of generation and scholarly communication of scientific knowledge. Most universities across the globe are encouraging their staff to upload all manner of content to the internet with the aim of increasing the visibility of their respective universities. Some of the content upload include class notes, class schedules, course outlines, and other related instructional materials. The aim of these rankings is to provide extra motivation to researchers worldwide for publishing more and better scientific content on the Web, making it available to

cxxxi

colleagues and people wherever they are located. The new innovations in education in terms of open and distance education have greatly benefitted from this resource. The Web indicators used are based and correlated with traditional Scientometrics and bibliometric indicators and the goal of the project is to convince academic and political communities of the importance of the web publication not only for dissemination of the academic knowledge but for measuring scientific activities, performance and impact too.

2.4.3 Research Productivity and Type of Institution Holcomb, Bartel and Thomson (1988) were concerned that little was known about the scholarly productivity of faculty members who taught in respiratory care programs. They studied the scholarly activities of respiratory care faculty members in southern academic health centers via a mailed survey. An analysis of the responses (n = 33, 86.8%) revealed: (1) The respondents' principal scholarly activity was the reporting of research findings in refereed journals, with a productivity index (number of articles/years on faculty) of 0.25, or one published article for every 4 years of employment in higher education, which was significantly less than that of other allied health faculty (productivity index 0.69, P less than 0.05). (2) Less than a majority of respondents had presented a paper

cxxxii

at a professional meeting during the 3 years preceding the survey. (3) Only a small percentage of respondents had been involved in research. (4) Promotion opportunities and academic preparation were the primary factors that encouraged scholarly pursuits, and heavy teaching responsibility was the primary discouraging factor. (5) Scholarly activity is perceived as an important consideration in academic promotion decisions. In conclusion they said, respiratory care program faculty and administration should take steps to increase the scholarly production of faculty members. This study focused on one department only. The present study has captured more faculties in this study. Ho (1998) conducted a study on measurement of research output among the three faculties of business, education, humanities and social sciences in six Hong Kong universities. Data was collected from the 1990-95 annual reports of research and publication outputs of each university. In order to have a fair comparison of publication outputs of each academic, rank, faculty and university, a framework was developed from practical experience and from literature to investigate the problem. Results indicated that the publication outputs of academics in Hong Kong were about the same as other countries in many aspects. The recommendation from this study, therefore, was that Hong Kong academics should not be motivated to do more research. Pressing these academics for more research

cxxxiii

publications could raise the figure in the start, but would not necessarily increase the output in the long run. I do not agree with these findings. Researchers should be motivated to continue working hard in their areas of specialization. If laxity creeps in then it will be hard to overcome it again. This study relied on annual reports only. It could have been better to seek more information from other sources. The present study has used both lecturers and heads of department as main respondents in this study. Under this sub heading, studies conducted at different levels are considered. For example, Adams (1996) in a study entitled research productivity in a system of universities in the USA explored some efficiency aspects of the university system. The findings suggested that leading schools had lower average and marginal costs of performing research than lesser institutions, and that leading institutions had a comparative advantage at generating higher quality, more highly cited research. Brocato (2005) carried out a study to determine the research productivity of faculty in family medicine departments at U.S. medical schools, as well as the individual and environmental characteristics and prior socializing experiences predictive of research productivity. In the year 2000, a 43-item questionnaire was mailed to 796 faculty to obtain descriptive data toward formulating a conceptual model of the research productivity of family medicine faculty. Prior to model construction and testing through full-model

cxxxiv

regression, the model's factors were reduced through factor analysis. A total of 474 questionnaires (63%) were returned. Eighty percent of respondents spent a half-day or less per week on research; on average they produced less than one scholarly product per year. Few had research experience, nor could identify a research agenda or current research project. Mixed messages were perceived related to research, both at institutional and disciplinary levels. In testing a conceptual model, psychological and cognitive characteristics were most predictive of research productivity, along with time spent on research. Psychological and cognitive factors included enhancing research skills, establishing a definable research agenda, fostering research networks, having multiple research projects underway, maintaining in-depth knowledge of a research area, and clearly understanding research expectations for promotion and tenure. The clinical and academic demands on family medicine faculty reduced the likelihood that they will engage in research. These demands prevent the development of a critical mass to provide mentorship and networking necessary for research productivity. The study concluded that resources were needed to recruit faculty with an interest in research and to provide faculty development in research skills, mentorship, and networking. This study did not perform a triangulation to find out what the heads of departments had to say about the research

cxxxv

productivity. The present study has filled this gap by getting the input of the heads of departments. In one way or the other, they are in agreement with the lecturers in terms of research productivity in Kenya. Waller, Wyatt, and Karni (1999) undertook a study to describe the research and scholarly productivity of faculty in four-year college and university clinical laboratory science (CLS) programs. To identify meaningful scholarship, to assign values to that scholarship, and to list the top 15 CLS programs according to faculty research productivity. In 1996, a national study involving 127 college and university CLS programs was conducted to determine whether faculty was participating in research. A questionnaire was distributed to 505 faculty members. Data from 286 respondents (57% response) representing 114 of 127 (90%) CLS programs were analyzed. The study took place at The Ohio State University with collaboration from the University of Tennessee-Memphis and the University of Minnesota. All CLS faculty within a four-year university or college sponsoring a CLS program were invited to participate. Research productivity included time spent in research, numbers of publications and presentations, and grantsmanship. Data indicated that those faculties who possess earned doctorates and are employed by research universities have higher levels of research productivity. While 46% of the CLS faculty held doctorates and 50% were tenured, 42% of all CLS faculty members had not

cxxxvi

published a research paper or abstract since 1990. On the other hand, 23% of the faculty responding had published six or more articles or abstracts since 1990, 46% were successful in obtaining external funding, and 15% of faculty members had been awarded grants larger than $100,000. Conclusions arrived at indicate that the top 10% of clinical laboratory science faculty researchers are performing approximately one-half of all scholarly activities. The top fifteen research programs in CLS were identified, and not surprisingly, they were located in research universities. Results showed that CLS faculty had made progress in scholarship including highest degree obtained, publications, presentations, and grantsmanship. A questionnaire used in this study was borrowed and adjusted to be used in the present study. Barhyte and Redman (1993) undertook a study to establish Department-level measures of productivity. These were constructed by using information reported by 180 nursing deans of schools with graduate (master's and doctoral) programs. Productivity was calculated in three ways: total (net), publications, and grants. The scores for each school were derived from nine categories of faculty scholarly activities. The following variables were examined for their contribution to productivity: three measures of environmental support budgeted and doctorally prepared faculty, students (Masters, doctoral), all graduate students-faculty ratio, scholarship time, and private faculty offices. The regressions of log-

cxxxvii

transformed variables yielded R2 = .59 for total (net) productivity, .54 for publications, and .50 for grants productivity. This study was limited in that it delved on departmental measures of productivity. This study has considered individual factors too. Barhytes study concentrated on the departmental level factors only. Individual factors have been brought on board in the current study. The findings of the current study as can be seen in the later chapters of this thesis is that both individual and institutional factors play a role in the influence on research productivity in Kenya. Marchiori (1998) undertook a study to investigate the research productivity of chiropractic college faculty and identified parameters associated with increased peerreviewed publication. A survey was administered to collect data in a cross-sectional design. Data were collected through a survey administered to all full-time chiropractic college faculties working in the United States. Although the survey addressed many scholarly activities, the criterion variable selected for this study was the number of peer-reviewed journal articles published over the past 3 years. Three groups of faculty had more publications: those primarily assigned to research, those with either a D.C. degree or a D.C. and Ph.D. degree, and finally those with a rank of full professor. Faculty age and gender were not associated with the reported number of publications.

cxxxviii

The majority of faculty members (72.2%) had not published a single peer-reviewed article in the last 3 years. Less than 2% of the faculty members had published 10 or more peer-reviewed articles in the last 3 years. Conclusion arrived at indicated that many faculty were not involved in research activities. Conclusion here was that; Faculty development and incentive programs needed to be implemented to stimulate these individuals. This study took into consideration one department only, the present study has brought on board more faculties and departments as can be seen in chapter four of this work. Ramsden (2005) describes results from a study of academic productivity in Australian higher education. It estimated the output (in terms of quantity of publications) of individual staff and academic departments across different subject areas and types of institution. Concerning research productivity, Australian academics resemble their colleagues in other countries: the average is low, while the range of variation is high. Most papers were produced by few academic staff. Several potential correlates of productivity, including level of research activity, subject area, institutional type, gender, age, early interest in research, and satisfaction with the promotions system, were examined. A model linking departmental context to personal research performance through department and personal research activity was developed and tested. The results supported earlier views that structural factors (such as how academic departments are managed and

cxxxix

led) combined with personal variables (such as intrinsic interest in the subject matter of one's discipline) to determine levels of productivity. There was evidence also that research and teaching do not form a single dimension of academic performance. The present study has borrowed much from this study particularly in investigating research activity, subject area, institutional type, gender, age, early interest in research and such. It was interesting to see how these factors reflect in institutions of higher learning in developing countries like Kenya.

Milgrom, Heima, Tomar, Kunzel

(2008) carried out a study to describe the

research productivity of the members of the International Association for Dental Research (IADR) Behavioral Sciences and Health services Research Group and examined personal and professional factors related to greater productivity. The findings from previous studies suggested there could be gender discrimination in opportunities for women faculty. Members on the active membership list for this IADR group were surveyed by email. Most were dentists, and three-quarters had external funding for their research. The primary outcome measure was the number of self-reported published articles in Pub Med in the preceding twenty-four months. The mean number of these publications was 4.9 (SD=5.1). Gender and time in research were the best predictors of research productivity of this

cxl

population. There was no difference in time for research between the men and women in this study. Controlling for gender, the best single predictor of research productivity remained percentage time spent in research. Overall, the members of the IADR group spent almost three times as much time in research and were more than twice as productive as faculty members as a whole as described in earlier studies. In view of the current emphasis in many countries on addressing the social and behavioral determinants of oral health disparities, the productivity of this area of dental research is very important. Trends toward clinically oriented, non-research-intensive dental schools in the United States and reductions in time and funding available to conduct research should be of concern. This study ignored institutional factors. The present study recognizes the critical role played by institutions in promoting research productivity, the institutional factors have therefore been considered as critical players affecting academic staffs research productivity. Dyrbye (2008) undertook a study to determine the research productivity related to required research experiences during medical school. The authors studied the research productivity of the 998 graduates at Mayo Medical School who had participated in a required third-year medical school research experience (21, 18, or 17 weeks long) between 1976 and 2003. Outcomes were verified by published research reports and abstracts, and

cxli

presentations at scientific meetings. Research reports and abstracts related or unrelated to the required research were distinguished. Seventeen of the graduates were excluded when considering authorship of research reports (ambiguous data). Four hundred (41%) of the remaining 981 graduates published one or more research reports related to their required research experience, 176/998 (18%) published one or more abstracts related to their required research project, and 375/920 (41%) presented research findings at an extramural meeting at least once. Graduates who published a research report or abstract related to their required research or presented research at a scientific meeting published more research reports unrelated to their required research than did their peers who did not publish or present their required research (all P < .05). More graduates in the 21-week group were first authors (203/584; 35%) than were those in the 17/18-week group (60/336; 18%, P = .001), but other outcomes were similar for different durations (P > OR = .17). The study concluded that required medical school research experiences facilitate tangible research products and may promote subsequent research productivity. Shorter experiences seemed to yield outcomes similar to longer experiences. Dyrbyes study focused on one discipline only. This could not give a true picture of the research activities in the universities. This study in

cxlii

Kenya has brought on board more faculties and departments so that it gives a true picture of the situation in Kenya today.

2.5 Factors that Influence Research Productivity Cole (1979) conducted a study on a scientist's productivity in regard with age. The study revealed that research productivity did not vary with age. The finding conflicted with accepted wisdom and with an earlier literature epitomized by Lehman's Age and Achievement, that had found that scientists' productivity often reached a peak relatively early in life and then declined steadily and significantly. The study used academicians who eventually became professors in highly ranked departments. Another related study found that researchers were productive, in terms of publishing, between 30 and 79 years old, reaching a peak of 1.76 papers per year by the time they are 53 years old GonzalezBrambila &Velosob (2003). This study used a unique data set of Mexican researchers to explore the determinants of research productivity. The study findings confirm a quadratic relationship between age and productivity. However, productivity peaks when researchers were approximately 53 years old, 5 or 10 years later than what prior studies had shown. These results suggested that age is not very important in terms of research productivity. Results also show great heterogeneity

cxliii

across areas of knowledge. Interpretations of other aspects, such as gender, country of PhD, cohort effect, among others are also discussed. The present study has studied both gender and age of the participating respondents.

Ostmoe (1986) carried out a study to identify factors which were important in the publication productivity of university nurse faculty. Two central research questions were addressed: 1) what relationship exists between selected professional, educational, and career variables and the publication productivity of university nurse faculty members? 2) What is the typical publication productivity profile of university nurse faculty? The population consisted of 422 full-time tenure tract nurse faculty teaching in seven nursing schools that offered baccalaureate, masters and doctoral programs and were located in public Research Universities. All data were obtained through the use of a questionnaire. Completed questionnaires were received from 80% of the respondents. Faculty not meeting the criteria for the study and all instructors were eliminated from analysis. Data were ultimately analyzed for 261 subjects. Thirty-two variables were found to have a significant relationship to faculty publication productivity. Eleven of these variables (highest degree, years since first master's, age, rank, teaching responsibilities, time spent teaching, time spent in research,

cxliv

hours of clinical instruction, teaching and research preferences, journals received, beliefs about the desirable relationship between publication and promotion and tenure) and five motivational variables were included in a regression analysis. These 16 variables grouped into three clusters, accounted for .4845% of the total variation in university nurse faculty publication productivity. Current job socialization factors and motivational factors accounted for a significant amount of variation in faculty publication productivity even when highest degree, years since first master's, age, and rank were controlled. This study focused on research universities only. It was not representative of the universities. The present study has sought to investigate factors influencing research productivity in public and private universities in Kenya. The results, conclusions and recommendations are presented in chapter five of this study. Richards (1987) conducted a study to explore patterns and correlates of research productivity of members (N=2,713) of the Population Association of America (PAA). Five measures of research productivity for the years 1981 to 1985 were examined: (1) number of presentations at PAA meetings; (2) number of times appeared as an author in the population journal "Demography"; (3) number of times served as editorial consultant for "Demography"; (4) number of publications listed in "Population Index"; and (5) number of times listed as principal investigator in the federal inventory of population research. The

cxlv

results revealed that the five criteria of research productivity were related to predictors, including demographic characteristics, educational and work background, and past productivity. Measures of research productivity were highly skewed. The best predictor of research productivity was previous productivity in the same area. Productivity was also associated with younger age, a research-oriented educational background, and perhaps a greater commitment to the population field. The results illustrate the usefulness of data from public records for research on the psychology of scientists. This study focused on one study only. It was also carried out in the late 80s. The present study has provided up to date information. Wilson and Wilson (1989) undertook a study on home economics education administrators' attitudes of factors contributing to research productivity of faculty. One hundred and sixty administrators responded to the questionnaire. Twenty-three percent came from public land-grant institutions, 50% from public non-land-grant institutions, and 26% from private institutions. Logistic regression models were used to analyze the data. Findings of the study showed that a majority of the administrators in all three types of institutions believed that the research productivity level of their faculty was not high. Results also indicated that administrators in the study perceived faculty morale, faculty development, expertise in writing for publication, and teaching productivity as factors

cxlvi

contributing significantly to faculty research productivity. The present study went way beyond administrators and focused on the university academic staff. Waworuntu & Holsinger (1989) carried out a study on factors influencing the research productivity of Indonesian faculty in public higher education. These factors were divided into three broad categories corresponding to: (1) ascriptive, (2) achievement and (3) organizational variables. As the validity of the productivity concept is difficult to ascertain, three different measures were employed. These were: (1) a simple summative count of all self-reported scholarly writing, (2) a summative scale with each separate item weighted, and (3) a subjective self-evaluation measure. Data was collected for the specific purpose of this study at the end of 1982. From the population of 20,945 tenured faculty in Indonesia, 11,269 representing 54% were interviewed using an extensively pretested interview schedule. Using multiple regression analysis, models were specified for each of the three dependent variables. Following initial breakdown analysis of means, a multiple regression analysis was conducted. The three groups of predictor variables did not have equal influence on research productivity. The ascriptive variables had the smallest association with all three dependent variables. The achievement block of variables had almost three times as many significant coefficients than did the organizational block in the two objective

cxlvii

count measures of productivity. In the specification for self-evaluation, the organizational block was a slightly larger number of significant coefficients. This study was carried out in the late 80s; the present study presents investigated current factors influencing research productivity. The use of a simple summative count of all self-reported scholarly productivity has been borrowed to be used in the present study. Yu (1998) carried out a study on Sex differences in research productivity. The study found empirical evidence based on a systematic and detailed analysis of data from four large, nationally representative, cross-sectional surveys of postsecondary faculty in 1969, 1973, 1988, and 1993 in the USA. The research yielded two main findings. First, sex differences in research productivity declined over the time period studied, with the femaleto-male ratio increasing from about 60% in the late 1960s to 75 - 80% in the late 1980s and early 1990s. Secondly, most of the observed sex differences in research productivity were attributed to sex differences in personal characteristics, structural positions, and marital status. This study undertook a study on one variable only. The present study has sought to find out all those factors that influence research productivity. Jones (1998) undertook a study whose purpose was to determine factors that were associated with increased individual research productivity among clinical faculty in 67 United States and Canadian schools of dentistry. Individual faculty research productivity

cxlviii

was defined as the total number of articles in refereed journals and book chapters published during an academic career. The 328 respondents represented a response rate of 62.8% from a 25% stratified random sample of faculty who (1) had full-time appointments and held at least a degree in dental surgery. Or foreign equivalent, (2) taught in a clinical department of the dental schools, and (3) were not department chairpersons and did not hold administrative positions (assistant dean, associate dean, or dean) within the dental school. Respondents reported a mean of 9.9 years in full-time dental education, a mean of 10.8 publications, and a mean of 7.5 hours spent in research per week. Forward addition multiple regression analysis demonstrated that five predictor variables, from a total of 20 variables evaluated, accounted for 59.9% of the variance in individual faculty research productivity. These predictor variables were total dollar amount of past research funding, career age, training status, colleague utilization in conducting research, and conducting research from planned goals. This study was narrow since it was concerned with individual factors that determine research productivity. The present study has investigated the institutional factors that influence research productivity. Babu (1998) designed an investigation with the sole purpose of establishing factors which determine the productivity of scientists in India. Nearly 200 variables influencing research productivity were collected through relevant literature, analysis of biographies of

cxlix

great scientists, and discussion with eminent scientists. Finally, through a critical examination, 80 variables were selected for the use of Q-sort technique. The sample for the study consisted of a cross section of scientists ranging from Fellows of Indian National Science Academy to young agricultural scientists. Mailed questionnaires and personal interview methods were used for collecting data. Out of a total of 912 respondents, reply was obtained from 325. On the basis of Q-sorted data, 26 variables were selected for further analysis and they were subjected to principal component factor analysis. The results indicated eleven factors affecting research productivity of scientists. They were: persistence, resource adequacy, access to literature, initiative, intelligence, creativity, learning capability, simulative leadership, and concern for advancement, external orientation, and professional commitment. This study delved on sole purpose of establishing factors which determine the productivity of scientists in India. Other data like demographic factors could have offered some meaningful insights into his study. The present study has investigated other factors like demographic factors to inform this study. Jones (1998) undertook a study to evaluate individual faculty research productivity with respect to gender among clinical faculty in 66 United States and Canadian schools of dentistry. A comprehensive survey instrument was developed to collect information on

cl

factors associated with research productivity of individual faculty. The investigation focused on time spent per week in various work related activities (teaching, administration, research, and private practice), external grant money obtained for research, and responses to 12 variables evaluating the subject's attitudes of their research background, work environment, attitude and outcome effects from publishing, and the use of colleagues in conducting research. The 833 respondents represented a response rate of 69.4% (833/1200) from a 50% stratified random sample of faculty who (1) had full-time appointments and held at least the DMD or DDS degree or the foreign equivalent, (2) taught in a clinical department of the dental school, and (3) were not departmental chairpersons or administrators. The majority of subjects responding to the survey were male (705 males, 84.6%; 128 females, 15.4%). Respondents reported a mean of 10.3 years (males = 10.8, females = 7.5) in full-time dental education and a mean of 10.1 career publications (males = 11.6, females = 6.5; P < 0.001). Although there was no significant difference in weekly hours devoted to academic responsibilities between males and females, several factors did demonstrate significant gender differences (external grant money obtained for research purposes, P < 0.03; feeling that the departmental chair did not emphasize research, P < 0.05; feeling a lack of autonomy within their institution, P < 0.007; and feeling a lack of available colleagues for

cli

research purposes, P < 0.001). The implications of the findings are discussed concerning strategies for improving research productivity for females in academic dentistry. These results suggest that sex differences in research productivity stem from sex differences in structural locations and as such respond to the secular improvement of women's position in science. Basically this study was on gender and present study may not wish to go into the issue of gender in research productivity, this can be made as a recommendation for future studies. Fox (2005) carried out a study in the United States on the relationship between marriage, parental status, and publication productivity for women in academic science, with comparison to men. Findings indicated that gender, family characteristics, and productivity were complex considerations that went beyond being married or not married, and the presence or absence of children. For women in particular, the relationship between marriage and productivity varied by type of marriage: first, compared with subsequent marriage and occupation of spouse (in scientific compared with non-scientific occupation). Further, type of family composition was important: women with preschool children had higher productivity than women without children or with school-age children. Women with preschool children were found to be a socially selective group in their characteristics, particularly in their allocation of time. Marriage and parental status were narrow factors to

clii

be used for a study. The present study has gone beyond gender and investigated other factors that influence research productivity. Sax, Hagedorn, Arredondo and Dicrisi (2004) came up with a study that explored the role of several family-related factors in faculty research productivity for a large, nationally representative sample of university faculty members. The role of marriage, children, and aging parents was examined after controlling other personal and environmental factors, such as age, rank, department, and intrinsic motivations to conduct research that previous research had shown to influence research productivity. Analyses were conducted on a sample of 8,544 full-time teaching faculty (2,384 women and 6,160 men) at 57 universities nationwide in the United States. Results showed that factors affecting faculty research productivity were nearly identical for men and women, and family-related variables, such as having dependent children, exhibited little or no effects. The present study did not go into details or characteristics of the family. Dundar and Lewis (1998) summarized results of several studies carried out in the United States on research productivity. In this summary they found individual and organizational attributes to affect research productivity. Individual attributes included: innate abilities - such as IQ, personality, gender and age - and personal environmental influences. Inconsistent results were reported when these variables were investigated.

cliii

They further identified as personal environmental factors the quality and culture of graduate training and the culture of the employing department. They argued that most research studies have found a positive correlation between departmental culture and research productivity. The departmental culture referred to shared values and attitudes within the academic unit. Faculty and administrators who learned to place a high value on research as graduate students tended to foster a research-oriented culture throughout their professional lives. This study was about summarizing what other studies had found out. Campion and Shrum (2004) carried out a study to find out why women were finding it more difficult to pursue research careers than men. Based on a survey of 293 scientists in Ghana, Kenya, and the Indian state of Kerala, the study examined gender differences on a variety of individual, social, and organizational dimensions. The results showed small or nonexistent differences between women and men in individual characteristics, professional resources, and the organizational conditions under which research was conducted. There was definitely a need to conduct comparative studies in research productivity but this has not been handled by the current study. The fact that this study was done in the developing world atmosphere is an assurance that more studies need to be done in this same line just as what the current study has done.

cliv

Bozeman (2005) conducted a study based on the curricula vitae and survey responses of 443 academic scientists affiliated with university research centers in the USA, an examination of the longstanding assumption that research collaboration had a positive effect on publishing productivity was done. Since characteristics of the individual and the work environment were endogenously related to both collaboration and productivity, the study focused on the mediating effect of collaboration on publishing productivity. By using the two-stage least squares analysis, the findings indicated that in the presence of moderating variables such as age, rank, grant, gender, marital status, family relations, citizenship, job satisfaction, perceived discrimination, and collaboration strategy, the simple number (normal count) of peer-reviewed journal papers was strongly and significantly associated with the number of collaborators. However, the net impacts of collaboration were less clear. When the same model was applied and examined, productivity by fractional count, dividing the number of publications by the number of authors, it was found that number of collaborators was not a significant predictor of publishing productivity. In both cases, normal count and fractional count, significant effects of research grants, citizenship, collaboration strategy, and scientific field was found. This led to a belief that it is important to understand the effects of the individual and environmental factors for

clv

developing effective strategies to exploit the potential benefits of collaboration. This study concentrated on research collaboration and did not involve the institutional and individual factors responsible for research productivity. These have been well covered in the present study. Bland et al (2005) conducted a study to test the ability of the Bland et al. (2002) model-based on individual, institutional, and leadership variables influencing faculty research productivity-to explain individual and group (department) research productivity within the context of a large medical school. This study used data from the University of Minnesota Medical School. Twin Cities vitality survey conducted in 2000 had a response rate of 76% (n = 465 faculty). A statistical software package was used to conduct t tests, logistic regressions, and multiple regressions on these data. The validity of faculty, department, and leadership characteristics identified in the Bland et al. (2002) model were confirmed as necessary for high levels of research productivity. Faculty productivity was influenced more by individual and institutional characteristics; group productivity was more affected by institutional and leadership characteristics. The study made two conclusions. First, that the characteristics and groupings (individual, institutional, and leadership) in the Bland et al. (2002) model predict faculty research productivity. Secondly, research productivity is influenced by the interaction of

clvi

the three broad groupings, and it is the dynamic interplay of individual and institutional characteristics, supplemented with effective leadership, that determines the productivity of individuals and departments. Hadjinicola (2006) carried out a study to investigate factors that promote research productivity of Production and Operations Management (POM) groups of researchers in US business schools. In this study, research productivity of a POM group was defined as the number of articles published per POM professor in a specific period of time. The paper also examined factors that affect research quality, as measured by the number of articles published per POM professor in journals, which had been recognized in the POM literature as an elite set. The results of this study showed that three factors increased both the research productivity and the quality of the articles published by professors of a POM group. These factors were (a) the presence of a POM research center, (b) funding received from external sources for research purposes, and (c) better library facilities. Doctoral students do assist in improving research quality and productivity, but they were not the driving force. These results had important implications for establishing research policy guidelines for business schools. For example, real-world problems were funded by external sources and had a higher probability of publication. Furthermore, schools could place more

clvii

emphasis on external funding, as most engineering schools do, since groups receiving external funding are more productive in terms of research. This study did not look at factors that hinder research productivity. The present study investigated both factors that promote and hinder research productivity in selected public and private universities in Kenya and brought on board more faculties as basis of this study.

Cepero (2007) undertook a study in the USA to expand knowledge about faculty productivity and the institutional and individual factors that may contribute to increased levels of faculty scholarly productivity. A sample of full time faculty from the restricted 1999 National Study of Postsecondary Faculty database was used to address the two main research questions. The first research question related to procedures that can be used to statistically model faculty productivity, and was answered using Structural Equation Modeling (SEM) techniques such as Confirmatory Factor Analysis (CFA), Multiple-group CFA, Latent Profile Analysis and CFA with covariates. The second research question related to the factors associated with high levels of faculty productivity, and was investigated using a set of explanatory models using Multilevel Logistic Regression Modeling. Three factors (inflation, sole productivity and joint productivity) were used to model the productive behaviors of the sample. Several

clviii

independent variables, such as gender, rank, and type of institution were positively and negatively associated to the three dependent variables. These variables differed across the three factors in the model of scholarly faculty productivity. These outcomes suggested some level of independence that existed among the variables associated with each of the factors in the model of productivity developed as part of the study. The results of this research had important implications for university administrators and agency directors who develop policies designed to foster faculty professional development. This study also provided valuable data for researchers in the field of gifted education and talent development related to fostering creative productivity and talent development in adulthood. A recent scholar noted the following; Future research should include a more comprehensive set of indicators of faculty productivity that also address other dimensions of faculty scholarly responsibilities such as teaching productivity. These future studies may lead to more accurate understandings of a broader conception of scholarship in higher education institutions (Cepero, 2007; 2). The present study has identified comprehensive indicators influencing faculty research productivity. Tower, Plummer, and Ridgewell (2007) noted that differences in terms of sex research productivity can contribute to differences in earnings. Scholarly production is typically a significant factor in determining earnings and promotions, and many authors

clix

note that women faculty members publish less on average than their male counterparts. Thus gender differences in publication rates explain, at least partially, differences in average earnings and promotion rates between men and women. Numerous studies have shown that research production in Africa is highly skewed; South Africa accounts for one third of Africa's publication output, Egypt and Nigeria jointly account for another one third, result that is congruent with the findings of previous studies. (Tijssen, 2007; Paraje et al, 2005; Uthman & Uthman, 2007) Tower et al (2007) examined top six journals in the world and found no difference between women and men productivity when the percentage of women participating in the academic work force is factored in. He noted that women had a 30-35% participation rate in academic university positions and represented almost 30% of the authors in the top tiered journals. There were also no significantly statistical differences in Journal Impact Factor ratings between men and women. These findings were found to be consistent across all the major disciplines, science, business and social science. Other trends were noted such as the significantly higher number of authors in science journals and the different trends between US and non-US authors. Science authors quality (as measured by Journal Impact Factor (JIF of 31.9) was significantly higher than non-science authors (JIF 6.5); thus differences in quality were discipline specific not a gender issue. The implications were

clx

that academic womens research contribution matches that of a mens productivity. This study relied on journal publications only for analysis. The present study carried out an investigation involving the lecturers themselves. Hassan, Tymms and Ismail (2008) carried out a study whose purpose was to explore the perspectives of Malaysian academics in relation to academic productivity and some factors affecting it. A large scale online questionnaire was used to gather information from six public universities. The most productive role in the eyes of the academics was found to be teaching, with research and administration coming second and third, respectively. Several factors were found to be related to productivity and some of these had policy implications. The universities themselves differed markedly and research productivity was related to the amount of time available, and linked negatively to the teaching load. Hassans study concentrated on public universities only. The present study sought to investigate factors that influence research productivity in selected private universities in Kenya. Usang, Udey, & Akuegwu (2007) carried out a study that examined academic staff research productivity in Universities in South-South zone of Nigeria. Ex post facto design was adopted for this study. The sample size comprised of 480 academic staff drawn from a population of 3120. Data collection was carried out using a researcher constructed

clxi

instrument called Academic Staff Research Productivity Inventory which was validated and pilot tested. The data obtained were treated statistically using Independent t-test and contingency Chi-square (X2) analyses. Results indicated that male and female academic staff differed significantly in their research productivity; married and single academic staff differed significantly in their research productivity and there is a significant influence of areas of specialization on academic staff research productivity. It was recommended that academic staff in universities should be encouraged to carry out research work irrespective of their gender, marital status and areas of specialization. It is interesting to note that the current study has not found significant difference on research productivity in terms of gender. Lertputtarak (2008) carried out a study in Thailand on one of the public Universities. The conceptual framework for her research was chosen to integrate empirical research findings on faculty role performance and productivity with two existing motivation theories, namely Expectancy Theory and Efficacy Theory. The research methodology used a qualitative research approach, based on in-depth interviews with eleven representative respondents from The Noble University. Lertputtarak came up with five factors that impact on academic research productivity. These were environmental factors, institutional factors, personal career development factors, social contingency

clxii

factors, and demographic factors. According to the findings of this study, these five factors were divided into three main groupings which were termed the essential factors, desirable factors, and side-affect factors. Each of these factors, she claimed, needed resolution, in a sequential way, by administrators of the university. This study did not provide room for triangulation. The present study has employed quantitative method though it has also had to involve some qualitative aspects particularly when analyzing the open ended items in the questionnaires for triangulation purposes. Perhaps the newest study in research productivity was conducted by Ogbogu (2009) who examined the research output of female staff and the factors that affect their research productivity in the Nigerian university system. The study was carried out with a view to promoting strategies that could enhance productivity and increase the research output of female staff in Nigerian universities. The study adopted a survey research design. The purposive sampling method was used in administering questionnaires to 381 female academic staff from twelve randomly selected universities in the six geo-political zones of Nigeria. The study revealed that female research output was generally low: 59.5 per cent of female academics published one paper annually; 23.6 per cent published up to two papers; 1.1 per cent published three papers; and 15.8 per cent did not publish on an annual basis.

clxiii

Although most female academic staff published annually, most wished to increase their publication rate. The study's results found that marital status, religion, academic position and number of hours of lectures per week had an impact on their ability to carry out research and publish the results. The study concluded that female academics made contributions that are more significant to teaching than research and that the Nigerian university system needs to develop strategies to enhance female research output. Religion is a sensitive issue and the present study has not used it as a factor influencing research productivity. However other factors such as gender, age group, rank have been used in the present study with very interesting results. 2.5.1 Factors associated with Personal Career Development Fulton and Trow (1974) observed that 29% of the full professors, 20 percent of the associate professors, percent of the assistant professors and 2 percent of the instructors has published five or more articles in a two-year period. This work accords with the findings of Bailey (1992) who pointed out that rank is a significant predictor of research productivity. Dundar and Lewis (1998) found that departments with higher ranked faculty resulted in higher research productivity (Vasil 1992).

clxiv

Beehr, Walsh and Taber (1976) indicated that role stresses can interfere with the way in which a person interprets the notion that working hard and effectively will bring about the satisfaction of higher order needs. These authors also suggested that role stresses may adversely affect workers who strongly value the task attributes of enriched work. In a similar study, Pfeffer and Langton (1993) reported job satisfaction was positively related to productivity, and noted that staff opinions of their personal circumstances may influence productivity, whether it is an opinion of job satisfaction, research/ training environment, funding adequacy or the freedom to collaborate. It has been suggested that interest in research can be the best predictor of research productivity. However Blackburn et al. (1991) found this variable not to satisfactorily predict productivity. The top academic institutions generally produce a high level of research productivity because high-status universities enjoy advantage in terms of financial resources and research support that encourage publication. Reskin, (1977) pointed out that a process of homosocial reproduction is common within business schools, so that graduates of high-status universities are hired by other high-status institutions. Homosocial is also referred to as cumulative advantage. This refers to a staff members prior academic and professional training. The attribute of accumulative advantage makes it easier to achieve success in publishing because of prior research project

clxv

experience, research membership, development of research skills, and collaboration on research project and research sponsorship (Creswell 1985; Fox 1996). According to Finkelstein (1984), academic rank is a significant predictor of publication success because the academic lecturers in higher ranks generally have more control over their workload assignment, allowing faculty of higher rank to produce more research than those of a lower rank. Personal career development factors are those factors that come from the academic and personal qualifications of academic lecturers themselves. These factors include such items as an individuals ability and interest, attitude toward conducting research, academic origin, the type of advance degree earned, research experience, skills and training, rank and tenure status. In a similar way, a staff members attitudes and commitment to scholarly work relates closely to their research productivity (Lertputtarak, 2008)

2.6 Research Productivity and Academic Disciplines Studies assessing the research productivity of departments in academic institutions or individual researchers have been conducted in various disciplines such as in general business (Niemi, 1998); Baden-Fuller et al. 2000), management (Stahl et al. 1988), marketing (Niemi,1988), finance (Klemkosky and Tuttle,1977; Ederington 1979;Niemi,

clxvi

1987) accounting (Jacobs et al. 1986), management information systems (Vogel and Wetherbe 1984; Grover et al. 1992), and economics (Laband, 1985). These studies ranked departments according to the number of articles published in the refereed journals of their relevant fields. They did not provide factors that influence research productivity of lecturers that is what the present study did. Biglan (1973) found that researchers from natural sciences and engineering published more journal articles, in contrast, researchers from social sciences and education tended to publish more monographs, results that seem to be consistent with more current studies (Brooks, 2006). In addition, Biglan found that researchers in applied fields (e.g., engineering, education, and economics) produced more technical reports than scholars in no applied fields. Brooks (2006), in a study that compared three cohorts of the NSOPF (1988, 1993, and 1999), found that social science and natural science/engineering faculty members were expected to demonstrate higher rates of research dissemination than faculty members from the field of humanities. In contrast, faculty members from humanities had a tendency to publish higher numbers of books and chapters in edited books. Hannington (1986) undertook a study to identify the characteristics of the active research producers within dental schools. A survey was mailed to 4,901 full-time faculty members in 53 U.S. dental schools, of which 1,481 (31%) were returned. Faculty Research

clxvii

Productivity (FRP) was defined as the number of publications generated by a faculty member during his or her academic career, as reported in response to a survey question. Seventeen characteristics of faculty members also were obtained from responses to survey items. Using stepwise multiple regressions, five variables predicted 38% of the variance in FRP: Interest in Research, Earned Ph.D., Number of Journal Subscriptions, Consulting Time per Week, and Research Time per Week. While the relationship between FRP and student contact time was linear (the more articles produced the less student contact time), time spent in teaching per week did not enter the prediction equation. The present study has focused on active researchers and has sought to get views from them with a view to finding out the factors that influence their research productivity. Kohlenberg (1992) carried out a study to identify the relationship between faculty research productivity and organizational structure in schools of nursing. This was necessitated by the fact that nursing research had been widely recognized by members of the nursing profession, yet comparatively few engaged in conducting research. He argued that although contextual variables that facilitate or inhibit nursing research had been investigated, the relationship between organizational structure and nursing research productivity had not been examined. This problem was examined within the context of the Entrepreneurial Theory of Formal Organizations. A survey methodology was used for data

clxviii

collection. Data on individual faculty research productivity and organizational structure in the school of nursing were obtained through the use of a questionnaire. A random sample of 300 faculty teaching in 60 masters and doctoral nursing schools in the United States was used. The instruments for data collection were Wakefield-Fisher's Adapted Scholarly Productivity Index and Hall's Organizational Inventory. The data were analyzed using Pearson Product-Moment Correlation Coefficients and multiple correlation/regression techniques. The study found out that the overall relationship between faculty research productivity and organizational structure in schools of nursing was not significant at the . 002 level of confidence. The present study has also studied individual factors responsible for research productivity. Mills, Zyzanski, Flocke (1995) carried out a study whose objective was to describe residency research productivity and identify the set of independent factors that best characterize programs at various levels of productivity. A 23-item survey was mailed to 226 randomly selected family practice residency directors. The survey included items on program demographics, mentoring, resident and faculty research activities, and program research resources. Factor and discriminant analyses were performed to identify the major independent factors associated with productivity. A total of 154 completed surveys were

clxix

received for a response rate of 68%. Based on a cross tabulation of grants per program and publications per faculty, 22% of programs had high productivity, 46% had medium productivity, and 32% had low productivity. The significant factors of mentor support, amount of research activity, and program size contributed independently to the classification of programs by relative level of research productivity. These associations remained significant when university programs were excluded. The study concluded that Family practice residencies with relatively higher research productivity are more likely to have three characteristics than lower productivity programs: availability of a research mentor, more faculty research activities, and larger program size. This study focused on directors instead of targeting the researchers themselves. This study sought information from both researchers and their heads of departments. Hemlin & Gustafsson (1996) explored the main factors influencing the research productivity in the arts and humanities in Sweden. A questionnaire was constructed to identify and assess the effects of various factors important for the productivity of the individual researcher as reflected in the number of papers and PhDs produced. First, respondents were given the opportunity to list in their own words a number of important factors influencing research productivity. Secondly, they evaluated on rating scales the importance of a number of pre-selected factors (e.g. individual characteristics,

clxx

organisational features, external factors) assumed to be important for research productivity. 50% of a sample of 256 researchers in the humanities responded. Ratings were grouped to produce a number of indices and these were subject to multiple regression analyses. The main results showed that the production of papers was predicted by the number of Ph.D.'s produced and inversely related to the importance of organisational factors. The production of Ph.D.'s was dependent on the year of the Ph.D. and the position of the respondent as well as on the number of papers s/he produced. A number of conclusions were drawn: a) there was support for the academic social position effect also in the humanities; b) organisational factors apparently played a minor role in comparison to individual characteristics in the humanities than in the sciences and; c) the differences in productivity of papers were also related to gender, but not to size, area or language of publications. The present study went beyond studies in humanities and social sciences and incorporates other faculties in the university setup. This has indeed yielded a more representative type of information from the universities. Bonzi (1992) collected data from the curriculum vitae of 411 senior faculty at Syracuse University which was analysed to uncover trends in productivity over time. Results showed that productivity was related to status and academic discipline. Overall, productivity earlier in a career was a good indicator of later productivity. Increase in

clxxi

productivity among females was greater than among males, but males were more productive overall. Humanities and science/mathematics faculty increased productivity to a greater extent than did social scientists and professional school faculty. Journal articles were cited much more than are book chapters, so faculties not tending to publish in journals did not show a high citation count. In addition, citation counts were biased toward older works, since they had greater exposure than more recent works. Focusing on senior faculty only could not give objective results to a study. This study has been all inclusive in terms of lecturers who were selected to take part in this study. Wanner, Lewis and Gregorio (1981) conducted a comparative study on Research Productivity in Academia in the US. He noted that a significant number of studies of scholarly productivity were accumulated in the past and a majority focusing on limited samples of specialists in one or only a few scientific disciplines, making it difficult to generalize findings across dissimilar academic disciplines. Wanner et als paper tested a model incorporating both academic and nonacademic factors as determinants of scholarly productivity with samples of physical and biological scientists, social scientists, and humanists taken from the 1972-73 in the American Council on Education survey of faculty at U.S. institutions of higher learning.

clxxii

The study found considerable variation in the process determining productivity both across the broad disciplinary categories as well as within categories when article and book productivity were compared. The study also examined the relative influence of the disciplinary context and attributes of scholars on productivity. The evidence found suggested that the decisive edge that physical and biological scientists enjoy over social scientists and humanists in article productivity was largely the result of the nature of work or a favorable disciplinary milieu, while the lower rate of productivity among humanists was more heavily determined by their attributes. Kotrik (2002) conducted a study to determine the factors that explained the research productivity of agricultural education faculty in colleges and universities. In this study, publications in refereed journals were used as a surrogate for research productivity. The study described the research productivity of agricultural education faculty, their attitudes of the organizational culture that exists in their department to support research productivity, and their self-assessment of their research competency. The population for the study included all full-time, professorial rank faculty employed by colleges and universities in the United States that offered agricultural education. The regression analysis was used and it revealed that three variables explained 50% of the variance in research productivity. These variables included number of doctoral

clxxiii

students advised to completion in the last five years, faculty members attitudes of their research confidence, and the number of graduate assistant hours allocated to the faculty member. The variables that did not explain a significant proportion of the variance were percentage of the faculty members time allocated to research, salary, organizational culture and support of research, age, gender, rank, number of masters students advised to completion in the last five years, and number of years they had held a tenure track position. This study gives an indication of the variables to consider in my present study. Borokhovich et al (1995) examined differences in finance research productivity and influence across 661 academic institutions over a five year period from 1989 through 1993. They found that 40 institutions accounted for over 50% of all articles published by 16 leading journals over the five year period; 66 institutions accounted for two-thirds of the articles. Influence was more skewed with as few as 20 institutions accounting for 50% of all citations to articles in these journals. They further found that the number of publications and publication influence increased with faculty size and academic accreditation. Prestigious business schools were associated with high publication productivity and influence. The present study has zeroed in on both academic staff and institutions. Print and Hattie (1997) carried out a study to demonstrate, via the use of the discipline of Education, a procedure to identify and weight the importance of various

clxxiv

indicators of research productivity which in turn have become significant components in determining quality within and between universities. The methodology allows for the identification of indicators that are most important, and ascertains if there are differences among academics as to the relative weighting of the various research indicators. Highly valued indicators of research productivity amongst the Education academics were refereed journal articles, peer reviewed books, and major competitive research grants. Refereeing was critical in the determination of quality in research productivity, and the findings generalized across many academics regardless of their own personal productivity. It is recommended that the methodology can serve to determine the tacit weights that academics within and across disciplines attach to various research products. At least, this method makes academics and administrators aware of the weightings they are actually using when making decisions about the quality of academic departments. Hasselbacka, Reinstein, Schwanc (2000) conducted a study which made use of comprehensive data on both the quantity and quality of research productivity of 3878 accounting faculty who earned their accounting doctoral degrees from 1971 to 1993. Publications in 40 journals were used to measure faculty publication quantity. Journal ratings derived from a compilation of the rankings of five prior studies and co-authorship were used to measure publication quality. Choosing benchmarks for an individual faculty

clxxv

required users of the data to determine four parameters: (1) what credit to give a faculty member for co-authored articles; (2) what level of journal quality was appropriate, (3) choosing appropriate levels of performance and (4) deciding the emphasis to place on the number of years since the doctoral degree was earned. It was discovered that the average number of authors per article was significantly correlated with time and growing at a pace of 0.017 authors per article per year.

Wierzbicki (2002) conducted a study to investigate the research productivity of all staff in chemical pathology in the United Kingdom. Chemical pathologists or biochemical scientists were identified from publicly available sources. All journals, their impact factors (IFs), and individual publications over the period of 1995 to 1999 were identified from electronic databases. Each publication was sub classified with respect to type of publication, number and position of author, and subspecialty to which the article referred. Results of this study indicated that research output over the period comprised 6162 articles, originating from 1399 individuals, 264 of who were medically qualified. Specialty initiated research accounted for 26% of the total publications and 80% of the research was performed in teaching hospitals.

clxxvi

Research output was highly skewed because 49% of individuals published a letter or more, 20% published one original piece of research over five years, but only 4% were research active, as defined by one publication each year. International standard research, defined as one paper each year in journals with IF > 4, was achieved by 1% of the profession, mostly aged > 55 years. Skewed distributions of publication rates were found in all age deciles. The possession of higher research degrees correlated with higher output in all age deciles. The conclusions formed from this study indicate that those working in chemical pathology are active in initiating and conducting research, although at a low level. Because long-term activity in research correlates with the possession of higher research degrees and the opportunity to carry out research from early in career pathways, priority should be given to encouraging research in training. The present study has gone beyond Chemical pathologists and investigated more researchers in different disciplines. Chen (2004) surveyed 320 faculty members in a study from 10 business schools to examine the intrinsic and extrinsic rewards that motivate faculty to conduct research. Of the thirteen rewards studied, receiving or having tenure was the most important reward, while getting a possible administrative position was the least important. There were significant differences in the importance of these rewards between tenured-untenured and

clxxvii

between male-female faculty members. Faculty perceives a strong link between research productivity and the attainment of the rewards of tenure and of promotion. However, in the minds of the faculty, the link between publications and the reward of salary increases was not strong. Associate professors reported lesser importance than either full professors or assistant professors on nine of the thirteen rewards and perceived a weaker link between research productivity and achieving the reward. This implies that the associate professors were the least motivated faculty rank to perform research. There was no significant difference in the number of journal articles either published or accepted for publication within the past 24 months by tenure status, gender, or faculty rank. Rewards are not the only factors that can influence research productivity. Even institutional working conditions can increase research productivity. The present study has investigated more factors that can influence research productivity. Dakik (2005) analysed the quality and quantity of scientific publications of the medical faculty at the American University of Beirut (AUB) during a six year period (19962001). The study included all faculty members in the medical school of AUB in the year 2001. A Medline search inclusive of the years 19962001 was done for each faculty member and a total number of 881 publications were obtained. The faculty consisted of 203 members. Their average productivity rate (mean (SD)) was 1.24 (1.38)

clxxviii

publications/faculty member/year (PFY), with a mean impact factor of 2.69 (4.63). Eighteen per cent of the faculty did not have any publication in the six year study period, and only 20% had two or more publications per year. There was a significantly higher publication rate among newly recruited faculty members (0.93 (1.40) PFY for those appointed before 1990, 1.45 (1.24) PFY for those appointed during 19901995, and 1.67 (1.43) for those appointed after 1995, p = 0.007), and among those who were younger in age (p<0.01). Collaboration with international investigators resulted in more original publications than work done only at AUB (65% v 35%, p<0.001), and a higher journal impact factor for the publications (3.20 (3.85) v 1.71 (2.36), p<0.05). This was one of the first studies that analysed the research productivity of the medical faculty in a university setting in a developing country. It showed a wide variation in the research productivity of the faculty members that seems to be related to individual as well as institutional characteristics. This study did not elaborate on the factors responsible for research productivity of researchers. That is what the present study has strived to realise. Ramona, Lee, Skakun, & Lorne (2007) carried out a study to determine whether the sequence of training to obtain MD and PhD degrees was associated with different career paths for physicians who had their PhD before medical school and those who

clxxix

obtained it after their MD, and to explore the factors that encouraged or dissuaded Canadian dual-degree physicians in pursuing a research career. In 2003, questionnaires from the University of Alberta, Edmonton, Canada, were sent to all 734 Canadian physicians having MDs and PhDs; they were identified through the Canadian Medical Directory. Data collected were gender, year and country of MD, sequence of obtaining degrees, portion of time on clinical, research, teaching, and administrative duties, number of publications and grant amounts held, and perceived incentives and disincentives to research careers. Two focus groups were held with a subset of physicians to further explore themes. The response rate was 64%. On the basis of the timing of the PhD relative to the MD, physicians were designated early PhDs (26%), concurrent PhDs (12%), or late PhDs (62%). Late PhDs spent more time in research and less time on clinical practice than the other two groups and spent more time teaching and had published more papers than the early PhDs. Grant amounts were highest for late PhDs. Lack of time and resources were the major disincentives to research, and noteworthy incentives were the opportunity for intellectual challenge and creativity, and previous research experience. The study concluded that Physicians who obtained a PhD after an MD had a more research-focused career than those who entered medical school with a PhD. Sequence of training was a

clxxx

narrow way of analyzing research productivity. The present study has employed more factors and how they influence research productivity. Wills et al (2008) carried out a study on a meta-analysis of international studies from accounting and related business fields, published between 1988 and 2008, the study examined factors influencing the research productivity of academics. In this study, more than 70 factors were identified from 25 studies, which were then reduced to clusters of factors, or themes. A data-driven approach to thematic analysis was used to identify the factors and to allocate them to nine themes. The study examined the relationships between the themes and proposed a model of how they were linked. The study suggested that three hierarchical clusters of factors at government, institution and individual levels influenced the research output of accounting academics. This study concentrated on business subjects. The present study has involved other disciplines

2.7 Research Productivity and Technology Transfer/Patents Recently, researchers have proposed other indicators of creative productivity such as the number of patents, presentations at conferences, or textbooks published (Antony and Raveling, (1998). Brooks (2006) insisted we know that these indicators [traditional indicators of scholarship productivity] reflect only a narrow slice of an institutions overall

clxxxi

research and scholarship, and do not accurately represent the full variety of scholarly activity of faculty in all disciplines. The main problem in only using traditional indicators of productivity is that this omission might introduce systematic bias against those disciplines characterized by different type of indicators Brooks (2006) Donald (2003) undertook a study in the United Kingdom on assessing the impact of university science parks on research productivity. This was against a background that University science parks stimulate technological spillovers. Results of this study indicated that there was virtually no empirical evidence on the impact of these facilities on research productivity. This was done by examining whether companies located on university science parks in the United Kingdom had higher research productivity than observationally equivalent firms not located on a university science park. The results showed that there was no relationship between the science parks and research productivity. The development of Science parks are at the infancy stage in Kenya. Jean and Schankerman (2004) carried out a study on Patent Quality and Research Productivity: Measuring Innovation with Multiple indicators. They analyzed the

determinants of the decline in research productivity using panel data on manufacturing firms in the US for the period 198093. They focused on three factors: the level of demand, the quality of patents and technological exhaustion. They developed an index of patent

clxxxii

'quality' using detailed patent information and showed that using multiple indicators substantially reduced the measured variance in quality. They further found out that research productivity at the firm level was inversely related to patent quality and the level of demand, as predicted by theory and patent quality. In this case, patent quality was positively associated with the stock market value of firms. This confirmed the fact that research productivity positively influences quality of a patent. This present study has not explored levels of patent applications with KIPI, ARIPO or WIPO but has made suggestions for further research to explore research productivity in Kenyan Universities in regard to patent applications. Torrez (2006) reports about a study by the Milken Institute, an independent thinktank based in Santa Monica in the US, which sought to determine the universities worldwide that were doing the best job at technology transfer and commercialization of their discoveries and inventions in biotechnology. In their study titled Mind to Market: A Global Analysis of University Biotechnology Transfer and Commercialization, the institute found that the University of California system was the most successful university in licensing income from its discoveries and inventions, a total average of about $100 million per year, followed by Stanford University ($50 million) and the Massachusetts Institute of Technology ($33 million).

clxxxiii

The study highlighted the importance of research to a university's bottom line and its positive economic effects on its region. It also stressed the importance of a technology transfer office to a university and of a campus's proximity to clusters of biotech firms that can fund research. The present study recognizes the critical roles played by patent institutions in Kenya and in the region; however, this study has concentrated on publications by university lecturers as a yardstick for research productivity. Welker (2006) conducted a survey of senior university research administrators on a broadly defined set of their institutions research activities. In spring 2005, the senior research administrator at each of the 250 Carnegie classified doctoral/research universitiesExtensive and Intensive was identified (Shulman, 2001). They were contacted by email and asked to answer an on-line survey. The purpose of the survey was to gain a better understanding of national trends in research planning and priority setting, research culture, research publicity, economic development, and technology transfer. Ninety-five senior research administrators completed the survey, forming a response rate of 38%. Responses came from 35 states. Two-thirds of the respondents were from Doctoral/Research Universities -Extensive and one-third was from Doctoral/Research Universities -Intensive. Two-thirds

clxxxiv

represented universities with undergraduate enrollments of over 10,000; 72% were public and 28% were private institutions. The survey was intended to stimulate thought, not necessarily to recommend specific actions, and to provide insights into the management of research at U.S. universities.

2.8 Research Productivity and Teaching Effectiveness Kenneth (1987) made an analysis review of the research that had been done on the connection between research productivity or scholarly accomplishment of faculty members and their teaching effectiveness (as assessed by their students). On average, it was found that there was a very small positive association between the two variables. The association between research productivity and teaching effectiveness was explored further by considering whether its size and direction varied by career stage of faculty members, their academic discipline, and the type of college or university in which they taught. The current study has not gone into making comparisons between the teaching and research aspects of faculty members but has looked at the contribution of both lecturers and students in terms of research output.

clxxxv

Centra (2005) investigated the relationship between research productivity and teaching effectiveness to shed light on the long-debated question of whether performance in one area enhanced performance in the other. The academic field and the stage of a faculty members career were both considered in the analysis. Two samples one of 2,973 and the other of 1,623 faculty members from a variety of institutions, were studied. In considering results of both analyses, teachers of social science courses were the only group for which there were consistent though modest relationships between the numbers of published articles and student ratings of instructor effectiveness. Thus spillover effects, or a general ability factor, or other reasons for a possible link between research and teaching performance were not totally supported. The relationship between performance in the two areas was either nonexistent or, where it appeared, it was too modest to conclude that one necessarily enhanced the other. This study did not indicate the method of data analysis used. 2.9 Summary of Literature Review The literature review done in this work is in the area of research productivity at various levels. The research work done range from groups of institutions, individual institution, fields of specialization to individual variables that contribute to research productivity. From this literature review, it is clear that few studies have combined the

clxxxvi

analysis of both institutional and individual factors, it is clear that most studies have not used the university heads of departments to get information for triangulation purposes. This study has tried to cover this gap. It is also clear and evident that most of the work has been done in the developed countries and very little has been done in the developing world. The literature also indicates that most of the work has been done in the recent past meaning that this is a new emerging area that developing countries should embrace to gauge the productivity of their research capacity. Furthermore, very little work has been done in Africa and in Kenya in particular in this area. Perhaps this is the first comprehensive study to be carried out in Kenya targeting on research productivity; this is evident from the literature reviewed. This by itself is a justification for this work to be carried out in Kenya.

clxxxvii

CHAPTER THREE RESEARCH DESIGN AND METHODOLOGY 3 Introduction This chapter presents and justifies the research design and methodology employed in this study. It describes the research design, target Population, sample and sampling procedures, description of research instruments, test for validity and reliability, description of data collection procedures, description of data analysis procedures.

3.1 Research Design In this study, survey research design was employed. The survey design collects data at a particular point in time with the intention of describing the nature of the existing conditions, indentifying the standards against which existing conditions can be compared and for determining the relationship that exists between specific events (Orodho, 2004). This design is preferred because the researcher was interested in describing the existing phenomenon without any manipulation. This method was appropriate for this study because it enabled the researcher to collect much information in less than one months time. It was also possible to make a wide coverage of both public and private universities in Kenya in that short duration. This design has also been used because the findings of the

clxxxviii

study would be generalised to the whole population. Therefore survey research design was found to be appropriate for this study. 3.2 Target Population This study targeted academic staff in Kenya drawn from selected public and private universities. This study focused on research productivity of the university academic staff; this is because lecturers are mandated to teach and to conduct research in their areas of specialization. Therefore the university academic staff drawn from selected faculties in the mentioned universities formed the population of this study. Others were heads of departments drawn from the selected universities. 3.3 Sample and Sampling Procedures Stratified random sampling was used in this study. This is because the population studied was heterogeneous. There are differences among the lecturers in terms of gender, rank and academic disciplines. A total of 400 questionnaires were administered to the university academic staff in the ratio of 70 to 30 for male and female academic staff respectively. On the other hand, simple random sampling was used to target university heads of departments. This sampling procedure was preferred because only those who are charged with the said responsibilities as heads of departments were used.

clxxxix

The sampled universities consisted of five private and six public universities. For ethical issues the universities were randomly assigned letters A, B, C, D, E, F, G, H, J, K, and L. Letter I was skipped in this naming. Public universities were A, B, C, D, J, K while the private universities were E, F, G, H and L. The names of faculties and departments remained the same. This confidentiality was maintained due to the nature of sensitivity associated with performance of universities. Other universities in the sample had also requested for the confidentiality of the findings. 3.4 Description of Research Instruments This study employed the questionnaires and document analysis in data collection. 3.4.1 The Questionnaire The questionnaire was the main research instrument used for this study. It was employed to collect information from the university lecturers and heads of departments. It had both open and closed ended items. This study mainly used the questionnaire as the instrument for data collection. This was partly because most of the studies done in this area have used this instrument for data collection. It has also been proven that self-reported data has correlated positively with the information collected from journal publications. The instrument was developed and pilot tested to ensure that it was valid and reliable.

cxc

3.4.2 Questionnaire for Lecturers (Appendix 1) The questionnaire was used to collect six types of information from lecturers which were required for the purposes of this study. The questionnaire was structured as follows: Demographic information: contained information of the lecturers with variables such as name of university, faculty, department, gender, age group, rank, highest degree obtained, and work load. Information on publications: information sought here included research publications done by the lecturer between the years 2004-2008. Individual factors: a wide range of questions specifically targeting the individual factors were presented to the lecturers. The respondent was supposed to tick against the appropriate responses. The response was a four level scale ranging from 1 4, 1- Large Extent; 2 Some Extent; 3 Little Extent; 4 No Extent. Institutional factors: questions here centred on institutional factors influencing research productivity. The respondent was supposed to tick against the appropriate responses. The response was a four level scale ranging from 1 4, 1- Large Extent; 2 Some Extent; 3 Little Extent; 4 No Extent.

cxci

Attitude on research and publications: this contained a five point Likert type scale SA Strongly Agree; A Agree; U - Undecided/No opinion; D Disagree; SD Strongly Disagree. Personal opinions on problems and possible solutions: this section had questions that sought information on any factors affecting research productivity that had not been covered in the items above. The lecturers were also asked to give possible solutions to the problems/factors they had identified.

3.4.3 Questionnaire for Heads of Departments (Appendix I1) The questionnaire was structured as follows: Demographic information: contained information of the lecturers with variables such as name of university, faculty, department, gender, age group, rank, highest degree obtained, and work load. Information on publications: information sought here included research publications done by the lecturer between 2004-2008. Individual factors: a wide range of questions specifically targeting the individual factors were presented to the lecturers. The respondent was supposed to tick against the

cxcii

appropriate responses. The response was a four level scale ranging from 1 4, 1- Large Extent; 2 Some Extent; 3 Little Extent; 4 No Extent. Institutional factors: questions here centred on institutional factors influencing research productivity. The respondent was supposed to tick against the appropriate responses. The response was a four level scale ranging from 1 4, 1- Large Extent; 2 Some Extent; 3 Little Extent; 4 No Extent. Personal opinions on problems and possible solutions: this section had questions that sought information on any factors affecting research productivity that had not been covered in the items above. The heads of department were also asked to give possible solutions to the problems/factors they had identified. The scales and items used in the instruments was selected after a review of the literature and grounded in the theoretical base of this study. Each of the questionnaires had a short introductory letter signed by the researcher attached to it. During the data analysis, the items on the five point Likert scale were reversed in scoring negative questions. Hence the higher the percentage or the mean, the more important the factor was perceived to be. Document Analysis

cxciii

This was the other document used in the data collection exercise. The instrument was used in collecting information on the research policies for the institutions that had developed them. Six universities were sampled here. Four public universities and two private universities were used in this case. Main concern here was the weight given by individual universities towards personal career development of the individual academic staff. It yielded interesting results in this regard.

3.5 Piloting/Pre-Testing The purpose for pre-testing was to ensure that items in the questionnaire were stated clearly and had the same meaning to all respondents. The questionnaire was pilot tested with 10 university faculty members who did not take part in the study. 3.6 Validity and Reliability of Instruments 3.6.1 Validity Nachmias and Nachmias (2005) observes that validity is about answering a question, Am I measuring what I intend to measure? Coolican (1996) observed that validity of the instrument is the extent to which it measures what it is intended to measure. In this case, three experts in the faculty of education CUEA assisted the researcher in

cxciv

validating this instrument so that it was able to measure what it was intended to measure. They looked at them independently and ascertained that they could solicit the kind of information they were to collect. Their views on the content and structure were incorporated in the final draft of the instrument. Peer review was also used to enhance face and content validity. At this stage the items on attitudes of the academic staff to research and publishing were added to the instrument. Appropriate changes were also made during validation and were incorporated into the instrument. Other changes were done in the wording of items, the design of scales, and in the instructions for completing the instrument. The suggestions from the experts and peer reviewers helped to achieve the content and face validity of the instruments. 3.6.2 Reliability Reliability refers to the extent to which the measuring instrument contains variable errors, that is the errors that appear inconsistently from observation to observation during any one measurement attempt or that vary each time a given unit is measured by the same instrument (Nachmias and Machmias,2005; 170). Instruments used in Social sciences yield more errors than those used in physical variables. This reliability measure varies between 0 1. An instrument with a figure closer to 0 has more errors than one which is closer to 1 (Nachmias & Nachmias, 2005; 172).

cxcv

In this case the researcher employed the Split half method. The questionnaires were administered to a group of lecturers in pilot testing. Received responses were split into two by using odd and even numbers. Cronbach Alpha Coefficient yielded a coefficient which indicated that the questionnaires were reliable. If the coefficient realized was too low, some work had to be redone on the questionnaire until it yields satisfactory coefficient. In most cases, a value lying between 0.70 1.00 is commonly accepted (Nachmias & Nachmias, 2005). The reliability index for closed ended questions for academic staff was calculated using SPSS for Windows (version 15). The split half technique was used to analyse the ten questionnaires. A total of ten (10) questionnaires were used for this calculation. For this case the Spearman Brown coefficient was calculated and it yielded a coefficient of 0.796. According to Nachmias and Nachmias (2005) this coefficient is enough to justify the instruments use. The realized coefficient indicated that the instrument was reliable enough to collect information from the academic staff in Kenya. On the other hand, six questionnaires of the Heads of Departments was subjected to Cronbach Alpha Coefficient and it yielded 0.753 reliability index. This coefficient was regarded as qualifying research instrument to be used in this study.

cxcvi

3.7 Description of Data Collection Procedures The Department of Postgraduate studies in Education at the Catholic University of East Africa provided the researcher with a clearance letter. This letter made it possible to make an application for a research permit from the National Council for Science and Technology (NCST), in the Ministry of Higher Education, Science and Technology. The researcher used this permit to contact the university vice chancellors for clearance to conduct research in their respective institutions. Permission was granted from all the universities that were contacted. The only problem was that some of the vice chancellors took long to grant approval letters. The questionnaires were administered to the selected university academic staff by the researcher and research assistants. The questionnaires were self administered while others were sent by email. Due to the large scope of this study, involving universities outside Nairobi and the fact that university academic staff are scattered in offices across the university, it became prudent to engage the services of research assistants. All the research assistants were trained by the researcher and were assigned to specific universities. The package to each lecturer contained a cover letter explaining the purpose of the study and the questionnaire. The same applied to the Heads of Departments questionnaires.

cxcvii

Research policies from six universities were collected and analysed according to the preset criteria. The main criteria here was to find out whether the issue of career professional development, among academic staff, had been sufficiently taken into account and catered for in the university research policies.

3.8 Description of Data Analysis Procedures Descriptive and inferential statistics were used to analyse the data. Quantitative data from responses to closed ended type of questions in the questionnaire were coded in the computer by applying the Statistical Package for Social Sciences (SPSS) version 15. Several statistical analysis were performed, the quantitative data analysis was performed to generate frequencies and percentages. Cross tabulation analysis of various independent variables like age group, gender, type of university, rank and highest degree obtained with the participants number of publications were performed. Information obtained from the attitude scale was used to test hypothesis using one way ANOVA at 0.05 level of significance. If the p- value was less than or equal to 0.05 level of significance the hypothesis was rejected. If the p-value was greater than 0.05 level of significance the

cxcviii

hypothesis was not rejected. The one way ANOVA was used to test all the hypothesis presented in this study. Factor analysis was employed in data reduction. It reduced large number of factors influencing research productivity into a small number of factors that explained most of the variance observed in a much larger number of variables (Obure, 2002; child, 2006). The results were presented in frequency and percentage Tables, graphical representations and pie charts. Means and standard deviations were also computed to determine the respondents attitudes towards research productivity. Data collected from the open ended items were analysed by grouping similar answers together across respondents in order to form emerging themes/factors. The theme that had many respondents thus formed the main factor influencing research productivity. Other themes attracted very few responses and were therefore regarded as minor factors. These factors were sorted in a descending format for ease of interpretation and presented in a tabular form. Besides the demographic information and the last section of the of the Heads of Department questionnaires other parts of the questionnaires had many missing variables and was therefore excluded from analysis. It looked like most HoDs were not willing to commit themselves on the issue of publications made in the department in the last five years, they were not willing to give their views on the close ended questions on individual

cxcix

and institutional factors affecting research productivity, nevertheless they contributed meaningfully to the open ended questions presented to them. These items have been analysed in the last section of chapter 4 and they are indeed invaluable input in this study.

cc

CHAPTER FOUR PRESENTATION AND DISCUSSION OF THE FINDINGS 4.1 Introduction This chapter presents the study findings, interpretations and analysis including discussions based on the structure of the research questions. Deliberate effort has been made to address the pre specified study questions with relevance to factors influencing research productivity.

4.2

Demographic characteristics of participants The demographic information is derived from item one to ten of both the

Universities' academic staff and heads of department questionnaires. These items were name of university, faculty, department, gender, age group, rank, highest degree obtained, and work load.

4.2.1 Academic staffs demographic information Presentations, discussions and analysis in this section are founded on the perceived participants background variables. These, the researcher deemed to be important indicators

cci

and determinants of their attitudes in regard to the factors hindering or promoting research output among Kenyan university academic staff. This information was obtained from question one to ten of the university academic staff questionnaire (Appendix 1) The participants were asked to place a tick or a mark in an appropriate variable that best presented their status i.e. Name of university and faculty, age group, gender academic rank, years since last degree obtained. Item nine and ten required the participants to fill in hours time spent on teaching load and research work per week respectively. The questionnaire response rate obtained from academic staff was 69.3%. Male participants constituted 70.2% while 29.8% was made up of female academic staff. Sixty nine percent of the questionnaires were responded to by academic staff in public universities while 31.0% were filled by academic staff in private universities as shown in the Tables below. Table 4.1: Frequency Distribution of academic staff by Gender

Gender Male Female No Response

Frequency 193 82 2

Percent 70.0 29.6 0.7

ccii

Total

277

100.0

n=277 Table 4.1 above indicates the composition of the sample. It is spread in the ratio of 70:30 for men to women academic staff. Somehow it was a satisfactory ratio given that the number of academic staff and even students is in that range.

Table 4.2: Distribution of academic staff by type of university

Type of University Public university Private university Total

Frequency 191 86 277

Percent 69.0 31.0 100.0

n=277 Table 4.2 indicates that more respondents were drawn from the public university than from the private Universities. This is a true reflection of the University enrolment in Kenya today where there are seven Public universities with 11 constituent colleges and

cciii

several campuses all forming the public Universities. On the other hand, there are over twenty private universities with smaller enrolment. Table 4.3: Distribution of academic staff by Age group

Age Group 20-29 30-39 40-49 50-59 60-69 Total

Frequency 30 75 104 54 14 277

Valid Percent 10.8 27.1 37.5 19.5 5.1 100.0

n=277

Table 4.3 above, illustrates a higher number of academic staff i.e. 37.5% who belonged to the 40-49 age group while 5.1% of the respondents were in the 60-69 age group. This implies that most of the respondents were lecturers in the 40-49 age category. This is in agreement with the analysis done below which indicates that most of the respondents were those in the lecturer category. Table 4.4 indicates that most of the participants held lecturer positions; this is represented by 39.7% of the respondents. 17.4% were senior lecturers while 17.0% were assistant lecturers. Seven point six percent were graduate assistants and 7.2% were tutorial fellows. About six percent represented associate professors while only 3.3% were full

cciv

professors as shown in Table 4.4. The fact that many of the respondents were lecturers is satisfying in the sense that this is the group that is struggling to climb the academic ladder of the university to such positions as senior lecturers and professors. At least this was an ideal group to be interviewed and to see how they are fairing on in terms of research publications. Some studies reviewed had eliminated the use of the graduate assistants and tutorial fellows from their analysis. The argument they were positing was that the graduate assistants, tutorial fellows and assistant lecturers had not published much and therefore they could not be taken into account. But for the case of this study, the researcher found this group to have published though not much. Secondly, the nature of Kenyan Universities is really changing. Most of the Universities are resorting to part time academic staff. They recruit lecturers at those lower levels as part time lecturers with the intention of minimizing on costs. Others are recruited but take long to move up the ladder. Lastly, the current study is not very much into the output of the academic staff but their attitudes and views on factors that influence research productivity. This was the basis for retaining this group of academic staff in the study. The study also captured about 9% of the respondents in the rank of associate professors and above.

ccv

Table 4.4: Distribution of academic staff by Position Held in Their respective Universities.
Position held Graduate Assistant Tutorial Fellow Assistant lecturer Lecturer Senior Lecturer Associate Professor Professor No Response Total Frequency 21 20 47 110 48 17 9 1 277 Percent 7.6 7.2 17.0 39.9 17.4 6.2 3.3 0.4 100.0

n=277

Figure 4.1 indicates that a majority (53.4%) of the academic staff had Masters Degrees, 37.9% had PhDs while 6.9% had bachelors degrees. This was a good representation for the academic staff in Kenya. It is a situation that is gradually changing. It has been known over the years, for example at some of the oldest university public universities in Kenya, a majority of the PhDs are found in the so called life and basic sciences. The other areas like humanities and social sciences have had very few PhDs. This has partly been explained by the abundance of university scholarships targeting these

ccvi

fields. Secondly, there were no opportunities for PhD courses at the PhD level for these members of staff. This is however changing due to more universities offering PhD programmes across the country. Figure 4.1: Distribution of academic staff by Highest Degree Obtained

n=277

It was further observed that 45.8% of the academic staff participants had obtained their last highest degree in less than 5 years, while 25.6% obtained degrees between 6 10 years ago. 17.0% represented participants who obtained their highest degree between 11 15 years ago. 5.4% obtained their last highest degrees between 16 20 years ago while

ccvii

only 4.0% obtained theirs more than 20 years ago as shown in Table 4.5. This information is basically indicates that a sizeable number of the academic staff in Kenyan Universities have masters degrees.

Table 4.5: Distribution of academic staff by years since last highest degree was obtained

Years since last degree was obtained <5 Years 6-10 Years 11-15 Years 16-20 Years More than 20 Years No Response Total

Frequency 127 71 47 15 11 3 274

Valid Percent 45.8 25.6 17.0 5.4 4.0 1.1 100.0

n=277 4.2.2 Head of departments demographic information This section presents an outline of HODs demographic information as provided in the HoDs questionnaires. This information was obtained from question one to ten of the Heads of Department questionnaire (Appendix 2).

ccviii

The participants were asked to place a tick or a mark in an appropriate variable that best suited their status i.e. Name of university and faculty, age group, gender academic rank, years since last degree obtained. The response rate obtained from heads of department was 60.7%. 76.5% of the questionnaires were responded to by male participants while 23.5% were filled by female participants. 88.2% of the questionnaires were responded by heads of departments in public universities while 11.8% were filled by those in private universities as shown in Tables 4.6 and 4.7.

Table 4.6: Distribution of Heads of departments by gender

Gender Male Female Total Frequency 13 4 17 Percent 76.5 23.5 100.0

n=17 Table 4.7: Distribution of Heads of departments by type of university

Type of University Frequency Percent

Public Private Total

15 2 17

88.2 11.8 100.0

ccix

n=17 Many of the heads of departments i.e. 41.2% belonged to the 40-49 age group, just like the lecturers, while 37.3% belonged to the 50-59 age group and only 17.6% in the 3039 age group. Basically most of these HoDs are over 40 years of age as shown in Table 4.8 below. Table 4.8: Distribution of Heads of Departments by Age Group
Age group Frequency Percent

30-39 40-49 50-59 No response Total

3 7 6 1 17

17.6 41.2 37.3 5.8 100.0

n=17 Table 4.8 illustrates that most of the HoDs were above 40 years of age. Majority of the head of departments held lecturer positions; this was represented by 59%. About 23% were associate professors while 12% were senior lecturers. One was a professor as shown in Figure 4.2 below Figure 4.2: Distribution of heads of departments by Position held in university

ccx

n=17 Approximately 35.3% of the heads of departments had masters degrees, 58.8% had PhDs while only 5.9% had bachelors degrees as shown in the Table below.

Table 4.9: Distribution of heads of department by highest degree obtained

Highest degree obtained Frequency Valid Percent

Bachelors degree Masters degree PhD Total

1 6 10 17

5.9 35.3 58.8 100.0

n=17

ccxi

It was further observed that 52.9% of the head of department participants had obtained their last highest degree between 6 - 10 years ago, while 17.6% obtained less than 5 years ago and between 16 20 years ago. Only 11.8% obtained their last highest degrees between 11 15 years ago as shown in Table 4.10. Table 4.10: Distribution of heads of departments by years since last highest degree was obtained
Frequency Percent

<5 Years 6-10 Years 11-15 Years 16-20 Years Total

3 9 2 3 17

17.6 52.9 11.8 17.6 100.0

n=17 Table 4.10 shows that most of the HoDs finished school in more than 9 years ago. This means that they are persons who have been in the university system for long. The responses that they provided to the rest of the items in the questionnaire reflected this maturity. 4.3 Nature of Research Productivity among public and private universities in Kenya between 2004-2008

ccxii

This section analysed data that helped to answer research question number one on the nature of research productivity among public and private universities in Kenya. 4.3.1 Average publication output by university between 2004 - 2008 The information presented below is a calculation of the average publication output of the academic staff in Kenya. The calculation was done by dividing the total publications from self reported data in this study by the number of academic staff in the sample. This was done for all the universities under study in each of the five years under study. Table 4.11: Average publication output By University 2004-2008
UNIVERS ITY A B C D E 2004 Mean % 0.4 1 0.5 7 0.3 8 0.2 3
No dat a No dat a

2005 Mean % 0.4 5 0.7 4 0.4 7 0.2 3

No dat a No dat a

2006 Mean % 0.5 5 0.8 9 0.3 8 0.3 1 0.0 8 12. 0 19. 4 8.3 6.8 1.7

2007 Mean % 0.6 11.2 0 0.8 9 0.5 2 0.2 3 0.1 7 16.6 9.7 4.3 3.2

2008 Mean % 0.7 12.6 1 0.9 0 0.1 9 0.1 5 0.3 3 16.0 3.4 2.7 5.9

Total Mean % 2.72 12.1 3.99 17.7 1.94 8.6 1.15 5.1 0.58 2.6

13. 1 18. 3 12. 2 7.4

11. 7 19. 3 12. 2 6.0

ccxiii

No dat

No dat

0.1 6 0.2 8 0.3 5 0.6 7 0.3 3 0.1 6

4.2

0.2 5

5.5

0.3 8

7.0

0.3 8

6.7

1.17 5.2

G H J K L

0.1 9 0.2 1 0.6 7 0.2 9 0.1 7

6.1 6.7 21. 5 9.3 5.4

7.3 9.1 17. 4 8.6 14. 2 10 0

0.2 8 0.3 6 0.7 5 0.2 4 0.5 0 4.5 9

6.1 7.8 16. 3 5.2 10. 9 10 0

0.3 1 0.4 6 0.8 8 0.4 3 0.5 0 5.3 7

5.8 8.6 16.4 8.0 9.3

0.5 6 0.4 6 0.8 8 0.5 7 0.5 0

10.0 8.2 15.6 10.1 8.9

1.62 7.2 1.84 8.1 3.85 17.1 1.86 8.2 1.83 8.1

TOTAL

3.1 2

10 0

3.8 4

100 .1

5.6 3

100 .1

22.55 100

National Average = 2.05 Table 4.11 summarizes average publication outputs of academic staff in the eleven Universities under study between 2004 2008. This can also be referred to as the research productivity of the said institutions. Here, the total research publications were divided by the total number of academic staff in that institution to yield this figure. The highest production, according to this table, was done in 2008 by university B which produced 0.90

ccxiv

publications per academic staff per year. This means that each academic staff produced 0.90 publications in that year. The publication outputs fluctuate between these five years and vary considerably across different universities and also across different years for the same university. From the information presented in this Table, the Universities can be divided into two groups, that is; high and low performers. Those above the ten percent average mark can be considered as a high group (B, A, J) while those with a performance less than ten percent can be grouped as low performing group (C, D, E, F, G, H, K, L). It is evident that all the high group universities are public universities whereas the low group has accommodated all the private universities in the sample with a few public universities. This may be partially due to the higher number of senior lecturer and above in the older universities unlike the private universities. Funding levels might be different too in these institutions. It is interesting to see here that B and J had a higher average rate of publications. One thing that is quite clear of University B is the large size of the institution and also being the first university to be set up here in Kenya as a branch of the University of East Africa. This university enjoyed this privilege for too long till 1984 when the second public university was established after the recommendations of the Mackay report of 1982. These findings are in disagreement with the findings by Ho (1998) who found that the average

ccxv

production rate for the universities he studied in Hong Kong was above 7.0 mark for the five year study he undertook the study. On the other hand, Milgrom et al (2008) conducted a study with the International Association for Dental Research (IADR) Behavioral Sciences and Health services Research Group whose primary outcome measure was the number of self-reported published articles in Pub Med in twenty-four months. The mean number of the publications he found was 4.9 (SD=5.1). Studies done in Beirut by Dakik (2005) analysed the quality and quantity of scientific publications of the medical faculty at the American University of Beirut (AUB) during a six year period (19962001). Their average productivity rate was found to be 1.24 publications/faculty member/year (PFY). Eighteen per cent of the faculty did not have any publication in the six year study period, and only 20% had two or more publications per year. For the case of the present study, no single university managed to produce a single digit in terms of research productivity of its individual members per year. The highest was 0.90 by university B in the year 2008. The conditions under which academic staff was working in Ho and Milgroms studies are different from the situation in Kenya and these differences in research productivity might have arisen due to that.

ccxvi

However, these findings resonates the findings of Lee and Bozeman (2004) who used the number of papers published in five years to calculate the research productivity of various countries around the world. For their sample of U.S. academicians, they found an average ranging from 14.40 papers for assistant professors to 25.75 papers for full professors. Their study used self-reported data too, the self-reported sum of articles in national and foreign journals was 4.5 for both academic and government researchers. both the African and Indian respondents in their study published at a significantly lower rate than Bozemans sample of U.S. scientists. Within developing areas; however, there were significant differences. The mean number of total articles (foreign and national journals) ranged from 7 articles in Kerala, to 3.6 in Ghana and 2.5 in Kenya.

ccxvii

Table 4.12: Frequency distribution of those who Published by rank

Graduate 2004 Assistant 0 0% 2 2005 1.6% 2006 3 2.1% 2007 3 1.9% 2008 7 3.8%

Tutorial Fellow 1 1.0% 4 3.1% 7 4.8% 9 5.6% 11 6.0%

Assistant lecturer 4 3.9% 9 7.0% 14 9.7% 25 15.4% 29 15.9%

Lecturer 41 39.8% 50 38.8% 58 40.0% 64 39.5% 70 38.5%

Senior Lecturer 32 31.1% 38 29.5% 38 26.2% 37 22.8% 40 22.0%

Associate Professor 16 15.5% 16 12.4% 15 10.3% 15 9.3% 16 8.8%

Professor 8 7.8% 9 7.0% 9 6.2% 9 5.6% 9 4.9%

Total 103 100.0% 129 100.0% 145 100.0% 162 100.0% 182 100.0%

The scenario presented in Table 4.12 shows that the position of lecturer and senior lecturer were the most productive of all the academic staff under study. The contribution by graduate assistants, tutorial fellows and professors was not as high. This is an expected scenario since most lecturers are aspiring to grow professionally. They are being motivated to produce more so as to climb the academic ladder. They have the energy because they fall under the most productive bracket of 40-49 years old. They have also gained enough

ccxviii

experience to propel them to writing more publications. This is a confirmation that the experience gained by the academic staff is crucial or instrumental in increasing ones productivity in terms of publishing. Table 4.13 Frequency distribution of those who did not publish by rank
Graduate 2004 Assistant 19 12.7% 19 13.1% 18 14.0% 19 16.9% 14 15.4% Tutorial Fellow 43 28.7% 16 11.0% 13 10.1% 11 9.8% 9 9.8% Assistant lecturer 70 46.7% 38 26.2% 33 25.6% 22 19.6% 18 19.7% Lecturer 16 10.7% 61 42.1% 53 41.1% 47 42.0% 41 45.1% Senior Lecturer 1 0.7% 10 6.9% 10 7.6% 11 9.8% 8 8.9% Associate Professor 1 0.7% 1 0.7% 2 1.6% 2 1.8% 1 1.1% Professor 0 0% 0 0% 0 0% 0 0% 0 0% Total 150 100.0% 145 100.0% 129 100.0% 112 100.0% 91 100.0%

2005 2006

2007

2008

Table 4.13 is a confirmation of the analysis in Table 4.12. Here, it is quite evident that a smaller number of lecturer and above is not engaged in publishing. A majority of the non publishers fall in the category of lecturer and below. There is an interesting scenario for the categories of assistant lecturer and lecturer between the years 2004-2008 The assistant lecturers started on the wrong footing with about 70 of them not publishing

ccxix

anything in 2004 but this figure continued to rise. This can be attributed to experience and the realization of the importance of publishing and publications. Unfortunately, the reverse is the case for the position of lecturer. This can be explained by the motivation by the lecturers to move to higher rank hence an increase in publication output. It is interesting also to note that there was no single professor who had not published in the five year period. They had made publications in their fields of specialization and were therefore were recognized as producers. There are many reasons that can be associated with this. The first one is the experience of the professors to attract funding, ability to allocate funds for research to their teams, ability to use graduate students to develop collaborative papers and of course the trust the donor community has on people with experience to conduct their research and evaluations. Most donors or projects usually require that the Principal Investigators be people who have handled work of same magnitude before. The findings of this study are in agreement with Kelchtermans et al (2005) who observed that most of the studies to date aim at explaining average productivity profiles, ignoring the often skewed distribution of research productivity, with many researchers nonactive and a few researchers accounting for the bulk of the publications. This issue of persistence of research productivity profiles is underexplored in the literature. Indeed the

ccxx

above Table 4.13 indicates that it is only the professors who have managed to publish consistently over the five year period. They have been able to realize the benefits of publishing whereas the majority of the academic staff have not had that privilege.

Table 4.14: Average publication output by gender 2004-2008 2004 Male Female TOTAL Mean 0.39 0.33 0.72 % 54.2 45.8 100 2005 Mean 0.46 0.49 0.95 % 48.4 51.6 100 2006 Mean 0.51 0.54 1.05 % 48.6 51.4 100 2007 2008 Total

Mean % Mean % Mean % 0.60 52.6 0.69 53.9 2.65 51.6 0.54 47.4 0.59 43.1 2.49 48.4 1.14 100 1.28 100 5.14 100 National Average = 2.57

Table 4.14 indicates that the publication output among gender fluctuates between these five years. It is evident that there is some difference between the publication levels when categorized by gender. The difference is however small. Though this table (Table 4.14) shows that women published fewer papers than their male counterparts between 2004 2008, the difference is not statistically significant (0.722) at the 95 percent level as can be seen in Table 4.21.

ccxxi

This is in agreement with Blackburn et al. (1978), who observed that gender does not appear to influence research output after other influences such as rank and discipline are controlled. The national average at 2.57 is however quite low and needs to be boosted across the board. Studies on gender and research productivity have been studied with mixed results. Bailey (1992) reported a higher level of research productivity by male faculty members. Other researchers have noted that female faculty members are lagging behind experienced male faculty members in research productivity (Smith, Anderson, Lovrich, and Nicholas, 1995). Blackburn et al. (1991) stated that the relationship between gender and research productivity had been addressed in many studies and that little if any, and sometimes contradictory, correlations have been found. Therefore for this case in Kenya, the information realized is satisfactory. Gender differences in scientific productivity have been given attention by many researchers in the recent past. Several studies have found that female scientists publish at lower rates than male scientists. Using a sample of American biochemists, Long (1993) finds that sex differences in the number of publications and citations are bigger during the first decade of the career but are reversed later. He attributes the lower productivity of females to their overrepresentation among non-publishers and their under representation

ccxxii

among the extremely productive. Definitely this has not been supported by the present study.

Table 4.15: Average publication output By Rank 2004-2008 2004 Graduate Mean % 0.05 2.2 2005 Mean % 0.10 2.7 2006 Mean % 0.14 3.5 2007 Mean % 0.14 3.2 2008 Mean % 0.33 6.7 Total Mean % 0.76 4.0

ccxxiii

Assistants Tutorial Fellows Assistant Lecturers Lecturers Senior Lecturers Associate Professors Professors TOTAL

0.2

8.6

0.2

5.4 5.2

0.35 0.30

8.8 7.5

0.45 10.3 0.55 11.2 0.53 12.2 0.62 12.6

1.75 9.1 2.51 13.0 2.49 12.9 3.51 18.2 4.11 21.4 4.11 21.4 19.24 100

0.87 37.5 0.19

0.29 12.5 0.45 12.3 0.53 13.3 0.58 13.3 0.64 13.1 0.33 14.2 0.79 21.5 0.79 19.8 0.77 17.7 0.83 16.9 0.47 20.3 0.94 25.6 0.88 22.1 0.88 20.2 0.94 19.1 0.11 2.32 4.7 100 1.0 27.2 1.0 25.9 1.0 33.0 1.0 20.4 3.67 100 3.99 100 4.35 100 4.91 100

National Average = 2.75 Table 4.15 presents data on the average publication output among various ranks held by the academic staff in Kenyan Universities between 2004-2008. This productivity index was calculated by dividing the total published data per year by the total number of academic staff in the sample. From this information, it is clear that the publication output increases with the rank held by the academic staff. For example the graduate assistants had 0.76 as their total mean for the 5 year period while the professors had 4.11 as their total mean for the same period. This may be due to the exposure of older academic staff to research funding opportunities, experience and professional networks in research. There is

ccxxiv

an obvious decline of publication output from the higher cadres of professorship to the lower cadres. Bailey (1992) found that rank is a significant predictor of research productivity. Dundar and Lewis (1998) found that departments with higher ranked faculty had higher research productivity. Vasil (1992) reported that rank was a significant predictor of research productivity. This gives the same explanation for the current study. Some argue that the professors have low teaching load and can therefore manage to have plenty of time to publish. Others argue that they are mainly in the administrative positions and that they can manage to maneuver research funds to their research teams. According to Finkelstein (1984), academic rank is a significant predictor to publication success because the academic lecturers in higher ranks generally have more control over their workload assignment, allowing faculty of higher rank to produce more research than those of a lower rank. Fulton and Trow (1974) found that 29 percent of full professors, 20 percent of associate professors, 13 percent of assistant professors, and 2 percent the instructors have published five or more articles in a two-year period.

ccxxv

Table 4.16: Average publication output by Age group 2004-2008 2004 20 29 30 39 40 49 50 59 60 69 TOTAL Mean 0.07 0.16 0.41 0.63 0.86 2.13 % 3.3 7.5 19.2 29.6 40.4 100 2005 Mean 0.1 0.27 0.53 0.72 0.86 2.48 2006 2007 Mean 0.2 0.51 0.62 0.80 0.79 2.92 2008 Total Mean 0.87 2.02 2.79 3.68 4.01 13.7 % 6.7 14.7 20.4 26.9 29.3 100

% Mean % 4.0 0.17 6.5 10.9 0.4 15.3 21.4 0.58 22.1 29.0 0.76 29.0 34.7 0.71 27.1 100 2.62 100

% Mean % 6.8 0.33 10.2 17.5 0.68 21.1 21.2 0.65 20.2 27.4 0.77 23.9 27.1 0.79 24.5 100 3.22 100

National Average = 2.74

Table 4.16 above illustrates that older age groups generate more publications than younger age groups. From this Table it is clear those persons who are 50 years and above (about 25% of the academics) account for more than a half of the average publications for the entire five year period. Again this shows that older academics in Kenya are producing more than younger academics. The life-cycle model (Diamond, 1986) predicts that faculty

ccxxvi

research productivity will decline as an individuals academic experience increases. Unfortunately this study was not able to sample those lecturers above 70 year maybe this could have given us an insight into this life cycle model. Age has been included in several studies with conflicting results. Bland and Berquist (1997) observed that the average productivity of faculty seemed to drop with age, however, many senior faculty members remained quite active in research activities and their products were comparable to those of younger faculty members. They also reported that there was no significant evidence that age determined a drop in productivity, but increased workloads and shifting emphasis was to blame. Gorman and Scruggs (1984) reported that age was related to research productivity. Blackburn et al. (1991) stated that the relationship between age and research productivity had been addressed in many studies and that little if any, and sometimes contradictory, correlations were found.

Table 4.17: Average publication output By University Type 2004-2008 2004 Public Private TOTAL Mean 0.47 0.15 0.62 % 75.8 24.2 100 2005 Mean 0.56 0.26 0.82 % 68.3 31.7 100 2006 Mean 0.63 0.29 0.92 % 68.5 31.5 100 2007 Mean % 0.67 65.0 0.36 35.0 1.03 100 2008 Mean % 0.74 60.7 0.48 39.3 1.22 100 Total Mean % 3.07 66.6 1.54 33.4 4.61 100

ccxxvii

National Average = 2.31 Table 4.17 is an indication that the private universities are lagging behind in terms of publications output than their counterparts, the public universities. An encouraging feature here is the steady progress of the increase in the productivity index between 20042008. Though this table (Table 4.17) shows that private universities published fewer papers than their counterparts, the public universities, between 2004 2008, this difference is not statistically significant (0.55) at the 95 percent level as can be seen in Table 4.19. However, there are many reasons that can account for the fewer number of papers published by the private universities in Kenya in the said period. This may be partly due to the fact that most of them were awarded charters more recently. These universities also have comparatively more newly recruited academic staff than the public universities. This can also be attributed to the higher number of graduate students in older universities than in the newer universities. The public universities have also benefitted a lot from central governments for a long time. This dependence on government enabled them to invest in infrastructure that is serving them well to date. Again, being the oldest universities in the country, all donor funding projects were targeted at them. They have continued to enjoy this good will and

ccxxviii

they have built on it. The facilities they built have been converted for use during this season of module II. In the process they prove to have a niche over the others who did not have this opportunity. Ho (1998) conducted a study on measurement of research output among the three faculties of business, education, humanities and social sciences in six Hong Kong universities. In order to have a fair comparison of publication outputs of each academic, rank, faculty and university, a framework was developed from practical experience and from literature to investigate the problem. Results indicated that the publication outputs of academics in Hong Kong were about the same as other countries in many aspects. Here, average publication output ranged from 7.0, 7.3 and 8.9 for faculties of business, social sciences and education respectively. The recommendation from this study, therefore, was that Hong Kong academics should not be motivated to do more research. According to Ho (1998), more motivation could just increase initial output but decline in the long run. On the other hand, Milgrom et al (2008) found out the research productivity of members of international dental association to be 4.9. These two scenarios presented indicates that the universities in Kenya have to work extra hard to redeem themselves from this underperformance. Table 4.18: Average publication output by highest degree obtained 2004-2008 2004 2005 2006 2007 2008 Total

ccxxix

Mean % Masters PhD TOTAL 0.15 0.18 0.33 45.5 54.5 100

Mean % 0.26 0.84 1.1 23.6 76.4 10

Mean % 0.36 0.85 1.21 29.8 70.2 100

Mean 0.48 0.85 1.33

% 36.1 63.1 100

Mean % 0.56 0.88 1.44 38.9 61.1 100

Mean % 1.81 33.5 3.6 66.5 5.41 100

National Average = 2.71 Table 4.18 presents data on the publications output by the highest degree obtained. It is clear that the PhD holders are producing more research publications than the Masters Degree holders. This is a confirmation on the publication output by age and rank held in the university. This is a challenge for universities in the country to increase the number of PhDs in their universities if they are to realize higher publication outputs.

4.3.2 Test of Hypotheses The purpose of this section is to test the results discussed above. The One way ANOVA has been employed in this exercise to test the results presented above. The Ho was tested at 0.05 level of significance.

ccxxx

Decision rule: When the P Value is less than or equal to .05, the null hypothesis is rejected meaning there is significant difference between the variables under study. If, however, the P Value is greater than .05 level of significance, the null hypothesis is not rejected meaning that there is no significant difference between the variables being tested. Ho1 There is no significant difference between the means for academic staffs type of university and research productivity.

Ha1 There is significant difference between the means for academic staffs type of university and research productivity.

Table 4.19: One way ANOVA test of difference in the mean score of academic staffs type of university and research productivity.
ANOVA SUMMARY TABLE

Source of variation Type of university

SS Between 9.34 Groups Within Groups Total 30.9 40.2

DF 36 175 211

MS .259 .177

F 1.47

Sig. .055

In the One way ANOVA Table above, the P-value for type of university is 0.055. This is

ccxxxi

greater than 0.05 level of significance. Decision: The null hypothesis is not rejected. Conclusion: Thus there is no significant difference between the means for academic staffs type of university and research productivity.

There were two types of universities in this study, the private universities and the public universities. The results obtained here indicate that there is no significant difference in terms of research productivity by the academic staff from whichever type of university. This is very positive to the private universities in Kenya in the sense that most of them are new and have entered the arena to compete with the public universities. The public universities have been around for long and have many advantages associated with that. These advantages range from the large number of qualified senior staff, long term association with the state for funding and other resources, and well established infrastructure. Ho2 There is no significant difference between the means for academic staffs age groups and research productivity.

Ha2 There is significant difference between the means for academic staffs age groups and

ccxxxii

research productivity.

Table 4.20: One way ANOVA test of difference in the mean score of academic staffs age group and research productivity.
ANOVA SUMMARY TABLE

Source of variation Age group

SS Between 68.1 Groups Within Groups Total 141.4 209.5

DF 36 175 211

MS 1.89 .808

F 2.34

Sig. .000

In the One way ANOVA Table above, the P-value for type of university is 0.000. This is lower than 0.05 level of significance. Decision: The null hypothesis is rejected. Conclusion: Thus there is significant difference between the means for academic staffs age groups and research productivity. This is indeed confirmed from the available data. The older age groups are producing more than the younger age groups. This is indeed true for most of the organization. One key factor that comes to play here is the issue of experience. The more the years spent in performing a particular task, the skills of doing the same work are

ccxxxiii

sharpened and improved. For this case, the university senior lecturers, associate professors and above are no doubt producing more in terms of research publications than the younger age groups. Ho3 There is no significant difference between the means for academic staffs gender and research productivity. Ha3 There is significant difference between the means for academic staffs gender and research productivity.

Table 4.21: One way ANOVA test of difference in the mean score of academic staffs gender and research productivity.

ANOVA SUMMARY TABLE

Source of variation Sex

SS Between 6.29 Groups Within Groups Total 36.3 42.6

DF 36 175 211

MS .175 .207

F .843

Sig. .722

In the One way ANOVA Table above, the P-value for type of university is 0.722. This is greater than 0.05 level of significance.

ccxxxiv

Decision: The null hypothesis is not rejected. Conclusion: Thus there is no significant difference between the means for academic staffs gender and research productivity. Basically this result indicates that the research output of both males and females in the university setup in Kenya is the same. This is a small confirmation that given the chance and opportunity, the female academic staff in Kenya is able to compete favorably with their male counterparts, if not better. This can be a reason to promote them to administrative positions in the universities to fulfill the 30% requirement for women in leadership positions in the universities as advised by Kenya Government (GOK, 2005).

Ho4 There is no significant difference between the means for academic staffs academic rank and research productivity. Ha4 There is significant difference between the means for academic staffs academic rank and research productivity.

Table 4.22: One way ANOVA test of difference in the mean score of academic staffs rank and research productivity.

ANOVA SUMMARY TABLE

ccxxxv

Source of variation Rank

SS Between 138.1 Groups Within Groups Total 235.6 373.7

DF 36 175 211

MS 3.84 1.35

F 2.85

Sig. .000

In the One way ANOVA Table 4.22, the P-value for type of university is 0.000. This is lower than 0.05 level of significance. Decision: The null hypothesis is rejected. Conclusion: There is significant difference between the means for academic staffs academic rank and research productivity. This is a confirmation of the earlier analysis of the academic staff in terms of age groups. This result indicates that the higher the rank, the higher the research productivity in the University set up in Kenya. This is quite true in that the senior lecturers and the professors fall in this category. For all obvious reasons, the senior members of the academic staff will tend to produce more than the junior members of the academic staff. This does not apply to Kenyan universities only; it is the same in most universities in the world. (Ho, 1998) confirmed this when he studied the research output of academic staff in Hong Kong, he found out the senior members of the academic staff were writing more than the junior members of staff. The senior members of staff may also be involved in research

ccxxxvi

in their universities due to their ability to draw funding for research. Most funding organizations prefer to deal with persons who have wide experience in research. Again, these are the people in university leadership positions, they have added advantages in securing internal resources meant for research. In most universities they utilize the graduate students in writing joint books, journals, and even conference papers. Many studies illustrate a correlation between academic rank and a scientists productivity. In a study sample of American academics, Blackburn and Al [1978] show that full professors publish at a higher average rate than associate professors and research staff. Dickson [1983] and Kyvik [1990] have illustrated the same effect of professional role on scientific productivity in their respective studies of Canadian and Norwegian universities. Academic rank was studied by Bailey (1992), Dundar and Lewis (1998), Gottlieb et al. (1994), Teodorescu (2000) and Vasil (1992). Each found rank to be a significant predictor of research productivity. Several studies have found seniority of academic rank to be correlated with research performance.

Ho5 There is no significant difference between the means for academic staffs highest degree obtained and research productivity. Ha5 There is significant difference between the means for academic staffs highest degree

ccxxxvii

obtained and research productivity.

Table 4.23: One way ANOVA test of difference in the mean score of academic staffs highest degree obtained and research productivity.

ANOVA SUMMARY TABLE

Source of variation Highest degree

SS Between 22.8 Groups Within Groups Total 45.9 68.8

DF 36 175 211

MS .634 .263

F 2.414

Sig. .000

In the One way ANOVA Table above, the P-value for type of university is 0.000. This is lower than 0.05 level of significance. Decision: The null hypothesis is rejected. Conclusion: There is significant difference between the means for academic staffs highest degree obtained and research productivity.

It goes without saying that the higher the academic qualifications of one, the higher the research output. In this case, the highest academic qualifications are the doctor of

ccxxxviii

philosophy. From the data available in this study, this groups output is comparably higher than that of the masters degree holders. The PhD holders are no doubt the leaders in publishing in Universities due to their long experience in the field of research.

Ho6 There is no significant difference between the means for academic staffs years since last highest degree was obtained and research productivity. Ha6 There is significant difference between the means for academic staffs years since last highest degree was obtained and research productivity.

Table 4.24: One way ANOVA test of difference in the mean score of academic staff s years since last highest degree was obtained and research productivity.

ANOVA SUMMARY TABLE

Source of variation Years since last highest degree obtained

SS Between 67.1 Groups

DF 36

MS 1.86

F 1.40

Sig. .079

In the One way ANOVA Table above, the P-value for type of university is 0.079. This is

ccxxxix

greater than 0.05 level of significance. Decision: The null hypothesis is not rejected. Conclusion: Thus there is no significant difference between the means for academic staffs years since last highest degree was obtained and research productivity.

Here, the period since last highest degree was obtained does not seem to matter much in terms of research output from the universities in Kenya.

Ho7 There is no significant difference between the means for academic staffs university and research productivity. Ha7 There is significant difference between the means for academic staffs university and research productivity.

Table 4.25: One way ANOVA test of difference in the mean score of academic staffs university and research productivity

ANOVA SUMMARY TABLE

ccxl

Source of variation Lecturers university

SS Between 560.1 Groups Within Groups Total 2314.8 2874.9

DF 36 175 211

MS 15.6 13.2

F 1.18

Sig. .244

In the One way ANOVA Table above, the P-value for type of university is 0.244. This is greater than 0.05 level of significance. Decision: The null hypothesis is not rejected. Conclusion: Thus there is significant difference between the means for academic staffs university and research productivity. In this study it was found out that some academic staff from particular individual universities was doing better in terms of research productivity than others. There are many reasons associated to this. Some of the universities are quite new and have just begun. Some are old public universities that have enjoyed the resources from the state for far too long. These differences in the university setups will definitely reflect in the manner o research output that they produce. There might be efforts placed by individual universities to promote research and many more reasons can account for these differentials in terms of research output.

ccxli

Gibbs and Locke (1989) insisted that research productivity was the most important criterion for making promotion and tenure decisions after surveying 59 chairs and committees in 93 universities. This increase in emphasis on research and decrease in importance of teaching and service has been recognized by faculty members since the 1980s (Cargile & Bublitz 1986; Schultz, Mead & Hamana, 1989). It is therefore clear why staff in traditional universities, where research has always featured more significantly in promotion and development of status, is expected to maintain and possibly increase research output. It should be clear, though, that a balance needs to be struck between these noble duties of the university, teaching, research and service

4.4 Lecturers views on individual and Institutional factors influencing research productivity

In this section, the views of universities academic staff have been explored in line with the research question on factors that promote or hinder research output among Kenya university academic staff. 4.4.1 Lecturers views on individual factors influencing research productivity

ccxlii

Table 4.26: Individual Factors influencing research productivity Large Extent f 1 By self motivation 2 Socialization with colleagues 3 Research content knowledge 4 Research skills gained 5 Job satisfaction 6 Simultaneous research projects 7 Parenting responsibilities 8 Early orientation to research work 9 Personal work discipline 175 49 124 136 97 42 48 106 154 % 63.2 17.7 44.8 49.1 35.0 15.2 17.3 38.3 55.6 f Some Extent % 17.7 45.1 29.6 27.8 27.8 36.1 27.8 31.0 23.5 f Little Extent % 10.1 19.9 13.4 12.3 20.9 28.2 26.0 18.8 11.6 No Extent f 22 45 28 24 39 45 75 26 23 % 7.9 16.2 10.1 8.7 14.1 16.2 27.1 9.4 8.3

49 125 82 77 77 100 77 86 65

28 55 37 34 58 78 72 52 32

From the Table 4.26 above the frequency distribution for each participant per item was calculated in an SPSS for Windows version 15. The information obtained can be

ccxliii

summarized as follows; self motivation (63.2%) Personal work discipline (55.6%) Research skills gained (49.1%) Research content knowledge (44.8%) and early orientation to research work (38.3%) received a rating of large extent. This means that these are the individual factors that influence research productivity to a large extent. This is in agreement with the findings of the Principal Component Analysis that shall be discussed later in this study. These findings indicate that even within the individual factor influencing research productivity, there are key factors that are making a big contribution than the rest. In this case, self motivation is seen as a key mover in the individual factors category. This is closely followed by personal work discipline. This basically means that the decision to publish or not to publish lies with individual academic staff. If he is motivated and disciplined and focused in that direction, then he will achieve his objective of publishing. 4.4.2 Lecturers views on Institutional factors influencing research productivity Table 4.27: Institutional Factors Influencing Research Productivity
Large Extent 1. 2. 3. 4. 5. 6. Resources available for research Rewards for research output Good salary Sufficient work time Clear research coordination goals Mentorship among colleagues f 183 117 80 126 92 78 % 66.1 42.2 28.9 45.5 33.2 28.2 Some Extent f 49 83 78 87 108 114 % 17.7 30.0 28.2 31.4 39.0 41.2 Little Extent f 23 51 71 34 44 51 % 8.3 18.4 25.6 12.3 15.9 18.4 No Extent f 21 24 43 26 24 26 % 7.6 8.7 15.5 9.4 8.7 9.4

ccxliv

7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

Communication with professional networks Library facilities Size of the university Recruitment and selection of academic staff Positive group climate Research emphasis by university Access to relevant journals teaching load Availability of technology e.g. internet and computers Equipment for research Number of graduates students supervised

93 132 34 38 69 112 125 147 149 158 68

33.6 47.7 12.3 13.7 24.9 40.4 45.1 53.1 53.8 57.0 24.5

119 74 74 79 109 78 88 65 71 65 86

43.0 26.7 26.7 28.5 39.4 28.2 31.8 23.5 25.6 23.5 31.0

42 41 74 79 58 52 36 20 26 24 60

15.2 14.8 26.7 28.5 20.9 18.8 13.0 7.2 9.4 8.7 21.7

20 25 88 76 37 28 26 32 22 22 51

7.2 9.0 31.8 27.4 13.4 10.1 9.4 11.6 7.9 7.9 18.4

Table 4.27 clearly indicates that particular institutional factors have a direct influence on research productivity of academic staff in Kenya. These factors are assumed to be provided by the respective institutions where the academic staff work. Topping the list is resources for research (66.1%), equipment for research (57%), availability of technology equipment (53.8%), large teaching load (53.1%) and rewards for research (42.2%). Contrary to general opinion, good salary, size of the university, recruitment and selection of academic staff have not been rated highly. These findings are in agreement with the Principal Component Analysis to be discussed later in this study. Several studies have reported the relationship between research productivity and salary (Jacobsen, 1992; Pfeffer & Langton, 1993; Tornquist and Kallsen, 1992). Since salary often reflects research

ccxlv

productivity levels, this was expected. Paying attractive salaries in return for performance may serve as an incentive for higher productivity from faculty members. Higher salaries may also attract productive faculty while at the same time minimizing the possibility of losing active faculty to other institutions (Pfeffer & Langton, 1993). Closer home, the University of Transkei in South Africa tried this and they succeeded. The number of publications shot up since early 1990s and the tempo has remained (Mwamwenda, 1994).

4.4.3 Heads of Departments views on Individual factors influencing research productivity

Table 4.28: Individual factors Influencing research productivity as perceived by Heads of Departments

ccxlvi

Large Extent

Some Extent

Little Extent

No Extent

f 1 Self motivation 2 Socialization with colleagues 3 Research content knowledge 4 5 6 7 8 9 Research skills gained Job satisfaction Simultaneous research projects Parenting responsibilities Early orientation to research work Personal work discipline 9 4 8 10 6 4 2 6 6

% f 52.9 5 23.5 47.1 58.8 35.3 23.5 11.8 35.3 35.3 3 4 2 6 5 4 3 6

% 29.4 17.6 23.5 11.8 35.3 29.4 23.5 17.6 35.3

f
No data

%
No data

f 2 3 3 1 2 3 2 3 1

% 11.8 17.6 17.6 5.9 11.8 17.6 11.8 17.6 5.9

6
No data

35.3
No data

3 2 4 6 3 3

17.6 11.8 23.5 35.3 17.6 17.6

Most heads of departments are in agreement that research skills gained (58.8%), self motivation (52.9%) and research content knowledge (47.1%) play a critical role in influencing lecturers research productivity. On the other hand socialization with colleagues (35.3%) is seen to be of no effect to the lecturers research productivity. These views by the heads of departments indicate that most academic staff are not well grounded in research skills.

4.4.4 Heads of Departments views on Institutional factors influencing research productivity

ccxlvii

Table 4.29: Heads of Departments attitudes on institutional factors affecting academic staff research productivity

ccxlviii

ccxlix

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

Resources available for research Rewards for research output Good salary Sufficient work time Clear research coordination goals Mentorship among colleagues Communication with professional networks Library facilities Size of the university Recruitment and selection of academic staff Positive group climate Extent to which research productivity is affected by research emphasis by university Access to relevant journals Teaching load Availability of technology e.g. Internet and computers Extent to which research productivity is affected by equipment for research Number of graduates students supervised

Large Extent f % 12 70.6 11 5 12 8 4 4 10 2 5 1 9 1 11 6 10 6 64.7 29.4 70.6 47.1 23.5 23.5 58.8 11.8 29.4 5.9 52.9 5.9 64.7 35.3 58.8 35.3

Some Extent f % 3 17.6 3 4 1 6 4 5 1 17.6 23.5 5.9 35.3 23.5 29.4 5.9

f 1 3 4 2 1 9 7 2 7

Little Extent % 5.9 17.6 23.5 11.8 5.9 52.9 41.2 17.6 41.2 17.6 23.5 17.6 29.4 5.9 509 17.6 5.9

f 1

No Extent % 5.9

3 2 2

17.6 11.8 11.8

1 3 8 2 2 3 2

5.9 17.6 47.1 11.8 11.8 17.6 11.8

6 10 2 9 5 7 2 5

35.3 58.8 11.8 52.9 29.4 41.2 11.8 29.4

3 4 3 5 1 1 3 1

3 1 4

17.6 5.9 23.5

ccl

The heads of departments postulate that resources available for research (70.6%), sufficient work time (70.6%), rewards for research output (64.7%), teaching load (64.7%), library facilities (58.8%) and equipment for research (58.8%) play a critical role as institutional factors influencing research productivity. This is very clear that without the requisite resources, no meaningful research output will be realized in any institution. The point that these HODs are making is that lack of financial support will lead to poor performance in terms of research output and that the more the resources for research the better for an institution to carry out research mandate. This researchers long association with the university system in Kenya, both public and private university setups has experienced situations where many theses, of complete researches have been lying in shelves without being published. This is testimony that in most cases, financial resources are a small component for publishing to take place. However, it can also be argued that financial resources are needed to run a journal or to print books for that matter. However, it is interesting to note that the heads of departments concur with lecturers that a good salary will not influence research productivity.

4.4.5 Principal Component Analysis

ccli

This part of the study also used Principal Component Analysis (PCA) to make an analysis of the factors that influence research productivity. PCA is one type of factor analysis. Factor analysis attempts to identify underlying variables or factors that explain the pattern of correlations within a set of observed variables. It is mainly used in data reduction to identify a small number of factors that explain most of the variance observed in a much larger number of manifest variables. (Obure, 2002; child, 2006) in this way the most important variable accounting for the variability in a set of data can be identified. This method has been used by various studies focusing on research productivity in the recent past. Babu (1998) designed an investigation with the sole purpose of establishing factors which determine the productivity of scientists in India. Nearly 200 variables influencing research productivity were collected, 26 variables were selected for further analysis and they were subjected to principal component factor analysis. The results indicated eleven factors affecting research productivity of scientists. Brocato (2005)

carried out a study to determine the research productivity of faculty in family medicine departments at U.S. medical schools, as well as the individual and environmental characteristics and prior socializing experiences predictive of research productivity. Cepero (2007) undertook a study in the USA to expand knowledge about faculty productivity and

cclii

the institutional and individual factors that may contribute to increased levels of faculty scholarly productivity. This study found Principal Component Analysis to be the best statistic to reduce the variables in this study and obtain variables that contribute a higher variation to the dependent variable. 4.4.5.1 Selecting the Variables for PCA The literature review done for this study initially extracted 33 variables that have been found to influence research productivity. This has been discussed in Chapter 2 of this study. These factors were broadly classified into institutional and individual factors. After several iterations (repetitions of analysis based on the criteria guiding the PCA) fourteen (14) variables were extracted. These variables were found to have high factor loadings and therefore contributed more variation to the dependent variable. These variables were divided into three categories namely institutional, individual and personal career development factors. These 14 variables are;
1. Age of the academic staff 2. Academic rank 3. Highest degree obtained

ccliii

4. Years since last highest degree 5. Self motivation 6. Research content knowledge 7. Research skills gained 8. Early orientation to research work 9. Personal work discipline 10. Resources for research 11.Rewards 12.Teaching load 13.Availability of technology 14.Availability of equipment for research.

ccliv

Table 4.30: Rotated Component Matrix (a)

1 Age group of respondent Position held in the University Highest degree obtained Years since you obtained last highest degree Extent to which research productivity is affected .807 by self motivation Extent to which research productivity is affected .858 by research content knowledge Extent to which research productivity is affected .830 by research skills gained Extent to which research productivity is affected .655 by early orientation to research work Extent to which research productivity is affected .720 by personal work discipline Extent to which research productivity is affected by resources available for research Extent to which research productivity is affected by rewards for research output Extent to which research productivity is affected .089 by teaching load Extent to which research productivity is affected by availability of technology e.g. internet and computers .229 .744 .031 .722 -.156 .097 .683 .025 .141 .766 .141 .181 -.078 .109 .005 .163 .053 .096 .005 .133 .054 -.014 -.035 .016 .055 Component 2 .072 .033 -.043 -.013 3 .850 .895 .808 .771

cclv

Extent to which research productivity is affected by equipment for research .129 .803 .013

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. A Rotation converged in 4 iterations.

The factor loadings of the variables of the three components are presented on Table 4.30. This table shows the factor loadings of each variable. Those variables that have the highest weights are the most important variables accounting for highest variations in the Principal components. For example, research content knowledge accounts for a high variation to the first Principal component. This is followed by research skills gained then the rest in that order. For principal component two, highest variations are accounted for by equipments for research, availability of technology in that order. In the last principal component, highest variation is provided by position held in the university, age group and highest degree obtained. The highest variations in this set are the academic position held in the university or the rank of the academic staff. These ranks are graduate assistants, lecturers, senior lecturers, professors and such. The result presented here basically confirms that the rank of the academic staff contributes highest to research productivity.

cclvi

Table 4.31: Correlation Matrix and Rotation Components 1 2 3 Extraction Sums of Squared Loadings Total % of Variance 4.061 29.0 2.818 20.1 1.964 14.0 Cumulative % 29.0 49.1 63.2

A further assistance was given in the interpretation of the three components, this involved performance of several rotations. The three components explained a total of 63.2% of the variance in the data with the first, second and third components contributing 29%, 20% and 14% respectively. The factor loadings of the variables of the three components are presented on Table 4.30. The eigenvalues extracted were 4.06, 2.82 and 1.96 for components 1, 2 and 3 respectively. These values were above the acceptable eigenvalue of 1. This is a clear indication that the data were sufficient for this analysis. It is these values that were used to construct the scree plot. Table 4.32: KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. Bartlett's Test of Sphericity Approx. Chi-Square df Sig. .770 1506.680 91 .000

cclvii

The Kaiser-Mayer-Olkin (KMO) measure of sample adequacy is .770 indicating that the data matrix has sufficient correlation to justify principal component analysis. Furthermore the Bartletts test of sphericity produced a high value and statistically significant which also meant that the data matrix was sufficient for PCA.

Figure 4.3: Scree plot for principal component analysis

cclviii

Scree Plot

Eigenvalue

cclix
1

Scree Plot

Eigenvalue

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Component Number

On the other hand the scree plot indicated a clear break after the fourth component. This confirmed the existence of the three components extracted (Figure 4.3).

4.4.5.2 Further Interpretation of the PCA

cclx

A total of 33 variables were used to extract the three components that affected research productivity in this study. After several removals of variables that did not meet the requirements, 12 variables were finally selected and used to extract three components that reflect various diversity of research productivity. They formed the basis to judge the factors influencing research productivity among academic staff in Kenya. The first component can be interpreted as personal career development factors that have contributed much in this component number one. This is in agreement with a study presented by Blackburn and Lawrence (1995). This component explained the highest variance of 29% in the data. The second component, contributing 20%, can be interpreted as institutional factors. Third factor, contributing 14% can be interpreted as demographic or experience of the individual researcher to publish. Component one and three are closely related in this analysis. This may mean that it is the individuals self determination, commitment, motivation, and stamina that do count in establishing whether an individual researcher is able to publish or not. The institutional component, explained 20% of the variance in the data involves all those resources, equipment or rewards that are supposed to be supplied by respective institutions.

cclxi

Ramsden (2005) argues that factors associated with high research performance in the study that he carried out, indicated that the strongest personal (i.e., individual) correlates were early interest in research, involvement in research activity, and seniority of academic rank. All these indicate that it was the individual factors that were coming out strongly to influence research productivity. This is in agreement with the findings of this study. Williams (2003) investigated the factors related to research productivity of human resource education and workforce development in the postsecondary faculty, and as a result classified related factors into three categories: environmental factors, institutional factors and individual interest and ability factors. This has also been reflected in this study. A study by Suwanwala (1991) investigated attitudes of research productivity of academic lecturers in Chulalongkorn University, the most famous institution in Thailand, and found that many lecturers did not realize the importance of conducting research, and many of them lacked the knowledge, skills, experience and resources to do research. This is basically a confirmation of the findings of this study. This study is quite clear that the extent of research mastery is a problem that needs to be solved first and foremost. This combines both lack of research skills and research content. Extent to which research productivity is affected by availability of technology e.g. internet and computers was one of the components raised in the PCA table above. It

cclxii

attracted a factor weighting of 0.744, this was quite significant in such a study. Internet is the new vehicle for access and delivery f information. Just like a vehicle, it is so useful to the person who knows how to use it. Wagner et al (1994) estimated that only 10% of academics at institutions with access to the Internet actually used it. He suggested that 30% of the users only used it for e-mail. Possible reasons for this lack of use were unawareness of available information sources on the Net and the lack of skills in locating the information needed. Adams and Bonk (1995) attributed barriers to use amongst academics to the lack of time, and lack of training on how to use. As observed by Lazinger,Barllan and Peritz (1997), most of the studies on Internet use by academics only focused on the users, leaving the reasons for non-use unexplored. Lazinger,Barllan and Peritz reported about 80.3% (371 out of 462) of academic staff members from the University of Jerusalem were Internet users. Similarly, White (1995) found that between 72% and 73% of the academics sampled consulted the Internet for their information needs. Other reported scholars have mixed response of the effect on the use of the Internet. This was an expected result in the early years of electronic networks when the interface was not so user- friendly. Cohen (1996) indicated that academic staffs main use of CMC (Computer mediated communications) is for the e-mail facilities to and from other faculties, both on-

cclxiii

and off-campus. Other usages of the Internet include FTP (File transfer protocol), Telnet and Gopher. The study also found low use of the electronic journals. In the United States, McClure, et al. (1996) reported on the impact of networking on the academic institution. Abel, Liebscher and Denman (1996) reported on academic scientists and engineers use of electronic networks mainly for e-mails, electronic discussion groups, access to databases, running programs and file transfer. Lazinger, Barllan and Peritz found that 362 outof 371 respondents used the Internet for e-mail and most e-mail correspondences were research related. White (1995) found that the younger academic staff tended to use the Internet more than those older. A reason for this may be that, the former group participated in electronic discussion groups and were less dominated by those higher in status. Cohen (1996) investigated the use of computer networks by 888 academic staff, and found higher use of the computer networks by younger academics. Applebee, Clayton and Pascoe (1997), however, did not find age a significant factor in the use of the Internet and proposed that older academics have caught on the use of computers in their work. The effect of age on computer use therefore is inconclusive, especially in the current situations where all academics regardless of age are dependent on the computers for their teaching, research and administrative work. The effect of gender on computer use has also been investigated.

cclxiv

A study by Ruthand Gouet (1993) found greater number of female academics using the computer networks. White (1995) discovered that female academic staff makes significantly higher use of the Internet than their male counterparts. In contrast, Applebee, Clayton and Pascoe (1997) found a higher number of male academics from the University of Canberra used the e-mail than the women. Cohens study (1996) also indicated a greater proportion of females using the network than males. Most of these studies are confirming that most lecturers at the university level do not have the capacity to use the internet for research work. However, most of these studies were done in the 90s when the usage of the internet was not wide spread. Efforts may be needed to train the university academic staff on the more ways in which the internet can be more useful for academic work than the emails use only. 4.5 Attitudes of academic staff on Research and Publications in their institutions The lecturers attitudes were measured on a five point Likert scale where strongly Agree = 5, Agree = 4 Undecided = 3, Disagree =2 strongly Disagree = 1. The scale was reversed in cases where the question was negatively constructed. The total score of frequencies for all responses from all respondents was calculated. This was used to calculate the mean attitude score. It is this mean attitude score that was used by SPSS Version 15 to run the one way ANOVA against the demographic variables identified.

cclxv

4.5.1 Lecturers total attitude scores on research and publications. Table 4.34 displays the total frequency distribution of the lecturers scores on their attitude towards research and publications. It gives a general picture on their attitude towards the listed statements. From this information it can be seen that most of the lecturers fall in the category of undecided. Therefore, the analyses that follow will give a true picture of the lecturers attitudes towards research and publishing. The analysis of this part was done in a 2 scale nominal scale that categorized the attitude scale into two. All the statements that scored more than 50% were categorized in another group. Under the strongly disagree, there were several statements. Notable ones were, publishing is important for any lecturer aspiring to grow professionally 70%; research and publishing increases the visibility of the university 64.6%. These two statements clearly indicate some tone of pessimism on the part of the academic staff. It gives a feeling of despair and hopelessness in regard to research and publishing. On the other hand, the academic staff are in agreement that academic staff should spend family resources in research and publishing. This is ironical, the statement made here is that research is important and therefore family resources can be used to conduct researches, on the other hand the academic staff are saying that publishing is not important for anybody striving to

cclxvi

grow professionally. Perhaps the message here is that the academic staff are not seeing a contribution of publishing as a criteria for professional promotion. They are saying that publishing is crucial but it should not be used as a criterion for promotion. Some academic staff in this study had indeed indicated that research and publishing alone should not be used as a yard stick for promotion. Some argument was that senior university academic staff should also be promoted on the basis of research and publishing in the areas of specialization. Another notable issue here is the case where the academic staff are implying that money meant for research should be used for academic staff salaries (50.2%). This is perhaps the first and only incidence in this study that the academic staff have raised the issue of salaries. This contradicts their individual suggestions in table 4.49 where the issue of salaries was raised by a paltry 4.0%. Perhaps this is because they had captured it in this section. The Kenyas UASU has done a good job in agitating for better salaries for its staff and it is hoped that will translate to better research output.

Table 4.33: Frequency distribution of lecturers attitude scores on Research and Publications in their institutions

cclxvii

Attitude Statements

Strongly Disagree f % 128 46.2 88 118 31.8 42.6

Disagree f 64 85 77 % 23.1 30.7 27.8

Undecided f 19 26 15 % 6.9 9.4 5.4 f 33 37 35

Agree % 11.9 13.4 12.6 f 25 38 29

Strongly Agree % 9.0 13.7 10.5

Means

SD

1 2 3

My department should offer additional pay for publishing Lecturers should be offered less teaching load for publishing in refereed journals. Universities should offer monetary rewards as an incentive for publishing in

3.88 2.46 2.20

1.37 1.41 1.38

refereed journals. Universities should spend money meant for research on improving salaries for

22

7.9

21

7.6

15

5.4

77

27.8

139

50.2

4.06

1.26

5 6 7 8 9 10 11 12 13

lecturers Promotions should be based on publishing alone Teaching alone without research is important Lecturers should spend family resources on research and publishing Research and publishing increases visibility of a university Publishing is important for any lecturer aspiring to grow professionally Universities should be ranked purely on the basis of their research productivity Teaching is more important than publishing I'll rather spend my time as part time lecturer than on writing articles Appointment to senior university management should be based on

20 52 5 179 195 57 18 10 52

7.2 18.8 1.8 64.6 70.4 20.6 6.5 3.6 18.8

26 45 7 57 54 73 16 25 89

9.4 16.2 2.5 20.6 19.5 26.4 5.8 9.0 32.1

14 22 19 10 6 17 35 23 29

5.1 7.9 6.9 3.6 2.2 6.1 12.6 8.3 10.5

123 59 90 14 8 82 104 109 64

44.4 21.3 32.5 5.1 2.9 29.6 37.5 39.4 23.1

88 88 152 15 13 44 97 108 41

31.8 31.8 54.9 5.4 4.7 15.9 35.0 39.0 14.8

3.86 2.68 4.38 1.65 1.51 2.94 3.91 4.02 2.83

1.19 1.55 0.87 1.13 1.02 1.43 1.15 1.08 1.37

14 15

academic writing Lecturers who do not publish have no business teaching in university It is advisable for lecturers to spend personal resources on research and

50 91

18.1 32.9

52 43

18.8 15.5

40 30

14.4 10.8

74 61

26.7 22.0

61 49

22.0 17.7

3.16 3.02

1.43 1.43

16

publishing Developing countries need research more than developed countries

72

26.0

66

cclxviii

23.8

19

6.9

57

20.6

61

22.0

2.89

1.54

4.5.2 University Academic Staffs Mean Attitude Score towards Research and Publishing The mean score for each question was worked out. All negatively states items were reversed so that all the statements became positive. The responses were assigned scores as follows; SA 5 A4 U3 D2 SD 1 The minimum total that a respondent could score was 16 assuming that he scored SD = 1 in all the 16 items). On the other hand, the maximum score was 80 (assuming that the respondents strongly agreed in all the 16 items in the questionnaire). It is these totals that were used to calculate the mean scores for each of the following independent variables.

Table 4.34: Academic Staff gender mean attitude score

Gender

Mean

Std. Deviation

cclxix

Male Female Total

169 74 243

49.8 49.2 49.6

9.83 7.75 9.23

From Table 4.34 above, both male and female academic staffs tend to have equal attitude mean scores towards research and publishing. Both sexes are in agreement

concerning their attitude towards research and publishing. However, these means have to be tested in order to make conclusions.

4.5.2.1 Hypotheses testing The Ho was tested at 0.05 level of significance. Decision rule: When the P Value is less than or equal to .05, the null hypothesis is rejected meaning there is significant difference between the variables under study. If, however, the P Value is greater than .05 level of significance, the null hypothesis is not rejected meaning that there is no significant difference between the variables being tested.

Ho8 There is no significant difference between male and female attitudes towards research and publishing. Ha8 There is significant difference between male and female attitudes towards research and

cclxx

publishing.

Table 4.35: ANOVA test of difference in the gender mean attitude scores towards research and publishing

ANOVA SUMMARY TABLE Sum of Gender Between Groups Within Groups Total Squares 22.844 20606.078 20628.922 df 1 241 242 Mean Square 22.8 85.5 F .267 Sig. .606

In the One way ANOVA Table above, the P-value for gender is 0.606. This is greater than the set 0.05 level of significance. Decision: The null hypothesis is not rejected. Conclusion: thus there is no significant difference in academic staffs attitude towards research and publishing. Therefore both male and female academic staff have almost same attitude towards research and publishing. This is supported by the actual figures produced by these two groups of academic staff in Table 4.14 and the total attitude mean scores presented in Table 4.34 above. Their output for the 5 year period was about the same. In the one way ANOVA Table 4.35, the lecturers attitudes on research and publishing when

cclxxi

categorized by gender is .606. This is higher than the P Value of .05. This hypothesis is therefore not rejected. This means that there is no significant difference in attitudes of university academic staff on research and publishing when categorized by gender. In other words, both male and female academic staff has same views on research and publishing in the universities in Kenya. On the other hand, Bonnett (2004) analyzed 900 research articles in nine major Evolutionary Ecology Journals in order to examine how gender influences research output. The study found that women and men differed in areas of research, with women much more likely to conduct projects on behavior rather than evolution or ecology. Studies on differences between gender and research productivity has been done by various scholars. These studies have tended to have contradictory findings. The findings of the present study has made a blanket finding among Kenyan scholars that there is no significant difference among them when categorized by gender. Table 4.36: Academic Staffs types of university mean attitude score

Type of university Public university Private university Total

n 171 74 245

Mean 50.3 48.1 49.7

Std. Deviation 10.4 5.58 9.22

From Table 4.37 above, public university academic staffs have a mean attitude score which is higher than that of the private universities. This is in agreement with the

cclxxii

age-group categorys attitude towards research and publishing. This can be explained by the large number of older academic staff teaching in public universities than at the private universities and the length of time that these institutions have been in operation since inception. Most public universities are older than most private universities in Kenya. .

Ho9 There is no significant difference between public and private universities academic staffs attitude towards research and publishing Ha9 There is significant difference between public and private universities academic staff attitude towards research and publishing Table 4.37: ANOVA test of difference in academic staffs type of university mean attitude scores towards research and publishing
ANOVA SUMMARY TABLE

Sum of Type of university Between Groups Within Groups Total Squares 244.288 20505.222 20749.510 df 1 243 244 Mean Square 244.3 84.4 F 2.895 Sig. .090

In the One way ANOVA Table above, the P-value for type of university is 0.090. This is greater than the set 0.05 level of significance.

cclxxiii

Decision: The null hypothesis is not rejected. Conclusion: The null hypothesis indicates that there is no significant difference in academic staffs type of university and attitude towards research and publishing. In other words the academic staff from both the public and the private universities has no differing attitudes towards research and publishing. This means that there is no significant difference in attitudes of university academic staff on research and publishing when categorized by the type of university. Here there were two categories of universities, the public and the private Universities. This finding therefore indicates that there is no significant difference in terms of research output of those lecturers from the public or private universities in Kenya. Therefore the type of university does not influence the amount of research output to be done by the university academic staff. Table 4.38: Academic Staffs age group mean attitude score
Age group 20-29 30-39 40-49 50-59 60-69 Total

n
26 70 90 48 11 245

Mean 45.7 47.0 49.8 53.5 58.2 49.7

Std. Deviation 7.58 7.52 7.94 10.5 14.9 9.22

From Table 4.38 above, it is clear that the older the age group of a lecturer, the

cclxxiv

more attitude score he/she has towards research and publishing. This is a reflection of the appreciation that one develops as he/she grows old in the profession. This is a common phenomenon even in other fields where young professionals are not fully satisfied with their jobs. They keep on searching for better jobs or the so called greener pastures. The life-cycle model (Diamond, 1986; Hu and Gill, 2000) predicts that faculty research productivity will decline as an individuals academic experience increases. Sometimes, the estimated regression coefficient of the variable years of academic employment is negative. One plausible reason for this decline in research productivity in the developed countries is the decline of extrinsic motivation as a result of attainment of tenure and promotion and the proximity of retirement (Diamond, 1986). Another factor may be that senior faculty members tend to have more service and administrative responsibilities, which may hinder their research productivity. Overall, this cannot be concluded for the case of Kenya because it was not supported by data. Another possible reason is that there was no respondent above the 70 year mark. Ho10 There is no significant difference between age groups attitude towards research and publishing. Ha10 There is significant difference between age groups attitude towards research and publishing.

cclxxv

Table 4.39: ANOVA test of difference in academic staffs age group mean attitude scores towards research and publishing

ANOVA SUMMARY TABLE

Sum of Age group Between Groups Within Groups Total Squares 2414.014 18335.496 20749.510 df 4 240 244 Mean Square 603.58 76.39 F 7.899 Sig. .000

In the One way ANOVA Table above, the P-value for type of university is 0.000. This is less than 0.05 level of significance. Decision: The null hypothesis is rejected. Conclusion: Thus there is a significant difference in lecturers age group and attitude towards research and publishing. Thus one age-group has differing attitude towards research and publishing. This means that there is a significant difference in academic staffs attitudes on research and publishing as categorized by the age group of the respondents. It also means that each age group has differing opinions on the management of research and publishing in the universities. This is in agreement with the earlier data where the older age groups are seen to be more active in research and publishing than the other groups. In a recent study,

cclxxvi

Turner and Mairesse (2003) analyzed the impact of research productivity relative to age, gender and education of French condensed matter physicists. The study found that there was a quadratic relation between the age of the scientists and the number of publications, with researchers productivity increasing before 50 and then declining after 51. The results of that study, using citations, were not significantly different from those obtained with publications. Table 4.40: Academic Staffs rank mean attitude score
Rank Graduate Assistant Tutorial Fellow Assistant lecturer Lecturer Senior Lecturer Associate Professor Professor Total

n
15 18 42 102 41 14 8 240

Mean 45.7 49.4 48.8 48.9 51.1 52.1 57.1 49.6

Std. Deviation 4.92 7.72 8.43 8.17 11.3 13.2 13.4 9.28

Ho11 There is no significant difference between academic staffs rank attitude mean scores to research and publishing. Ha11 There is significant difference between in academic staffs rank attitude mean scores to research and publishing. Table 4.41: ANOVA test of difference in academic staffs rank mean attitude scores towards research and publishing

cclxxvii

ANOVA SUMMARY TABLE Sum of Rank Between Groups Within Groups Total Squares 940.554 19633.942 20574.496 df 6 233 239 Mean Square 156.759 84.266 F 1.860 Sig. .089

In the One way ANOVA Table above, the P-value for type of university is 0.089. This is greater than the set 0.05 level of significance. Decision: The null hypothesis is not rejected. Conclusion: thus there is no significant difference in academic staffs rank and attitude towards research and publishing. This hypothesis is therefore not rejected. This means that there is no significant difference in attitude of academic staffs attitudes on research and publishing as categorized by the rank of the respondents. It also means that rank of academic staff has no differing opinions on the management of research and publishing in the universities. This is not in agreement with the earlier data where the older age groups are seen to be more active in research and publishing than the other groups. The other section in this study especially section 4.3.2.4 and table 4.22, there is a clear indication that; rank of academic staff is significant in research productivity.

cclxxviii

However, since this was just the attitude of the academic staff, the results indicated here could just mean what it has measured in attitudes. Table 4.42: Academic Staffs highest degree mean attitude score

Highest degree Bachelors degree Masters degree PhD Total

n
16 133 91 240

Mean 46.9 48.7 51.5 49.6

Std. Deviation 4.63 8.36 10.8 9.29

From Table 4.42 above, it is evident that Bachelors degree holders teaching at the university have a lower attitude score towards research and publishing. On the other hand, the masters and PhD holders have a higher attitude score towards research and publishing. However, the PhD holders have a higher attitude score than their masters counterparts. This is a repeat of earlier scenarios where the persons who have served in the university environment for long, have favourable attitude to research and publishing than those who have been in the university in a shorter period.

Ho12 There is no significant difference between highest degree obtained and attitude towards research and publishing.

cclxxix

Ha12 There is significant difference between highest degree obtained and attitude towards research and publishing.

Table 4.43: ANOVA test of difference in academic staffs highest degree mean attitude scores towards research and publishing
ANOVA SUMMARY TABLE Sum of Highest degree Between Groups Within Groups Total Squares 558.867 20085.128 20643.996 df 2 237 239 Mean Square 279.434 84.747 F 3.297 Sig. .039

In the One way ANOVA Table above, the P-value for type of university is 0.039. This is less than 0.05 level of significance. Decision: The null hypothesis is rejected. Conclusion: Thus there is a significant difference in lecturers highest degree obtained and attitude towards research and publishing. . This means that there is significant difference in attitudes of university academic staff on research and publishing when categorized by highest degree obtained. In other words, both PhD and Masters Degree holders have differing views on research and publishing in the universities in Kenya

cclxxx

Table 4.44: Academic Staffs years since highest degree was obtained mean attitude score

Years since last highest degree <5 Years 6-10 Years 11-15 Years 16-20 Years More than 20 Years No response Total

n
113 63 43 12 8 3 242

Mean 47.1 50.5 52.9 51.8 55.6 51.7 49.6

Std. Deviation 7.18 10.7 9.01 9.91 12.7 5.86 9.16

From Table 4.44 above, it is evident that those members of academic staff, who got their last highest degree in less than 5 years, have an unfavourable attitude towards research and publishing whereas those who have taught for more than six years have a favourable attitude towards research and publishing.

cclxxxi

Ho13 There is no significant difference between years since highest degree obtained and attitude towards research and publishing. Ha13 There is significant difference between years since highest degree obtained and attitude towards research and publishing.

Table 4.45: ANOVA test of difference in academic staffs years since highest degree was obtained mean attitude scores towards research and publishing
ANOVA SUMMARY TABLE Years since last highest degree Between Groups Within Groups Total Sum of Squares 1608.602 18596.703 20205.306 df 5 236 241 Mean Square 321.720 78.800 F 4.083 Sig. .001

In the One way ANOVA Table above, the P-value for type of university is 0.001. This is less than 0.05 level of significance. Decision: The null hypothesis is rejected. Conclusion: Thus there is a significant difference in lecturers years since last highest degree was obtained and attitude towards research and publishing. Table 4.46: Academic Staffs University mean attitude score

cclxxxii

Name of university A B C D L E F G H J K Total

n
34 67 19 12 3 11 8 26 26 22 17 245

Mean 46.8 55.4 51.1 42.6 47.3 47.3 45.9 49.7 47.7 47.1 45.9 49.7

Std. Deviation 6.2 11.9 9.70 11.3 6.41 3.98 5.06 5.64 6.15 5.49 5.74 9.22

From Table 4.46 above, it is evident that University B has a higher attitude towards research and publishing than the other universities in the study. A majority of the universities had an attitude score less than 50 towards research and publishing. Ho14 There is no significant difference between universities and attitude towards research and publishing. Ha14 There is significant difference between universities and attitude towards research and publishing. Table 4.47: ANOVA test of difference in academic staffs university mean attitude scores towards research and publishing

cclxxxiii

ANOVA SUMMARY TABLE Sum of Name of university Between Groups Within Groups Total Squares 3781.416 16968.095 20749.510 df 10 234 244 Mean Square 378.142 72.513 F 5.215 Sig. .000

In the One way ANOVA Table above, the P-value for type of university is 0.000. This is less than 0.05 level of significance. Decision: The null hypothesis is rejected. Conclusion: Thus there is a significant difference in different universities attitudes towards research and publishing.

4.6 Lecturers perceived factors hindering research productivity The data presented below was synthesized from the last section of the lecturers questionnaires. This data were collected by an open ended question that sought the lecturers views on the problems and possible solutions facing academic staff in their pursuit to publish. The qualitative data were grouped and analysed into emerging themes. Manual tallying was employed in this section. It is important to note that the qualitative data in this section supports the quantitative data presented earlier.

cclxxxiv

This is the only section that has highlighted the issue of individual and institutional factors that hinder research productivity in this study. The lecturers have given about 30 issues that they feel are hindrance to research productivity in this country. The data collected was grouped into three main categories in terms of their percentages. These groups are presented below; 0 8% = 1 9 17% = 2 18 22% = 3

Table 4.48: Academic staff views on problems hindering research productivity

cclxxxv

Possible Problems 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. Inadequate research fund Heavy teaching load Inadequate facilities and equipments Lack of research motivation Lack of enough time for research Poor salaries Inadequate research knowledge in some lecturers Lack of institutional support Poor attitude towards research Lack of journals specifically African or Kenyan Engagement in Consultancy services Poor leadership/politics Low recognition of research publications Long procedures in accessing research funds Lack of research monitoring systems Poor promotion criteria Lack of implementations of research findings Research proposal writing problem Lack of research coordination No mentorship and encouragement by seniors Poor terms of employment Expensive publication process Poor family support/family responsibilities Inadequate exchange programs in universities Commercialization of university education Delay in disbursement of research funds Lack of research supervision Laziness Promotion not based on merit and performance Misuse of research fund

Frequencies 100 76 56 46 34 19 18 13 13 11 9 9 8 7 6 5 5 5 5 5 4 3 3 3 3 2 2 2 1 1

Percentage 21.1 16.0 11.8 9.7 7.2 4.0 3.8 2.7 2.7 2.3 2.0 2.0 1.7 1.5 1.3 1.1 1.1 1.1 1.1 1.1 0.8 0.6 0.6 0.6 0.6 0.4 0.4 0.4 0.2 0.2

GROUPS 1 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

Under category 3 there is only one problem presented here and this has a percentage of 21.1% this was explained as inadequate research fund. The academic staff who said this constituted slightly less than a quarter of all the respondents in this study. Their argument is

cclxxxvi

that due to lack of financial resources to conduct research is an hindrance to research efforts in this country however, one respondent, from the open ended items opposed this and said that there is too much money outside there for research, it is only that our academic staff are not willing to write winning proposals. Others said that the little available financial resources are not well administered; instead it takes too long to disburse the research funds. Available evidence indicates that most universities in the study have research funds of some sort, but the administration of these funds could not be established by this study. Efforts are needed to find out the problem facing research funding in this country. Others indicated that there was gross misuse of research funds. It was not clear whether this was on the part of the fund administrators or it was the research fund recipients. Under group two, there were three issues raised that hinder research productivity, these were heavy teaching load (16%), inadequate research facilities (11.8%) and lack of research motivation (9.7%), when combined, this forms about a third of the total respondents. Arguments placed under heavy teaching load are that the academic staffs are allocated many teaching subjects that make them less effective in discharging their research mandates. This can also be due to the many lessons that academic staff take especially when teaching other colleges. This is the so called moonlighting. The solution provided by

cclxxxvii

themselves here is employment of more academic staff. The issue of employment of academic staff in Kenya is a responsibility of individual universities councils. The highest administrative organ in the university set up). In this case it is not easy to manipulate all these institutions to effect a change in their employment policies. This has led to recruitment of staff in part time basis across most universities across the country. This is in a way to cut costs and the overheads that come along with hiring permanent staff. The second factor under this category is the issue of motivation. They are putting forward an argument that they are not well motivated to conduct research in their respective universities. Again this is an issue that can be well handled by their respective university councils. It is true that any worker needs positive motivation to enable him/her perform better in his/her place of work. The first category comprises of 25 items or issues that are attracting percentages of between 0.2% to 7.2%. The top one in this list is lack of enough time to conduct research. This problem is quite related to the issue of teaching load. The assumption that the academic staff are making is that heavy teaching load is reducing or denying them enough time to conduct research. This is followed by poor salaries. This is in position six from the top. This attracted 4% of the respondents. It is evident here that most academic staff in Kenya is not keen on salary increase but rather enabling environment to conduct research.

cclxxxviii

The next item in this group is lack or inadequate research knowledge with some academic staff. This problem has also been raised by the Heads of departments in the section that follows. This is true because any job requires regular updates on the changing ways and techniques of performing the tasks involved. In most organizations this is the issue of refresher courses and seminars to shape skills. The area of proposal and article writing is competitive especially when the publications are to be made in journals outside the country. It is interesting to see that this problem has been raised by the academic staff themselves. Other issues raised by the academic staff are; lack of institutional support. This goes along with poor institutional leadership, politics and poor promotion criteria. It is no secret that some universities have been accused of poor leadership. This weak leadership transcends to all sectors of the university even the acquisition of journals for the university library may be hampered. This also demoralizes the academic staff in their endeavor to publish. A poor promotion criterion was also highlighted. The argument put forward here indicated that some promotions were done irregularly hence demoralized the academic staff. One academic staff member mentioned the issue of promotion with fake publications. I think the point being made but the publications are not properly authenticated.

cclxxxix

Expensive publication process was also mentioned by the academic staff. The argument fronted by these academic staff was that it was expensive to publish an article in local university publication centers. Some journal publishers request the contributors to raise some fee for their publications to be made. It is indeed expensive to run a journal regularly. The duration taken before a publication is finally printed might be long too. These are the challenges that face academic staff in Kenya in trying to publish. Issues raised in this section by the academic staff mainly hinge on lack of the number of the academic staff who feel that the little finances available should be well managed and administered.

ccxc

Table 4.49: Heads of Department views on problems hindering research productivity

Possible Problems
1. 2. 3. 4. 5. 6. 7. Inadequate research funding Lack of research motivation Inadequate facilities and equipments Heavy teaching load Poor salaries for lecturers Lecturer engagement in consultancy work Inadequate research knowledge among some lecturers

Frequencies 4 4 4 4 2 1 1 1

Percentage 16.7 16.7 16.7 16.7 8.3 4.2 4.2 4.2

8. Lack of enough time for research

ccxci

9. Lack of implementations of research findings 10. Low recognition of research publications 11. Poor attitude towards research

1 1 1

4.2 4.2 4.2

From the above Table 4.49 it can be seen that four key factors influence research productivity among university academic staff in Kenya as observed by the heads of departments. These factors are; inadequate research funding, lack of motivation, inadequate research equipment and heavy teaching load. Some of these factors have also been highlighted by lecturers in this study. Basically this is a confirmation to the concerns of lecturers. It is interesting to note that the issue of salaries has not featured prominently by both lecturers and heads of departments as a factor hindering research productivity. Inadequate research knowledge among some lecturers has also been identified here by the heads of departments. This is confirmation by the findings of the PCA analysis. The issue of poor attitudes towards research has also been raised by the academic staff, the root causes of these attitudes have not been established in this study. A revisit of this should be done so as to find out reasons that make the academic staff to have a poor attitude towards research. 4.6.1 Enhancement of research productivity. The following section deals with the last research question that sought to establish how research productivity among academic staff in the selected universities could be

ccxcii

enhanced. After studying the problems that face or hinder academic staff in their quest to publish, the academic staffs and Heads of Departments gave some possible solutions to tackle the problems at hand. These solutions that were provided, forms a basis for enhancing research productivity of academic staff. Therefore, this section reviews the possible solutions to the problems raised.

4.6.1.1 Academic staff views on possible solutions to problems hindering research productivity The information presented below (Table 4.50) provides possible solutions to the problems perceived to influence lecturers ability to engage in productive research activities. This information was analysed and grouped into emerging themes. The information provided clearly indicates that it is a true reflection of the possible solutions to the factors that influence research productivity as perceived by the lecturers. The Principal Component Analysis identified individual factors to be the main component that is affecting research output among academics in Kenya. The key among the individual factors

ccxciii

was the content knowledge and skills gained in engaging in meaningful research. This is the area that needs to be addressed if the academics in Kenya are to make meaningful contribution to research output.

Table 4.50: Academic staff views on possible solutions to problems hindering research productivity

Possible Solutions

Frequencies

Percentage

ccxciv

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

Allocation of more funds for research Employ more teaching staff Reward researchers by promotions and money Buy modern equipments for research Universities to allocate less teaching hours Revise lecturers pay Engage private sector in funding research Create training in proposal and paper writing Develop research policies Build modern facilities for research Provide clear institutional support Encouragement by the old professionals Consider other research contributors when giving promotions Give research and teaching equal time Promotions strictly on productivity and junior staff should not be in administration Implement research findings Establish credible journals in Kenya Clear guidelines in ones role in research Link and exposure to funding organizations Employ lecturers that are on contract permanently Universities should enter into exchange programs with other universities Universities to establish part time lecturers positions and pay for their services Reduce the number of enrollment for students in universities Early orientation in research work Initiate external research supervision Family to support research

79 48 42 32 28 28 19 18 17 16 11 10 10 10 10 9 8 8 6 4 4 3 3 2 2 1

18.5 11.2 9.8 7.5 6.5 6.5 4.4 4.2 4.0 3.7 2.6 2.3 2.3 2.3 2.3 2.0 1.9 1.9 1.4 0.9 0.9 0.7 0.7 0.5 0.5 0.2

The lecturers have given the solution to this problem as trainings in proposal and paper writing skills. Ironically, this has only been proposed by 4% of the respondents. The

ccxcv

focus of lecturers possible solutions has mainly dwelt on institutional solutions to be provided by their respective institutions. The key among the institutional factors raised was the issue of inadequate research funds. The academic staff are of the opinion that there is no enough funds allocated to do research in this country. But they are also arguing that the available funds are mismanaged, come in too late and is not enough. Over 18% of the academic staff is of the opinion of boosting the research fund with the aim of increasing research productivity at the university level. This is an issue that has been raised by both the academic staff and the Heads of Departments. It is indeed true that no meaningful research can take place without financial inputs. The academic staff are also suggesting that more teaching staff should be employed to reduce the existing teaching load among the teaching staff. Rewards and recognition of outstanding scholars or researching academic staff has also been proposed by the academic staff. Any organization must employ a rewards system for the best performers in their organizations. This will indeed boost the morale of the writing academic staff and also encourage the rest of the academic staff to be engaged in research work. The issue of buying modern laboratory equipments for research has also been used here. There is need to heed this call because there is no way a lecturer will conduct research

ccxcvi

if he doesnt have the requisite tools to enable him/her too perform the work at hand. Most of these equipments in Kenyan universities are archaic and need and overhaul. The changing times needs better equipment. The national polytechnics in Kenya have gotten a facelift from a number of countries, particularly Italy to revamp their laboratory equipments to acceptable standards. The private sector has been proposed as a key partner in research. The private sector has been engaged in government business of late in what is referred to as the Public Private Sector Partnership (PPSPs). This has been effectively carried out under the auspices of the Kenya Private Sector Development Strategy (KPSDS) whose aim is to reduce the wide gap between the mainstream research institutes in the country. Basically there is need to engage the private sector in the management of research in this country. They will provide the funds, and the academic staff will provide the expertise. Another possible solution raised by the academic staff is the issue of training the academic staff is the issue of training the academic staff on proposal and paper writing skills. This has been discussed in separate section of this study. This is an issue that can be implemented within a short period of time. The senior members of staff should be encouraged also to write training manuals and handbooks on topics such as, writing wining

ccxcvii

proposals, research techniques and methods and such. Such materials together with on the job training will help the academic staff to engage more in research work. Establishment of academic journals in Kenya was another solution offered. Here, the argument was that there are no credible journals in Kenya. This may not be true; there are a number of journals being produced by different faculties in different universities in Kenya. The only problem is that some of the journals do not publish regularly. They are not up to date. This is mainly due to financial crisis or lack of articles to publish. The Chairman, Kenya National Academy of Sciences confided with this researcher that they did not have enough articles to publish in their three national journals. This is partly due to the fact that most academic staff prefers to publish in higher notch journals that are based outside the country. Implementation of research findings is another issue raised. This is possible only if the policy makers are able to agree with the findings and push for their implementation. This is a complicated issue for a developing economy like Kenya. The Royal Society in the UK has researchers and scientists attached to some of the members of parliament. They provided a link between policymakers and the academic world. This is working so well too far for Britain. It is called MP- Scientist pairing scheme, this has started to be implemented

ccxcviii

in Kenya in a small way. This will maybe enable the implementation of the science findings in Kenya. These views are in agreement with Doyle (2006) who suggested that reward and promotion system in colleges in which teaching has been the primary role of faculty should be changed. These changes in priorities will allow increases in the incentive system to encourage scholarly productivity, especially in higher education institutions which have not been traditionally focused toward research endeavors. A second group of strategies has been focused on fostering publication outputs. McGrail, Rickard and Jones (2006) examined 17 studies that described strategies used by higher education strategies to foster faculty publication rates. The strategies identified by the authors were oriented to fostering the writing abilities and motivation. There were three types of intervention identified by the authors: writing courses, writing support groups, and writing coaches (McGrail et al., 2006). McGrail et al. (2006) found that all three models can be beneficial to increasing publication rates. The third strategy is to establish support systems for research in addition to writing. Goodwin et al. (2006) described an example of a center to support faculty research established at the school of education at the University of Colorado at Denver. This center included services such as: consultation with research associates; literature search, editing

ccxcix

and manuscript preparation; data transcription; and funding for conferences presentations. Goodwin et al. (2006) found in the analysis of the first year of operation of the center that faculty publication increased. They suggest that it is important to include services to foster processes involved in research grant-seeking. These issues have been raised by the academic staff who participated in this study. It is important for the academic staff to be involved in refresher courses on the main issues involved in research and publishing. Researchers (Buchheit et al., 2001;Hu and Gill, 2000; Tien, 2000) have previously shown that faculty research productivity was a result of the interaction among many endogenous and exogenous variables, including individual personal characteristics; academic discipline; educational background; previous employment; institutional characteristics; and teaching, research, and service assignments. This basically calls for a multi-dimensional approach to enhancing research productivity in Kenya. Effort is needed in solving all problems hindering research productivity both at the institutional level and at the individual level. 4.6.1.2 Heads of Departments views on possible solutions to problems hindering research productivity Table 4.51: Heads of Department views on possible solutions for problems hindering research productivity

ccc

Possible solutions 1. Allocation of more research funds 2. Buy modern equipments for research 3. Employ more teaching staff 4. Reward researchers 5. Revise lecturers pay 6. Private sector intervention 7. Building modern facilities for research 8. Early orientation to research work 9. Implementation of research findings 10. Establish research refresher courses 11. Promotions on research productivity 12. Clear guidelines in ones role in research 13. Universities should enter into exchange programs

Frequencies 3 3 3 3 2 1 1 1 1 1 1 1 1

Percentage 13.6 13.6 13.6 13.6 9.1 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5

Table 4.51 presents possible solutions which are a true reflection to the issues raised by both lecturers and heads of department. It is evident that research funding, equipment for research, recruitment of more staff and rewards for publishing are seen as possible solutions to the challenges facing researchers in Kenya. It is interesting to note here that the respondents, in this case the heads of departments are recommending the introduction of research refresher courses. A research fund is one of the issues that the heads of departments have recommended. Indeed nothing much can be achieved without the input of financial resources to any enterprise. The Universities must look for alternative sources of funding to support research in their universities. One of the most talked about

ccci

sources is to have an endowment fund associated with the alumni fund to raise funds for research. Another one can be venture capital. This is whereby a discovery or invention made by the university can be sold out or partnered with an entrepreneur interested in the new created knowledge. Most of the universities in the west have benefited from such kind of arrangement. Income generating projects by the universities can also raise funds for universities to conduct research. In this study, the heads of departments have indicated that research productivity is not high because of lecturers perceptions of a lack of a motivating environment; for instance, lecturers have insufficient equipment and materials to pursue research in a satisfactory manner. Tools for doing the job so as to motivate the academic staff to work hard in their areas of specialization. Efforts should be targeted towards this direction to overhaul the equipments in the universities so as to reflect the modern society. At present the National polytechnics are in the process of overhauling their archaic machinery and equipment. Both members of the academic staff and heads of departments are recommending the employment of more staff in the universities. They are making a point that they are over worked in their duties. This is a debatable issue that cannot be exhausted. This is because due to the rapid expansion of the university in Kenya, particularly after the

cccii

introduction of the so called module II programmes in Kenya in 1998, there has been pressure on lecturers to accept more teaching load. This has led some of them to moonlight in several institutions across the country. It is therefore possible that the academic staffs are now feeling the pressure of the heavy teaching load. The mushrooming of satellite campuses means that more admissions are being made to university this increase in enrollment demands more fro the academic staff. Respondents in a study conducted by Traynor and Rafferty (1998) identified 'the need for an improved research capacity' as a major issue facing nurses in academia, 'Research capacity building' has been defined as, 'a general term for a process of individual and institutional development which leads to higher skills and greater ability to perform useful research' (Trostle, 1992). Levin and Stephan (2001) hypothesized that receipt of federal research support would be a determinant of academic scientists research productivity. Although they do not report these specific results, they indicate that their findings were consistent with their expectations. Some universities in Kenya, like the Catholic University of East Africa, have about two hours once a week for research seminars for its members of staff and the postgraduate students. The academic staff also gave another recommendation to buy modern equipments for research. This is in agreement with the findings by Dundar and Lewis (1998) who

ccciii

developed and tested a more comprehensive model of faculty research productivity and found that library expenditures represented one of the important institutional attributes. Unequal facilities and funds are important since departments with more money and better laboratories, libraries, and other facilities are better equipped to train their staff and students, resulting in higher publication rates (Payne & Spieth 1935). Ideally, the enhancement of research productivity in Kenya cannot be complete without development of relevant research policies particularly at the institutional level. The research policies will address issues such as; Sourcing of funding for research Training of academic staff in research Awards for outstanding institutional academic staff Financing academic journals Developing of institutional centres of excellence Research partnerships and collaborations IPR sharing of benefits

ccciv

Ethical considerations in research Regular review of the policy

The research policies will not be complete without the IPR policies too. These two documents go hand in hand and by developing them, the academic staff will be in a position to maximize their potential in research productivity in cases where the IPR and research policies exist, there is need to review them so as to reflect on the challenges raised by the academic staff.

4.7 Document Analysis. Document analysis was done on the various research policies presented by a sample of the universities selected for this study. Some universities were not wiling to share their research policies. It is however interesting that most of the universities have research policies. Some universities have gone to the extent of putting the in the internet. Those that do not have the research policies are in the process to complete them. Other institutions are

cccv

reviewing their policies. The research policies sampled here were developed between 2001 and 2008. At University C there was the training for research competence. Here, the university recognized that ongoing research in the University was mainly individual effort with little inter-discipline collaboration within the university and was usually developed as projects as apposed to programs. In addition, there was limited capacity to write fundable proposals and to manage projects when such proposals were funded. The university therefore saw the need to develop the capacity of the academic and research staff to develop skills for the development and management of inter-discipline research programs through training workshops and courses including proposal writing, project management and reporting, whole-system-in-the-room workshops, donor relationships, financial management for nonfinancial managers etc. A percentage of the research fund (10%) has been set aside for in-house training of academic staff in such workshops and/or to send trainers to such training who will in turn return to build capacity within the university. This percentage caters for ccapacity building, awards, researchers/investigators, centre for research and development, product development etc. These items seem to be too many for the allocated 10% of revenues from the successful funding. There is need to have the resources for capacity building as a stand

cccvi

alone item. The university also recognized an important aspect of the professional development of all academic staff in attendance to seminars, workshops and conferences. Other aspects have been well covered in this research policy and these includes offices, committees and panels that support research chapter which contains documents related to University C, its research mandate, and bodies governing the university research including, the Centre for Research and Development, the University Research Review and Advisory Board, the Ethics Review, Inter-School Research Review Committee, Institutional Animal Care and Use and, Intellectual Property and Copyrights Committees. It also explains why University C needs a policy, the creation of a research fund and how that fund will be disbursed for various actions and finally how research coordination will take place in the University. Academic policies chapter contains documents related to principles concerning research, academic freedom, eligibility for Principal Investigatorship, openness in research, scientific misconduct, authorship, retention of research data, establishment of Independent Laboratories, student relationships with outside entities, and others. Financial aspects of sponsored projects and administration chapter contains documents related to fiscal responsibilities of PIs; indirect costs, their application to different types of

cccvii

projects, and procedures for obtaining waivers; cost sharing; tuition remission; property; and others Conflicts of commitment and interest chapter contains documents related to conflict of commitment and interest, and related forms; consulting policies for faculty and for Academic Staff-Research & Extension; policies related to start-up companies and equity acquisition. Intellectual property chapter contains documents related to patents, copyrights, and tangible research property. Environmental health and safety chapter contains basic policies, as well as specific requirements related to chemical hygiene, radiological hazards, lasers, and biohazardous agents; emergency procedures. Human subjects in research chapter is a description of University C's Human Research Protection Program (HRPP), including assurance of compliance with Ministry of Health requirements, and special guidance on the use of women, students and laboratory personnel. Laboratory animals in research chapter are on University C's Assurance of compliance with Public Health Services requirements, and other documents related to the care and treatment of animals.

cccviii

Non faculty research appointments chapter contains documents related to graduate student research assistants, academic staff-research & extension, postdoctoral scholars, visiting scholars and visiting researchers, and consultants.

This is indeed a brilliant idea by the university to recognize the need for staff development. However, the funding for this has been pegged on the availability of funding. It could have been better to start off the program by manpower training. At University A, there were no deliberate efforts made towards academic staff professional development. Instead, a recommendation was made to allocate an unspecified amount of funding towards training of academic staff involved in any successful research funding. Other areas dealt with under this policy include; Research Funding, Research Planning and Administration Approval, Monitoring and Control of Research Projects Contributions to the Research Administration costs Sharing/Disposal of Research Projects Resources

cccix

Remuneration of Research Staff Creating a Conducive Research Environment Dissemination of Research Results Proprietorship of Research Outputs Monitoring-of Research Impact Ethical and Environmental Considerations

At University F there was a detailed research policy and particularly on the research forum/seminars and awards for its academic staff who participates in contributing to refereed journals. University F Research Forum aims at bringing the teaching scholars and students of the University to come close to each other and interact in research activities. The forum encourages inter-disciplinary research in different subject areas in the University. All faculty members and students are encouraged to participate and be involved in a healthy academic activity in order to take the University to new heights in teaching and research activities. Seminar programmes are held every month, each

faculty/school/institute has a set day for a one hour session dedicated to give a seminar by faculty member(s) and students. Research Week is organized once every year.

cccx

Concerning publications, Research articles presented by the academic community and staff in seminars and research week are published in a monthly Bulletin of Research Forum. Some articles will are selected for publication in reputed refereed international journals. An Annual Research Report outlining all the activities for the whole year are prepared and presented at the end of each year. Awards and incentives have been set up for academic staff who have their findings published in reputable journals and presented to international conferences and workshops. For example, where a researcher has published at least one article in a recognised reputed International Journal (University Journal included); the University will reward such a researcher with monetary reward of $200. Where a researcher has presented at least one paper in International Conferences or workshops, the University will reward such a researcher with monetary reward of $100 over and above financial support by the University to attend such conferences and workshops. Awards for Research Excellence will be given to the best research (individuals or groups) in the faculty/school/institute. The award will be in the form of financial grant to further the research activities of the group or individual. University J has made strides in developing research policy in promoting research work in the university. The University Council has established a University Research Fund

cccxi

(URF) and ensures that significant amount of funds are annually set aside for various research activities. The activities form part of core business activities of the University that complements teaching and gives University national and international recognition. The university council has also set up the Apportionment of Annual Research Grant (ARG) which is used for: (a) Training of Junior Researchers to be able to participate in

competitive research. The Junior Researchers will be exposed to research techniques and methodologies through attendance of workshops and seminars organized at the Universities. The professional researchers in various fields shall also have junior researchers as a requirement attached to them. This opportunity will prepare the Junior Researchers to participate in bidding for competitive project proposals. (b) Funding Competitive Research and/or Project Proposals Research

proposals will be considered bi-annually. (c) Awards. The University will reward the best publication arising

from research projects. The University will also reward innovators who have achieved excellent transfer of technology or knowledge and are not necessarily academic publications.

cccxii

(d) (e)

Conference attendance. Universitys Annual Research Conference. The University organizes

an annual research conference for the purpose of disseminating research findings. (f) Financing of Universitys Refereed Journal(s) will support journals

to publish and disseminate research findings resulting from researches in and outside the university.

The establishment and maintenance of research relationship with private sector/industry has been identified. University Js IPR Policy has also been developed to take into consideration the intellectual property rights of the academic staff. University B has developed a research productivity framework . This is a 3dimensional matrix. The first dimension consists of the Administrative, Financial and Human Resources roles and responsibilities that are necessary for the successful implementation of research. The second dimension consists of the research career progression methodology which will be applied to progressively develop a researcher from postgraduate studies up to research mentorship role at Professorial level. The third

cccxiii

dimension consists of the main organs of the University which have a responsibility to carry out the roles and activities required to develop the researcher. These organs consist of: Academic staff members The Department The Faculty, Institute or School The College The Office of Deputy Vice-Chancellor (Research and Development)

The matrix will thus document the policies that guide the organs of the University with the aim of increasing the research productivity of members of staff. The research productivity conceptual framework has the following guiding principles. Encouragement of the implementation of research projects that are relevant to Kenya and the world and that are carried out using sound methodology and honest reporting in a resource-efficient and ethical manner. Facilitation of research proposal drafting and sourcing of research funds.

cccxiv

Facilitation of the disbursement of research funds to the researcher in a transparent and efficient manner.

Provision of cost-effective research incentives and management and auditing of research activity without overburdening research funds.

Facilitation, automation and decentralization of research management, carrying out all research management activity at the lowest possible organ.

Establishment of common benchmarks for research activity, including methods for monitoring and evaluating research progress.

Maintenance of a sustainable research system by ensuring a steady, successful flow of postgraduate students into the research system of the University B.

The research policy aims to facilitate the development of research careers, assuming that researchers will become increasingly productive and useful for the fulfillment of the research objectives of the University B as they grow in seniority.

The table below classifies academics as postgraduate students, research trainees, researchers or research mentors for administrative purposes and outlines the qualifications,

cccxv

establishment positions, roles, research activities, research-to-teaching load and expected outputs of each category.

Table 4.52: Research Career Development Framework

Stage of

Postgraduate

Research

Researcher Ph.D.

Research Mentor Ph.D.

research student trainee Qualifications Bachelor Master s or Masters Positions degree Masters. or Role Ph.D. s. or Ph.D. Tutori al Fellow or Lecturer Carry out research project. other or

Lecturer Senior

and

Associate Full

student Qualify with expected degree. Assist in

Lecturer Professor Carry out Mentor all research. Supervise research activity. Recruit

cccxvi

Stage of research

Postgraduate student research.

Research trainee

Researcher researchers and students. Attract and administer research funding.

Research Mentor and supervise other researchers and students. Attract and administer large research

Research activity

Postgrad uate research project or thesis activity. al or

Doctor

Successfu lly implement funded research projects.

grants. Successfu lly administer large grants. Participat e in research administration in consultation with head of

postdoctoral research.

Research /

dept. (for research institutes the general load is 70% research / 30%

cccxvii

Stage of research teaching load

Postgraduate student teaching)

Research trainee

Researcher

Research Mentor

Academic departments shall observe the research and teaching load shown in the table below, that is, 30% Research / 70% other missionExpected Outputs (indicative) oriented activities Dissertati on, Thesis, Conference papers. 1 Thesis in 4 years 2 conference papers in 4 years s, conference, journal papers. 1 conference / journal paper per annum Thesi Reports, conference and journal papers, books. 1 journal paper or 2 conference papers per annum 1 patent Reports, conference and journal papers, books. 2 journal papers or 3 conference papers per annum or 1 book every four years. Source University B Research Policy

cccxviii

Perhaps this is the most comprehensive research policy developed in Kenya that is geared towards encouraging research productivity among academic staff. University H has research capacity building funds, which may be used by the University departments to develop the capacity of their staff to conduct research. The funds may be employed for organizing short courses on research methods. The amount available for each year is Kshs 100,000/- This amount is quite low and may not benefit many of the academic staff who would be engaged in career development. The university has organised research forums for engaging staff and graduate students in research methods. This is an important forum where an exchange on information gained is shared with junior researchers and graduate students.

4.7.1 Summary of Document Analysis Most of the research policies covered here had made omissions on several issues that are deemed to promote research productivity. Funding for research was an item that was clearly outlined and mechanisms for the same spelt out clearly. Issues like career development were highlighted by few universities. There is need therefore for the documents to be reviewed with the aim of refocusing them to research productivity of their academic staff.

cccxix

CHAPTER FIVE SUMMARY, CONCLUSIONS AND RECOMMENDATIONS

5.1 Summary The current academic climate in higher education in Kenya threatens the Kenyan universities ability to sustain the conditions that support research productivity. Increased demands on government and private funding, a deteriorating physical infrastructure, increased pressure on undergraduate programs, university expansion strategies and general economic climate in the country have raised concerns about the continued capacity of universities to maintain teaching, research productivity and service to the community. This calls for regular update on the level of research productivity in the country. There are very few studies done in Kenya to analyze the factors that influence research productivity in institutions of higher learning. This study therefore, examined the factors that influence research productivity among academic staff in selected public and private universities in Kenya. The themes discussed included the status of research output in Kenya between 2004-2008, nature of publishing infrastructure, individual and institutional factors influencing research productivity, the attitude of university lecturers in Kenya in regard to research and publishing and lecturers and heads of departments views

cccxx

of problems and possible solutions to addressing the same problems. The study finishes with a discussion of some of the strategies that can be used to enhance research productivity among academic staff in Kenyan Universities. Literature review was done from wide sources that had explored studies related to research productivity. Areas covered under this included, research productivity and related theories, research productivity and type of institution, factors influencing research productivity, technology transfer and research productivity, the teaching nexus, research productivity and publication and a summary of all reviewed literature was done. A survey research design and document analysis were used in this study. The target population was the university academic staff in selected 5 private and 6 public universities in Kenya. These formed 11 universities in total. The sample was selected by using simple random sampling. The sample size was 277 (70.2% male and 29.8% female) for academic staff and 17 for the heads of departments. Data were collected using three instruments namely the questionnaire for lecturers and the other one for heads of departments. The questionnaires were self administered to the academic staff. Three questionnaires were filled and sent to the researcher through email. Data collected were analysed using quantitative procedures. It was coded and analysed into frequencies and percentages with the help of SPSS for windows (version 15). The

cccxxi

SPSS tool was employed in calculation of the means, standard deviation, frequencies and percentages and the generation of the ANOVA tables. It also helped to compute the principal components using the Factor Analysis method. Lastly the summary, conclusions and recommendations of the study were done. There were five alternative hypotheses that were tested using one way ANOVA. There is a significant relationship between gender and research productivity, there is a significant relationship between age groups and research productivity, there is a significant relationship between academic rank and research productivity, there is a significant relationship between highest degree obtained and research productivity, there is a significant relationship between years since last highest degree was obtained and research productivity. The PCA used in this study revealed that there were three main factors that influence research productivity of Kenyan academicians. The first component was

interpreted as personal career development which explained the highest variance of 29% in the data. The second component, contributing 20%, could be interpreted as institutional factors whereas the third factor, contributing 14% could be interpreted as demographic or experience of the individual researcher to publish. Component one and three are closely related in this analysis. This may mean that it is the individual self determination, commitment, motivation and stamina that do count in establishing whether an individual

cccxxii

researcher is able to publish or not. The second component, institutional factors, explained 20% of the variance in the data and involves all those resources, equipment or rewards that are supposed to be supplied by respective institutions. Other findings indicated that the research productivity of Kenya was below par, the output among academic staff was skewed, and that there was differing views on academic staffs attitude towards research and publishing. Document analysis dwelt on the analysis of the sampled research policies of the selected universities. Information from these documents revealed that most universities were not keen in the area of staff professional development. Main recommendation of this study is for government and individual institutions to try and come up research policies that can guide the research process in this country by addressing the pertinent issues raised by academic staff. The issues raised here include individual and institutional factors. And they should be addressed together. It is only by addressing the raised issues that the country can think of making strides in increasing research productivity of this country.

5.2 Conclusions

cccxxiii

On the basis of the findings of this study, the following are some of the conclusions that can be drawn.
1.

The nature of research productivity among public and private universities in Kenya,

between 2004 2008 is below par when compared with those of other countries of the world.
2.

The research productivity produced indicated skewness in production. The older

universities had more production in terms of publications compared to the younger upcoming universities. The public universities seem to be better in terms of research output than the private universities. It also emerged that the status of research productivity in Kenya is still low compared to the developed nations. 3. There are differences in publication output between different age groups, ranks and

highest degree obtained. Publication output is high among the senior members of the academic staff. The senior members of staff have more publications than the younger members of staff. This is also reflected in the position held in the university and the higher qualification s held by the individual lecturers. It was evident that many of the academic staff below the rank of lecturer from the majority of those who did not publish anything between 2004-2008. Majority of the publications were done by professors.

cccxxiv

4.

Data collected also revealed that there was no much difference between the

research output produced by academic staff when categorized by gender, type of university, faculty and years since last highest degree was obtained. 5. It was established that a combination of individual and institutional factors acted to

influence the research productivity of academic staff in Kenya. Research content knowledge and self motivation were the key factors among individual factors that had the greatest influence on researchers. On institutional factors it was resources for research, equipments and availability of technology based equipments like internet and computers. 6. Academic staffs attitude toward research and publication is significant with age

groups, highest degree obtained, years since last highest degree and by name of the university. However the academic staffs attitude to research and productivity is not significant with gender, rank and type of university. 7. Academic staff and heads of department are in agreement on possible solutions to

problems affecting research productivity in Kenya. Some of these problems are inadequate research fund, heavy teaching load, inadequate research motivation, poor

cccxxv

salaries, inadequate research knowledge among some academic staff, lack of institutional support, poor attitude toward research and lack of appropriate journals. 8. The possible solutions they offer here include, allocation of more research funds,

employment of more academic staff, reward for researchers, purchase of modern equipments for research, revision of salaries for academic staff, engagement of the private sector in research, introduction of research refresher courses for lecturers, development of appropriate research policies, pegging promotion on administration on publishing. The heads of departments had almost the same views on this. One issue that comes out, though, is that the press for good salaries comes a distant sixth position.
9.

Document analysis on the analysis of the sampled research policies of the selected

universities revealed that most universities had not made sufficient provision for staff professional development. 5.3 Recommendations It is evident from this study that indeed there are challenges to be overcome if at all research productivity are to be realized in the higher institutions of this country. There is a need for the stakeholders in this area to have an all inclusive dialogue to address the pertinent issues raised by the academic staff.

cccxxvi

This calls for the development of a national research policy that shall cascade down to the institutional level.
1. Review of universities research policies that shall lead to;

Providing adequate research funds for researchers Staff professional career development Providing research training courses to raise research productivity and for publishing results worldwide.

Developing overall research management systems. Providing and establishing effective research assessment systems. Supporting and encouraging private organizations to participate in the universitys research by establishing systems which increase the opportunities for the university and private organizations to work collaboratively.

2. Engagement of the private sector in supporting research in the so called private

public sector partnership (PPPs). KEPSA (Kenya Private Sector Alliance) is functional and may need to be approached on this.

cccxxvii

3. Provide an effective publication system to promote the research productivity of

academic staff and graduate students. Again this is in agreement with the development of the research policies that will guide the entire research process in each university and the country at large.
4. Create avenues for publication and dissemination of work which is lying in

university shelves. Most of the theses done by undergraduate and graduate students are lying in the university shelves. It is high time that these materials were published so that they advance the mission of the universities for dissemination of knowledge.
5. Encouraging collaboration between universities and research institutions by

providing an information system that is able to show availability and access to utilize research facilities and data. The collaboration can extend to institutions outside the country. The linking of professionals in particular fields will promote scholarship at the university level.
6. Creation of a fund for the publication of tertiary books and journals; provision of

special funding for and revitalization of university publishing houses. In the modern

cccxxviii

world the idea of popularizing the electronic journals in the universities will be a great idea.
7. Fostering of a climate of reward for academic excellence particularly for

international publishing. The University should also consider the possibility of increasing a researchers salary and rewards, ensuring that the income derived from doing research is equal to or higher than income derived from teaching.
8. Rewards for staff who engage with research. Appropriate rewards can be

promotion, recognition, and money, providing a difference in terms of rewards between staff with research productivity and those without, in order to encourage research motivation. In addition, the University might consider a kind of special leave for the development of research, allowing teachers to take leave for one semester to be completely free from their workload. 5.4 Suggestions for further studies
1. The current study explored the most frequently used measure of the quantity or

amount of research productivity as a numerical publication count of number of research articles. There is need therefore for future studies to examine, quality,

cccxxix

peer review rating and citation analysis as new tools to assess the value of the contributions of research to the discipline.
2. There is need also to undertake research productivity analysis at the university,

faculty or departmental level.

3. There is a need for research that analyzes and describes the efforts currently used

by institutions to foster their academic staffs productivity levels. 4. Need for research to find out why individual professional development is a major motivator for research productivity.
5. There is need to find out how the universities have implemented their research

policies especially in regard to individual professional development. 6. The current study focused on universities. There is need to carry out a study on the factors influencing research productivity in other knowledge producing institutions e.g. National research institutes.

cccxxx

REFERENCES

AAU. (1987). Harare Statement on the Role of African Institutions of Higher Learning in Africa's Recovery and Development. Accra: Association of African Universities. Abagi, O. (1999). "Revitalizing Financing of Higher Education in Kenya: Resource Utilization in Public Universities", IPAR Policy Brief, IPAR, Nairobi. Adams, J. (1965). Inequity in Social Exchange Journal of Abnormal and Social Psychology, vol. 62, pp. 335-343. Adams, J. D. & Griliches, Z. (1996). Research Productivity in a System of Universities. NBER Working Paper No. W5833. Available at SSRN:

http://ssrn.com/abstract=225619. Adams, J. A. & Bonk, S. C. (1995). Electronic information technologies and resources: use by university faculty and faculty preferences for related library services. College and Research Libraries, Vol.56: 119-131. Ali, N. S., Young, H. C. & Ali, N. M. (1996). Determining the quality of publications and research for tenure or promotion decisions: A preliminary checklist to assist. Library Review, 45(1): 39-53.

cccxxxi

Allison, P. D. & Stewart, J. A. (1974). Productivity among Scientists: Evidence for Accumulative Advantage. American Sociological Review, Vol.39: 596-606. Almind, T. C & Ingwersen, P. (1997). Methodological issues of webometrics studies, handbook of quantitative science and technology research. Pages 339-369 Springer. Netherlands. Antony, J. S., & Raveling, J. S. (1998). A Comparative Analysis of Tenure and Faculty Productivity: Moving Beyond Traditional Approaches. Paper Presented at the Meeting of the Association for the Study of Higher Education, Miami, FL. Atwoli, L., (2008). Mushrooming Campuses Alarming. East African Standard November 24, 2008. Applebee, A. C., Clayton, P. & Pascoe, C. (1997). Australian academic use of the Internet. Internet Research, Vol. 7, no. 2: 85-94. Babber, S. et al. (2000). An empirical assessment of institutional and individual research productivity in international operations management. Internationa Journal of Operations and Production Management 20(12): 1392-1410. Babu, A. R & Singh, Y. P. (1998). Determinants of Research Productivity. Scientometrics. Volume 43, Number 3 / November, 1998.

cccxxxii

Baden-Fuller, C. F., Ravazzalo, & Schweizer, T. (2000). Making and Measuring Reputations: the Research Ranking of European Business Schools, Long Range Planning 33, no. 5, 621650. Bailey, T. G. (1992). Faculty Research Productivity. Minneapolis, MN: Association for the Study of Higher Education Annual Meeting. (ERIC Document Reproduction Service No. ED352895 ). Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behaviour change, Psychological Review, vol. 84, pp. 191-215. Bandura, A. (1986). Social foundations of thought and action: a social cognitive theory. New Jersey: Prentice-Hall Inc. Bandura, A. (1991). Self foundations of thought and action: a social cognitive theory. Englewood Cliffs, New Jersey: Prentice-Hall Inc. Bandura, A. (1997). Self-efficacy: toward a unifying theory of behavioral change, Psychological Review, vol. 84, no. 2, pp. 191-215. Bandura, A. & Locke, E. (2003). Negative self-efficacy and goal effects revised, Journal of Applied Psychology, vol. 88, pp. 87-89.

cccxxxiii

Bandura, A, Reese, L & Adams, N. (1982). Microanalysis of action and fear arousal as a function of differential level of perceived self-efficacy, Journal of Personality and Social Psychology, vol. 43, pp. 5-21. Barhyte, D.Y & Redman, B. K. (1993). Factors Related to Graduate Nursing Faculty Scholarly Productivity. Nursing Research [Nurs Res] 1993 May-Jun; Vol. 42 (3), pp. 179-83. Barry, J., Chandler, J. & Clark, H. (2001). 'Between the Ivory Tower and the academic assembly line', Journal of Management Studies 38(1): 87101. Bean, J. P. (1982). A Causal Model of Faculty Research Productivity. A Paper Presented at the Annual Meeting of the American Educational Research Association, New York. Beehr, T., Walsh, J. & Taber, T. (1976). Relationship of stress to individually and organizationally valued states: higher order needs as a moderator, Journal of Applied Psychology, vol. 61, no. 1, pp. 41-47. Biglan, A. (1973). Relationships between Subjects Matter Characteristics and the Structure and Output of University Departments. Journal of Applied Psychology, 57, 204213.

cccxxxiv

Blackburn, R. T., Behymer, C. E. & Hal, D. E. (1978). Research Notes: Correlates of Faculty Publications. Sociology of Education, Vol. 51: 132-141. Blackburn, R., Bieber, J., Lawrence, J., & Trautvetter, L. (1991). Faculty at work: Focus on research, scholarship, and service. Research in Higher Education, 32(4), 385. Blackburn, T. & Lawrence, H. (1995). Faculty at work: motivation, expectation, satisfaction, London: The Johns Hopkins University Press. Bland, C., Seaquist, E., Pacula, J., Center, B. & Finstad, D. (2002). One Schools Strategy to Assess and Improve the Vitality of its Faculty. Academic Medicine. Vol. 77,pp. 368376. Bland, Carole J B.; Center, Bruce A.; Finstad, Deborah A.; Risbey, Kelly R.; Staples, T & Justin G. (2005). A Theoretical, Practical, Predictive Model of Faculty and Department Research Productivity. Academic Medicine: March 2005 - Volume 80 Issue 3 - pp 225-237. Bland, C., & Berquist, W. (1997). The vitality of senior faculty members. Snow on the rooffire in the furnace. (ERIC Document Reproduction No. ED 415733). Blume, S. S. & Sinclair, R. (1973). Chemists in British universities: a Study of the Reward System in Science. American Sociological Review, Vol. 38: 126-138.

cccxxxv

Boice, R. (1987). Habits of research productivity, Accounting Education News, Late Fall, p. 7. Bonnet, X., Shine, R. & Lourdais, O. (2004). Does Gender Affect a Scientist's Research Output in Evolutionary Ecology? Journal of Women and Minorities in Science and Engineering4 (2004): 353-360. Bonzi, S. (1992). Trends in research productivity among senior faculty. Information Processing and Management. Volume 28, Issue 1, 1992, Pages 111-120 Published by Elsevier Science Ltd. Borokhovich, K. A., Bricker, R. J., Brunarski, K. R. & Simkins, B. J. (1995). Finance Research Productivity and Influence. Journal of finance, Vol 50 No 5, December 1995. Available at SSRN: http://ssrn.com/abstract=6598 Bottle, et al. (1994). Citation analysis and the quality of scientific productivity. Bioscience, Vol. 27, no. 1: 26-31. Boyer, E. L. (1990). Scholarship Reconsidered: Priorities of the Professoriate. Princeton, NJ: Carnegie Foundation for the Advancement of Teaching. Bozeman, B. (2005). The Impact of Research Collaboration on Scientific Productivity. Social Studies of Science, Vol. 35, No. 5, 673-702.

cccxxxvi

Braun, T.; Glanzeland, W. & Schubert, A. (1990). Publication productivity: from frequency distribution to scientometric indicators. Journal of Information Science, Vol.16: 3744. Brewer, G. A., Douglas, J. W, Facer, R. L., & O'Toole, L. J. (1999). Determinants of Graduate Research Productivity in Doctoral Programs of Public Administration Public Administration Review. Vol. 59, 1999. Brocato, J. J., & Mavis, B. (2005). The Research Productivity of Faculty in Family Medicine Departments at U.S. Medical Schools: a National Study. Academic Medicine: Journal of The Association of American Medical Colleges [Acad Med] 2005 Mar; Vol. 80 (3), pp. 244-52. Brooks, R. L. (2006). Analyzing Faculty Scholarly Activity across Disciplines: Individual and Structural Influences on Research Processes and Products. Manuscript Submitted for Publication. Buchheit, S., Collins, A. B., & Collins, D. L. (2001). Intra-institutional factors that influence accounting research productivity. The Journal of Applied Business Research, 17(2), 1731.

cccxxxvii

Budd, H. M. (1995). Faculty publishing productivity: an institutional analysis and comparison with library and other measures. College and Research Libraries, Vol. 56, no.6: 547-554. Burke, K., Fender, B. & Taylor, S. (2007). Walking the Tightrope: The Impact of Teaching and Community service on Scholarly Productivity for Accountants, Academic Business World International Conference, Nashville, Tennessee, USA, May 2830, pp. 113. Business Dictionary (2009). Motivation definition. Retrieved June.10th 2009 from http://www.businessdictionary.com/definition/motivation.html. Campion, P. & Shrum, W. (2004). Gender and Science in Development: Women Scientists in Ghana, Kenya, and India Science, Technology and Human Values, Vol. 29, No. 4, 459-485. Cargile, B. & Bublitz, B. (1986). Factors contributing to published research by accounting faculties, The Accounting Review, January, pp.158-178. Carnegie Foundation. (1991). The payoff for publication leaders, Change, March/April, pp. 27-30. Centra, J. (2005). Research Productivity and Teaching Effectiveness. In Research in Higher Education. Volume 18, Number Pages 379-389 Springer. Netherlands.

cccxxxviii

Cepero, M. C. (2007). Institutional and Individual Factors Associated with Faculty Scholarly Productivity. Unpublished PhDs Thesis. University of Connecticut. CHE (2009). Commission for Higher Education. Website www.che.go.ke Chege, M. (2006). The State Of Higher Education In Kenya: Problems And Prospects Paper Presented At the Mijadala on Social Policy, Governance and Development in Kenya Sponsored By Development Policy Management Forum on 22 June, 2006 at Nairobi Safari Club. Chen, Y., Gupta, A., & Hoshower, L. (2004). Faculty Attitudes of Research Rewards. Journal of College Teaching and Learning December 2004 Volume 1, Number 12. Chen, Y., Gupta, A. & Hoshower, L. (2006). Factors that Motivate Business Faculty to Conduct Research: An Expectancy Theory Analysis, Journal of Education for Business, Vol. 81, No.4, pp. 179189. Child, D. (2006). The essentials of factor Analysis. 3rd edition. London: Continuum International publishing Group. Clement, R & Stevens, G. (1989). Performance appraisal in higher education: Comparing department of management with other business units, Public Personal Management, vol. 18, pp. 263-278.

cccxxxix

Cohen, J. (1996). Computer mediated communication and publication productivity among faculty. Internet Research, Vol.6, no. 2/3:41-63. Cole, S. (1979). Age and Scientific Performance. The American Journal of Sociology, 84 (4), 958-977. Coolican. (1996). Introduction to Research Methods and Statistics in Psychology (Second Edition). London: Hodder and Houghton. Creswell, J. (1985). Faculty Research Performance: Lessons from the Sciences and Social Sciences, Washington DC: Association for the Study of Higher Education. Creswell, J. (1986). Measuring Faculty Research Performance, San Francisco:Jossey-Bass Inc., Creswell, J. W. & Plano Clark, V. L. (2006). Designing and Conducting Mixed Methods Research. Thousand Oaks, CA: Sage Publications. Dakik, H. A, Kaidbey, H, & Sabra, R. (2005). Research Productivity of the Medical Faculty at the American University of Beirut. Postgraduate Medical Journal 2006; 82:462-464. David, H. G.; Piip, L. & Haly. A. R. (1981). The examination of Research Trends by Analysis of Publication Numbers. Journal of Information Science, Vol. 3: 283-288.

cccxl

Diamond, A. M. (1986). The life-cycle research productivity of mathematicians and scientists. Journal of Gerontology, 41(4), 520525. Deci, E., & Ryan, R. (1991). Motivational Approach to Self: Integration in Personality. R. Dienstbier (Ed.), Nebraska Symposium on Motivation: Perspectives on Motivation (Vol. 38). Lincoln: University of Nebraska Press. Diamond, A. M. (1986). The life-cycle research productivity of mathematicians and scientists. Journal of Gerontology, 41(4), 520525. Dickson, V. A. (1983). The determinants of publication rates of faculty members at a Canadian university. Canadian Journal of Higher Education, 13 (2): 4149. Donald, S. S, Paul, W. & Mike (2003). Assessing the Impact of University Science Parks on Research Productivity: Exploratory Firm-Level Evidence from the United Kingdom. In International Journal of Industrial Organization Volume 21, Issue 9, November 2003, Pages 1357-1369. Doyle, W. (2006). State accountability policies and Boyers domains of scholarship: Conflict or collaboration? New Directions for Institutional Research, 129, 97-113. Dundar, H. & Lewis, D. R. (1998). Determinants of Research Productivity in Higher Education. Research in Higher Education, 39, 607-630.

cccxli

Dyrbye, L. N., Davidson, L. W. & Cook, D. A. (2008). Publications and Presentations Resulting from Required Research by Students at Mayo Medical School, 19762003. Journal of the Association of American Medical Colleges [Acad Med] 2008 Jun; Vol. 83 (6), pp. 604-10. Ederington, L. H., (1979). Aspects of Production of Significant Financial Research, Journal of Finance 34 (1979), 777786. Education Encyclopedia (2008). Faculty Roles and Responsibilities - The Teaching Role, the Research Role, the Community service Role, Integration of Faculty Roles and Responsibilities: Education Encyclopedia - StateUniversity.com. Fielden, J. & Gibbons, J. (1991). Merit myopia and business school faculty Publications, Business Horizons, vol. 34, no. 2, pp. 8-12. Finkelstein, M. (1984). The American Academic Profession: A synthesis of Social Scientific Inquiry Since World War II, State University Press, Columbus, Ohio Fox, M. K. (2005). Gender, Family Characteristics, and Publication Productivity among Scientists. Social Studies of Science, Vol. 35, No. 1, 131-150 (2005) DOI: 0. 1177/ 030 63127 05046630 Georgia Institute of Technology.

cccxlii

Fulton, O. & Trow, M. (1974). Research activities in American higher education, Sociology of Education, vol. 47, Winter, pp. 29-73. Gaston, J. (1970). The reward system in British Science, American Sociological Review, vol. 35, pp. 718-732. Geiger, R. (1986). To Advance Knowledge: The Growth of American Research universities: 1900-1940, Oxford University Press, New York. Gibbs, P. & Locke, B. (1989). Tenure and promotion in accredited graduate social Work program, Journal of Social Work Education, vol. 25, no. 2, pp.126-133. Gibbons, M. (1998), Higher Education Relevance in the 21st Century, Association of Commonwealth Universities. Gist, M. & Mitchell, T. (1992), Self-efficacy: a theoretical analysis of its determinants and malleability, Academy of Management Review, vol. 17,pp. 183-211. Gonzalez-Brambila & Francisco Velosob. (2003). The Determinants of Research Productivity: Department of Engineering and Public Policy, 5000 Forbes Avenue, 129 Baker Hall, Pittsburgh: PA Carnegie Mellon University. Goodwin, L., Kozleski, E., Muth, R., Rhodes, L. K., & White, K. K. (2006). Establishing a center to support faculty research. Innovative Higher Education, 30, 251-268.

cccxliii

Gorman, A. M., & Scruggs, M. M. (1984). Characteristics of home economics researchers, Home Economics Research Journal, 12(4), 461-69. Government of Kenya. (1981). The Report of the Presidential Working Party on the Second University in Kenya (The Mackay Report, 1981). Government of Kenya. (2005). Sessional Paper No. 1 of 2005 on a Policy Framework for Education, Training and Research: Meeting Challenges of Education, Training and Research in Kenya in the 21st Century, Nairobi: Government Printer. Government of Kenya. (2007). The National Strategy for University Education 2007-2015: Investing in The Future of University Education. Unpublished Document. Government of Kenya. (2008). Kenya National Bureau of Statistics: facts and Figures. Government Printer, Nairobi: Government Printer. Government of Kenya. (2008). Vision 2030. First Medium Term Plan. A Globally Competitive and Prosperous Kenya 2008 2012. Nairobi: Government Printer. Government of Kenya. (2009). Africa union Scientific Awards, Call for Proposals. Ministry of Higher Education, Science and Technology. www.scienceandtechnology.go.ke. Government of South Africa. (2003). Policy and Procedures for Measurement of Research Output of Public Higher Education Institutions. Ministry of Education http://www.education.gov.za

cccxliv

Government of South Africa. (2001). National Plan for Higher Education. Ministry of Education http://www.education.gov.za. Grover, V., Segars, A. H., & S. Simon, (1992). An Assessment of Institutional Research Productivity in MIS, Database 23 (1992), no. 4, 59. Grunig, S. D. (1997). Research, Reputation, and Resources: The Effect of Research Activity on Attitude of Undergraduate Education and Institutional Resource Acquisition. Research in Higher Education, 68, 17-52. Grover, V. et al. (2002). An Assessment of Institutional Research Productivity in MIS, Data Base 23(4), 5-8. Gu, Y. & A. N. Zainab. (2001). Publication productivity of Malaysian researchers in the field of computer science and information technology. Malaysian Journal of Library and Information Science, Vol.6, No.1: 1-23. Gupta, D. A. (1987). Lotka's Law and Productivity Patterns of Entomological Research in Nigeria for the Period, 1900-1973. Scientometrics, 12(1-2), 33-46. Hadjinicola, G. C., & Soteriou, A. C. (2006). Factors Affecting Research Productivity of Production and Operations Management Groups: An Empirical Study. Journal of Applied Mathematics and Decision Sciences Volume 2006, Article ID 96542.

cccxlv

Hannington, M. S., & Levine, D. U. (1986). Relationship between Faculty Characteristics and Research Productivity. Journal of Dental Education. 1986 Sep; Vol. 50 (9), pp. 518-25. Hassan, A., Tymms, P. & Ismail, H. (2008). Academic Productivity as Perceived by Malaysian Academics. Journal of Higher Education Policy and Management, Volume 30, Issue 3, pages 283 296. Hasselbacka, J. R, Reinstein, A. B., & Schwanc, E. S. (2000). Benchmarks for Evaluating the Research Productivity of Accounting Faculty. In Journal of Accounting Education Volume 18, Issue 2, spring 2000, Pages 79-97. Hemlin, S., & Gustafsson, M. (1996). Research Production in the Arts and Humanities: A Questionnaire Study of Factors Influencing Research Performance. Scientometrics. ISSN 0138-9130 Vol. 37, no.3, pp. 417-432. Ho, K. K. (1998). Research Output among the Three Faculties of Business, Education, Humanities and Social Sciences in Six Hong Kong Universities. Higher Education Vol. 36: 195 208. Holcomb, J. D., Thomson, W. A. & Bartel, R. E. (1988). An Analysis of the Scholarly Activities of Respiratory Care Faculty in Southern Academic Health Centers. Respiratory Care [Respir Care] 1988 Dec; Vol. 33 (12), pp. 1108-13.

cccxlvi

Horton, R. (1998). Publication and promotion. A Fair Reward. Lancet 352: 892 Hu, Q., & Gill, T. G. (2000). Is faculty research productivity: Influential factors and implications. Information Resources. Management Journal, 13(2), 1525. Jacobs F. A., Hartgraves, A. L., & Beard, L. H., (1986). Publication Productivity of Doctoral Alumni. The Accounting Review 61 (1986), 179186. Jacobsen, R. L. (1992). Colleges face new pressure to increase faculty productivity, Chronicle of Higher Education, 38(32), 1, 16-18. Jean, O. L. & Schankerman, M. (2004). Patent Quality and Research Productivity: Measuring Innovation with Multiple Indicators. The Economic Journal: 441465AD. Jone, J. E., Preusz, C., & Finkelstein, S. N. (1989). Factors Associated with Clinical Dental Faculty Research Productivity. Journal of Dental Education [J Dent Educ] 1989 Nov; Vol. 53 (11), pp. 638-45. Jones, J .E. (1998). Gender and research productivity in US and Canadian schools of dentistry. A preliminary investigation. European Journal of Dental Education: Vol. 2 (1), pp. 42-5.

cccxlvii

Joughin, L, (ed.) (1969). Academic Freedom and Tenure: A Handbook of the American Association of University Professors. Madison: University of Wisconsin Press (pp. 163164). Kelly, M. E., & Warmbrod, J. R. (1986). Developing and Maintaining Productive Researchers in Agricultural Education, Journal of the American Association of Teacher Educators in Agriculture, 27(1), 27-32. Kelchtermans, S. & Veugelers, R. (2005). Top Research Productivity and its Persistence. Centre for Economic Policy Research (CEPR) Brussel: Catholic University of Leuven (KUL). Kennedy, D. (1997). Academic Duty. Cambridge, MA: Harvard University Press. Kenneth, A. F. (1987). Research Productivity and Scholarly Accomplishment of College Teachers as Related to their Instructional Effectiveness: A Review and Exploration. In Research in Higher Education Volume 26, Number 3 / September, 1987 Pages 227-331 Springer Netherlands. Klemkosky, R. C. & Tuttle, D. L. (1977). The Institutional Source and Concentration of Financial Research, Journal of Finance 32, 901907. Kohlenberg, E. M. (1992). Faculty Research Productivity and Organizational Structure in Schools of Nursing. Journal Of Professional Nursing: Official Journal Of The

cccxlviii

American Association Of Colleges Of Nursing [J Prof Nurs] 1992 Sep-Oct; Vol. 8 (5), pp. 271-5. Kostoff, R. (1995). Federal research impact assessment: axioms, approaches Applications, Scientometrics, vol. 34, no. 2, pp.163-206. Kotrlik, J. W, Bartlett, J. E, Higgins, C. & Williams, H. A, (2002). With Research Productivity of Agricultural Education Faculty. Journal of Agricultural Education 9 Volume 43, Number 3, 2002. Krugman, P. (1991). Geography and Trade, Cambridge: IT Press. Kuh, G. D. & Whitt, E. J. (1988). The Invisible Tapestry: Culture in American Colleges and Universities. Higher Education Reports. Washington D.C. Kyvik, S. (1990), Motherhood and Scientific Productivity. Social Studies of Science, 20(1), 149-160. Laband, D. N., (1985). An Evaluation of 50 Ranked Economics Departments-by Quantity and Quality of Faculty Publications and Graduate Student Placement and Research Success, Southern Economic Journal 52 (1985), 216240. Landino, R. & Owen, S. (1998). Self-efficacy in university faculty, Journal of Vocational Behaviour, vol. 33, pp. 1-14.

cccxlix

Lange, L. (2001). Citation counts of multi-authored papers First-named authors and further authors. Scientometrics, Vol. 52, No. 3 (2001): 457470. Lazinger, S. S., Barllan, J. & Peritz, B. C. (1997). Internet use byfaculty members in various disciplines: a comparative case study. Journal ofthe American Society for Information Science, Vol. 48, no.6: 508-518. Lertputtarak, S. (2008). An Investigation of Factors Related to Research Productivity in a Public University in Thailand: A Case Study. A PhD dissertation, School of Education, Faculty of Arts, Education and Human Development, Victoria University, Melbourne, Australia. Liu, N. C. & Cheng, Y. (2005). Academic ranking of world universities - methodologies and problems. Higher Education in Europe, Vol. 30, No. 2. Lee, S. & Bozeman, B. (2004). The Impact of Research Collaboration on Scientific Productivity. Manuscript under review at Social Studies of Science. Long, J. S., Allison, P. D. & McGinnis, R. (1993) Rank Advancement in Academic Careers: Sex Differences and the Effects of Productivity. American Sociological Review Vol. 58, No. 5 (Oct., 1993), pp. 703-722. Lotka, A. J. (1926). The Frequency Distribution of Scientific Productivity. Journal of the Washington Academy of Sciences, Vol.16, no.12: 317-322.

cccl

Malhotra, K. & Kher, H. (1996). Institutional research productivity in production and operations management, Journal of Operations Management, 14(1), 55-77. Marchiori, D. M., Meeker, W., Hawk, C. & Long, C. R. (1998). Research productivity of chiropractic college faculty Journal of Manipulative And Physiological Therapeutics. 1998 Jan; Vol. 21 (1), pp. 8-13. Martin, B. R. (1996). The Use of Multiple Indicators in the Assessment of Basic Research. Scientometrics, Vol.36, no.3: 343-362. Maske, K. L., Durden, G. C., & Gaynor, P. E. (2003). Determinants of scholarly productivity among male and female economists. Economic Inquiry, 41(4), 555564. McGrail, M. R., Rickard, C. M., & Jones, R. (2006). Publish or perish: A systematic review of interventions to increase academic publication rates. Higher Education Research and Development, 25, 19-35. Milgrom, P., Heima, M., Tomar, S., Kunzel, C. (2008). Research Productivity of members of IADR Behavioral Sciences and Health services Research Group: Relationship to Professional and Personal Factors. Journal of Dental Education. 2008 Oct; Vol. 72 (10), pp. 1142-8.

cccli

Mills, O. F, Zyzanski,

S. J; Flocke, S. (1995). Factors Associated with research

productivity in Family Practice Residencies. Family Medicine. 1995 Mar; Vol. 27 (3), pp. 188-93. Mowday, R., & Nam, S. (1997). Expectancy Theory Approaches to Faculty motivation. In J. Bess (Ed.), Teaching well and liking it (n.p.). Baltimore, MD: John Hopkins Press. Muchie, M. (2008). Research Varsities Crucial to Development. East Africa Standard September 12, 2008. Mwamwenda, T. S. (1994). Academics stance on the slogan Publish or Perish In Onsongo, J. (2005). The Role of Research and Publication in the Promotion of Academics in Kenyan Universities. In Eastern Africa Journal of Humanities and Sciences. Vol. 5 No. 2 Catholic University of Eastern Africa, Nairobi. Mwiria, K. (2007). Public and Private Universities in Kenya. New Challenges, Issues and Achievements. Nairobi: EAEP. Nadler, D., & Lawler, E. (1977). Motivating Individuals in Organizational Settings. In J.Hackman, E. Lawler and L. Porter (Eds.), Perspectives on Behavior in Organizations (n.p.). New York:McGraw-Hill.

ccclii

Nachmias, C. F & Nachmias, D. (2005). Research Methods in the Social Sciences, 5th Edition. London: Arnold. Nafukho, F. M. (1999). Role of Universities in Promoting Cooperation between Social and Pure Scientists for Human Development in Kenya. The Journal of Third World Studies, Spring, 1999. Nahavandi, A. (1997). The Art and Science of Leadership. Upper Saddle River, New Jersey: Prentice Hall. Nahrwold, D. L., Pereira, S. G. & Dupuis, J. (1995). United States Research Published in Major Surgical Journals is Decreasing. Ann Surg 1995; 2 22: 263-266 Nelkin, D. (1998). Publication and Promotion. The Performance of Science. Lancet; 352: 893-894. NCST. (2009). Call for Proposals. National Council for Science and Technology. www.ncst.go.ke Ngome, C. (2003). "Kenya." Country Higher Education Profiles. In D. Teferra and P. G. Altbach (Eds.), African higher education: An international reference handbook. 359-371. Bloomington: Indiana University Press. Niemi, A. W. (1987). Institutional Contributions to the Leading Finance Journals, 1975 through 1986, a note: Journal of Finance 42 (1987), 13891397.

cccliii

Niemi, A. W. (1988). Publication Performance of Marketing Departments: 19751985, Journal of Marketing Education 10 (1988), 812. Niemi, A. W. (1998). Research productivity of American business schools, 197585, Review of Business and Economic Research 23 (1998), 117. Nwagwu, W. (2006). A Bibliometric Analysis of Productivity Patterns of Biomedical Authors of Nigeria During 1967-2002. Scientometrics, Vol. 69, No. 2 (2006), 259 269. Nyaigotti-Chacha. (2004). Reforming Higher Education in Kenya Challenges, Lessons and Opportunities: State University of New York Workshop with the Parliamentary Committee on Education, Science and Technology. Naivasha, Kenya. Obure, M. J. (2002). Handbook on Data Analysis Using SPSS version 10. M and O data Experts Training and Consultants, Nairobi. Ogbogu, C. O. (2009). An Analysis of Female Research Productivity in Nigerian Universities. Journal of Higher Education Policy and Management, Volume 31, Issue 1 February 2009, pages 17 22. Olson, J. E. (1994). Institutional and technical constraints on faculty gross productivity in American doctoral universities. Research in Higher Education 35(5): 549567.

cccliv

Onsongo, J. (2005). The Role of Research and Publication in the Promotion of Academics in Kenyan Universities. Eastern Africa Journal of Humanities and Sciences. Vol. 5 No. 2. Nairobi: Catholic University of Eastern Africa press. Oppenheim, & Ellerslie, C. S. (2008). The Effect of Motivation on Publication Productivity of UK LIS Academics. Library and Information Research Volume 32 Number 101 2008. Orodho, J. A. (2004). Techniques of Writing Research Proposals and Reports in Education and Social Sciences. 1st Edition. Nairobi: Masola Publishers. Ostmoe, P. M. (1986). Correlates of University Nurse Faculty Publication Productivity. The Journal of Nursing Education [J Nurs Educ] 1986 May; Vol. 25 (5), pp. 20712. Ouma, G. W. (2008). Higher education marketisation and its discontents: the case of quality in Kenya. Journal Higher Education Issue Volume 56, Number 4 / October, 2008. Oxford University. (1995). Oxford Advanced Learners Dictionary, 7th edn, NewYork: Oxford University Press. Paraje G, Sadana, R., & Karam, G. (2005). Public health. Increasing international gaps in health-related publications. Science, 308 (5724):959-960.

ccclv

Payne, F. & Spieth, E. (1935). An open letter to college teacher, Bloomington, Indiana: Principia Press, Pfeffer, J., & Langton, N. (1993). The effect of wage dispersion on satisfaction, productivity, and working collaboratively: Evidence from college and university faculty. Administrative Science Quarterly, 38, 382. Phillips, J. & Russell, R. (1994). Self-efficacy as a predictor of academic performance in science, Journal of Advanced Nursing, vol. 27, no. 3, p. 596. Price, D. J. De Solla. (1963). Little science, big science. New York: Columbia University Press. Print, M., Hattie, J. (1997). Measuring Quality in Universities: An Approach to Weighting Research Productivity, Higher Education, Vol. 33, No. 4 (Jun., 1997), pp. 453-469. Psacharapoulos, G. (1982). The Economics of Higher Education in Developing Countries. Comparative Education Review, (26 June), 139-59. Psacharapoulos, G. & Woodhall, M. (1985). Education for Development. An Analysis of Investment Choices. New York: Oxford University Press. Ramona, A. K., Lee, S. Y., Skakun, E, N., & Lorne, D. (2007). The Research Productivity of Canadian Physicians: How the Timing of Obtaining a PhD Has an Influence. Academic Medicine Volume 82 - Issue 3 - pp 310-315.

ccclvi

Radhakrishna, R. B. & Jackson, G. (1993). Agricultural and Extension Education Department Heads Attitudes of Journals and Importance of Publishing, Journal of Agricultural Education, 34(4), 8-16. Radhakrishna, R. B., Yoder, E. P., & Scanlon, D .C. (1994). Determinants of Faculty Productivity: Perspectives of Agricultural and Extension Education Faculty. Dallas, TX: National Agricultural Education Research Meeting. ERIC Document Reproduction service no. ED 380 549. Radhakrishna, R. & Jackson, G. (1993), Agricultural and extension education department heads attitudes of journals and importance of publishing, Journal of agricultural Education, vol. 34, no. 4, pp. 8-16. Radhakrishna, R. B., Yoder, E. P., & Scanlon, D. C. (1994). Determinants of faculty productivity: Perspectives of agricultural and extension education faculty. Dallas, TX: National Agricultural Education Research Meeting. ERIC Document Reproduction Service No. ED 380 549. Ramsden, P. (2005). Describing and Explaining Research Productivity. In Journal of Higher Education Publisher Springer Netherlands ISSN 0018-1560 (Print) 1573174X (Online) Issue Volume 28, Number 2 / September, (1994) Pages 207-226 Tuesday, April 05, 2005.

ccclvii

Rennie, D., Yank, V. & Emanuel, L. (1997). When Authorship Fails. A Proposal to Make Contributors Accountable. JAMA; 278: 579-585. Reskin, B. F. (1977). Scientific Productivity and the Reward Structure of Science. American Sociological Review, Vol.42:491-504. Rhodman, A. (2002). The effects of academic origin and academic affiliation on research productivity: an empirical investigation of information system faculty, Doctoral Dissertation, Mississippi State University, USA, viewed 4 May 2005, retrieved from UMI database. Richards, J. M. (1987). Patterns and Correlates of Research Productivity in Population Scientists. Unpublished Paper Presented at the Annual Convention of the American Psychological Association (95th, New York, NY, August 28-September 1, 1987. Rotten, J. (1990). Research productivity, course load, and ratings of instructors, Perceptual and Motor Skills, vol. 71, p. 1388. Sax, L. J, Hagedorn, L.S., Arredondo, M. & Dicrisi, F. A. (2004). Faculty Research Productivity: Exploring the Role of Gender and Family-Related Factors. In Research in Higher Education Springer Netherlands ISSN 0361-0365 (Print) 1573188X

ccclviii

Schultz, J., Mead, J. & Hamana, I. (1989). The changing roles of teaching, research,and service in the promotion and tenure decisions for accounting faculty,Accounting Education, Spring, pp. 109-119. Levin, S. G. & Stephan, P. E. (2001). Research Productivity Over the Life Cycle: Evidence for Academic Scientists, American Economic Review, 81(1), pp. 114 132. Skinner, B. (1953), Science and human behaviour, Macmillan, NY. Thorndike, E. (1911). Animal intelligence, New York: Macmillan. Stahl, M. J., Leap, T. L., & We, Z. Z. (1988). Publication in Leading Management Journals as a Measure of Institutional Research Productivity, Academy of Management Journal 31, 707720. Starbuck, W. H. (2006). The Production of Knowledge: The Challenge of Social Science Research, Oxford: Oxford University Press. Stossel, T. P, Stossel, S. C. (1990). Declining American representation in leading clinicalresearch journals. N Engl J Med; 322: 739-742. Suwanwala, J. (1991). The road of University, Chulalongkorn University Press, Bangkok, Thailand.

ccclix

Taylor, M. Lockes, E. & Gist, M. (1984). Type of a behaviour and faculty research productivity: what are the mechanism? Organizational Behaviour and Human Performance, vol. 34, pp. 402-418. Teferra, D., & Altbach, P. G. (2004). African Higher Education: Challenges for the 21st Century. Higher Education vol.47 pp. 21-50. Tien, F. F. (2000). To What Degree Does the Desire for Promotion Motivate Faculty to Perform Research? Testing the Expectancy Theory. Research in Higher Education, Volume 41, Number 6, December pp. 723-752(30) Publisher: Springer. Tijssen, R. (2007). Africa's contribution to the worldwide research literature: New analytical perspectives, trends, and performance indicators. Scientometrics 2007, 71(2):303-327. Toutkoushian, R. K., Porter, S. R., Danielson, C. & Hollis, P. R. (2003). Using publications counts to measure an institutions research productivity. Research in Higher Education, Vol. 44, No. 2. Tompkins, R. K, Ko, C. Y, Donovan, A. J. (2001), Internationalization of general surgical journals: origin and content of articles published in North America and Great Britain from 1983 to 1998. Arch Surg 2001; 136: 1345-1351. Tornquist, K. M., & Kallsen, L. A. (1992). Out of the ivory tower: An Analysis of the institutional characteristics of universities from which firms acquire scientific

ccclx

results. Minneapolis, MN: Association for the Study of Higher Education. ERIC Document Reproduction Service No. ED352913. Torrez, P. G. (2006). Study Names UC System as Most Successful University. In Technology Transferwww.milkeninstitute.org Tower, G., & Ridgewell, B. (2006). Benchmarking: An International Comparison, International Business and Economics Research Journal, 5, 1-14 Tower, G., Plummer, J. & Ridgewell, B. (2007). A_Multidisciplinary_Study_Of_GenderBased Research Productivity in the Worlds Best Journals. Journal of Diversity Management Fourth Quarter 2007 Volume 2, Number 4. Traynor, M. & Rafferty, A. M. (1998). Nursing Research and the Higher Education Context. Centre for Policy in Nursing Research, London School of Hygiene and Tropical Medicine,London. Trostle, J. (1992). Research capacity building in international health: definitions, evaluations and strategies for success. Social Science and Medicine 35, 1321-1324. Tsay, Ming-yueh. (2004). Literature growth, journal characteristics, and author productivity in subject indexing, 1977 to 2000. Journal of the American Society for Information Science and Technology, 55(1): 64-73. Turnage, J. (1990). The challenge of the new workplace technology for psychology,

ccclxi

American Psychologist, vol. 45, no. 3, pp. 171-178. Turner, L. & Mairesse, J. (2003). Explaining Individual Productivity Differences in Scientific Research Productivity: How important are Institutional and Individual Determinants? An Econometric Analysis of the Publications of French CNRS Physicists in Condensed Matter (1980-1997). Submitted to Annales dEconomie et de Statistiques for the special issue in honor of Zvi Griliches. Trow, M. T. (1996). Markets and Accountability in Higher Education, a Comparative Perspective, Higher Education Policy, 9, (4), 309-324. Usang, B., Akuegwu, U. & Udey, U. (2007). Academic staff research productivity: a study of Universities in South-South Zone of Nigeria. In Educational Research and Review Vol. 2 (5), pp. 103-108, http://www.academicjournals.org/ERR. U N E S C O (2006). Forum on Higher Education, Research, and Knowledge. Global Colloquium on Research and Higher Education Policy Universities as Centers of Research and Knowledge Creation: An Endangered Species? Paris UNESCO (1998), World conference on higher education. Paris: UNESCO. University of Nairobi (2009). A Preamble to the University Of Nairobi. Downloaded on 20th April, 2009 from www.uonbi.ac.ke.

ccclxii

Uthman, O. A & Uthman, M. B. (2007). Geography of Africa biomedical publications: An analysis of 19962005. International Journal of Health Geographics.6:46. Doi: 10,1186/1476-072X-6-46. Vasil, L. (1992). Self-efficacy expectations and causal attributions for achievement among male and female university faculty. Journal of Vocational Behavior, 41(3), 259-269. Vogel, D. & Wetherbe J., (1984). MIS Research: a Profile of Leading Journals and Universities, Database 16 (1984), 314. Vroom, V. (1964). Work and Motivation, New York: Wiley. Wagner, P. J., Hornsby, J. L., Talbert, F. S., Hobbs, J., Brown, G. E. & Kenrick, J. (1994). Publication Productivity in Academic Family Medicine Departments. Family Medicine [Fam Med] 1994 Jun; Vol. 26 (6), pp. 366-9. Waller, K. V., Wyatt, D., & Karni, K. R. (1999). Scholarly activities among clinical laboratory science faculty. Journal of the American Society for Medical Technology [Clin Lab Sci] 1999 Jan-Feb; Vol. 12 (1), pp. 19-27. Wanner, R. A., Lewis, L. S. & Gregorio, D. I (1981). Research Productivity in Academia: A Comparative Study of the Sciences, Social Sciences and Humanities, American Sociological Association. Sociology of Education Vol. 54 (October) 238-253.

ccclxiii

Waworuntu, B., & Holsinger, D. B. (1989). The Research Productivity of Indonesian Professors of Higher Education. Journal of Higher Education Volume 18, Number 2 / March, 1989. 167-187 Welker, M. E. & Cox, A. R. (2006). A Report on Research Activities at Research Universities. Wake Forest University Research Management Review, Volume 15, Number 1 Winter/Spring 2006. West Virginia University (1996). Reinforcement theory, viewed 25 December 2007, http://www.as.wvu.edu/~sbb/comm221/chapters/rf.htm White, C. M. (1995). Uses and impacts of computer mediated communication: asurvey of faculty in mass communication and related disciplines. Unpublished

doctoraldissertation. University of Georgia, Athens. Wierzbicki, A. S & Reynolds, T. M. (2002). Total Research Productivity in a Pathology Discipline. Journal of Clinical Pathology, 2002 July; 55(7): 495498. PMCID: PMC1769702. Wikipedia. The Online Encyclopedia. Williams, A. (2003). A mediated hierarchical regression analysis of factors related to research productivity of human resource education and workforce development

ccclxiv

postsecondary faculty, Doctoral Dissertation, Louisiana State University and Agricultural and Mechanical College, retrieved from UMI database. Wills, D., Ridley, G. & Mitev, H. (2008). Research Productivity of Accounting Academics in Challenging Times: A Meta-analysis, Submitted to the International Conference on Business and Management Education, Bangkok, Thailand, Nov 2008. Wilson, P. M. & Wilson, J. (1989). Research Productivity of Home Economics Education Faculty: Administrators' Perspectives. Family and Consumer Sciences Research Journal, Vol. 18, No. 2, 148-155 (1989). Witzel, M. (1999). Dictionary of Business and Management, International Thomson Business Press, England. World Bank (1988). Education in Sub-Saharan Africa: Policies for Revitalization and Expansion. Washington D.C.: World Bank. Yu, X., & Shauman, K. A. (1998). Sex Differences in Research Productivity: New Evidence about an Old Puzzle. American Sociological Review. Vol. 63(Dec. 847870).

ccclxv

APPENDIX I Joash Migosi (Mr.) C/O Directorate of Research Management and Development PO BOX 30568 00100 Utalii house, 9th Floor Room 935 NAIROBI migosi@scienceandtechnology.go.ke jmigosi@yahoo.co.uk +254 723 869 169 June 18th, 2009

Dear participant, I am writing to kindly request you to give your views and experiences as a university academic staff in Kenya. This questionnaire is designed to study aspects of research output among university academic staff, it will help to obtain information in regard to factors hindering or promoting research output among Kenya university academic staff. The

ccclxvi

findings of this study might be used by university administration, and government, to make decisions on the management of research in this country. The information obtained from this survey will be used for this study only. All ethical issues in regard to this study will be observed. Thank you very much for your time and cooperation. I greatly appreciate you and your institutions help in furthering this research endeavor.

Joash Migosi Postgraduate student, Catholic University of Eastern Africa

UNIVERISTY ACADEMIC STAFF QUESTIONNAIRE A. DEMOGRAPHIC INFORMATION 1. Name of your university___________________________________________ 2. Name of your faculty _____________________________________________ 3. Name of your department _________________________________________

ccclxvii

4. Your age group (kindly tick appropriately) 20-29 Years ( 30-39 Years ( 40-49 Years ( 50-59 Years ( 60-69 Years ( >70 Years ( No response ) ) ) ) ) )

5. Sex (kindly tick appropriately) Male Female ( ) ( )

6. Teaching position/ rank held in the University (kindly tick appropriately) a. Graduate assistant b. Tutorial fellow( ) c. Assistant Lecturer d. Lecturer e. Senior Lecturer f. Associate professor ( ) ( ) ( ) ( ) ( )

ccclxviii

g. Professor

( ) )

h. Any Other (please Specify) (

7. Highest degree obtained, (kindly tick appropriately) a. Bachelors degree b. Masters c. PhD ( ) ( ) ( ) )

d. Any Other (please Specify) (

8. Years since you obtained last highest degree (kindly tick appropriately) a. <5years b. 6-10 years c. 11-15 years d. 16-20years e. More than 20 years f. Any Other (please Specify) ( ( ) ( ) ( ) ( ) ( ) )

9. Kindly estimate time spent on teaching load per week__________________

ccclxix

10. Kindly estimate time spent on research work per week___________________ B. NUMBER OF PUBLICATIONS Kindly enumerate your research publications done between 2004-2008 in the Table below Type of Publication Book(s) Authored Singly Book(s) coAuthored Publications Made in Refereed Journals Papers Presented in Conferences Total number of Publications 2004 2005 2006 2007 2008 Total number of Publications

ccclxx

C. FACTORS AFFECTING RESEARCH PRODUCTIVITY 1. Kindly tick the extent to which the following individual factors affect your research output. Use the key: 1 Large Extent; 2 Some extent; 3 Little Extent; 4 No Extent
1 2 3 4

Self Motivation Socialization with colleagues Research content knowledge Research skills gained Job satisfaction Simultaneous research projects Parenting responsibilities Early orientation to research work Personal Work discipline 2. Kindly tick the extent to which the following institutional factors affect your research output. Use the key: 1 Large Extent; 2 Some extent; 3 Little Extent; 4 No Extent 1 Resources available for research Rewards for research output Good salary Sufficient work time Clear research coordination goals Mentorship among colleagues 2 3 4

ccclxxi

Communication with professional networks library facilities (Books, journals other materials) Size of university Recruitment and selection of academic staff Positive group climate Research emphasis by university Access to relevant journals Teaching load Availability of Technology e.g. internet, computers Equipment for research Number of graduate students supervised D. ATTITUDE ON RESEARCH AND PUBLICATIONS 1. Please indicate how strongly you agree or disagree with the statements regarding the issue of publishing in your institution. (SA Strongly Agree; A-Agree; UUndecided/No opinion-Disagree; SD-Strongly Disagree Item SA My department should not offers additional pay for publishing Lecturers should be offered less teaching load for publishing in refereed journals. Universities should offer monetary rewards as an incentive for publishing in refereed journals. Universities should spend money meant for A U D SD

ccclxxii

research on improving salaries for lecturers Promotions should be based on publishing alone Teaching alone without research is not important Lecturers should spend family resources on research and publishing Research and publishing increases visibility of a university Publishing is important for any lecturer aspiring to grow professionally Universities should be ranked purely on the basis of their research productivity Teaching is more important than publishing Ill rather spend my time as part time lecturer than on writing articles Appointment to senior university management should be based on academic writing Lecturers who do not publish have no business teaching in university It is not advisable for lecturers to spend personal resources on research and publishing Developing countries need research more

ccclxxiii

than developed countries

2. Kindly list factors hindering research productivity among lecturers in your institution and suggest possible solutions to these problems. (Those factors not captured above) Problem Possible Solution

ccclxxiv

THANK YOU FOR YOUR PARTICIPATION APPENDIX II Joash Migosi Catholic University of Eastern Africa PO BOX 62157 00200

ccclxxv

NAIROBI jmigosi@yahoo.co.uk +254 723 869 169 June 18th, 2009

Dear participant, I am writing to kindly request you to give your views and experiences as Head/Dean of department/faculty in a Kenyan University setup. This questionnaire is designed to study aspects of research output among university academic staff, it will help to obtain information in regard to factors hindering or promoting research output among Kenya university academic staff. The findings of this study might be used by university administration, and government, to make decisions on the management of research in this country. The information obtained from this survey will be used for the purposes of this study only. All ethical issues in regard to this study will be observed. Thank you very much for your time and cooperation. I greatly appreciate you and your institutions help in furthering this research endeavour.

ccclxxvi

Joash Migosi Postgraduate student, Catholic University of Eastern Africa

ccclxxvii

UNIVERISTY HEADS OF DEPARTMENTS QUESTIONNAIRE A. DEMOGRAPHIC INFORMATION 11. Name of your university__________________________________________ 12. Name of your faculty_____________________________________________ 13. Name of your department _________________________________________ 14. Your age group (kindly tick appropriately) 20-29 Years ( 30-39 Years ( 40-49 Years ( 50-59 Years ( 60-69 Years ( >70 Years ( ) ) ) ) ) ) )

a. Any Other (please Specify) ( 15. Sex (kindly tick appropriately) Male Female ( ) ( )

16. Teaching position/ rank held in the University (kindly tick appropriately) a. Graduate assistant ( )

ccclxxviii

b. Tutorial fellow( ) c. Assistant Lecturer d. Lecturer e. Senior Lecturer f. Associate professor g. Professor ( ) ( ) ( ) ( ) ( ) )

h. Any Other (please Specify) (

17. Highest degree obtained, (kindly tick appropriately) a. Bachelors degree b. Masters c. PhD ( ) ( ) ( ) )

d. Any Other (please Specify) (

18. Years since you obtained last highest degree (kindly tick appropriately) a. <5years b. 6-10 years c. 11-15 years d. 16-20years ( ) ( ) ( ) ( )

ccclxxix

e. More than 20 years

( ) )

f. Any Other (please Specify) (

19. Kindly estimate time spent on teaching load per week ________________ 20. Kindly estimate time spent on research work per week________________ B. NUMBER OF PUBLICATIONS Kindly enumerate research publications done by your staff between 2004-2008 in the Table below Type of Publication Book(s) Authored Singly Book(s) coAuthored Publications Made in Refereed Journals Papers Presented in Conferences 2004 2005 2006 2007 2008 Total number of Publications

ccclxxx

Any Other category of publications. (Kindly Specify) Total number of Publications C. FACTORS AFFECTING RESEARCH PRODUCTIVITY 1. Kindly tick the extent to which the following individual factors affect your staffs research output. Use the key: 1 Large Extent; 2 Some extent; 3 Little Extent; 4 No Extent
1 2 3 4

Self Motivation Socialization with colleagues Research content knowledge Research skills gained Job satisfaction Simultaneous research projects Parenting responsibilities Early orientation to research work Personal Work discipline

ccclxxxi

2. Kindly tick the extent to which the following institutional factors affect your staffs research output. Use the key: 1 Large Extent; 2 Some extent; 3 Little Extent; 4 No Extent 1 Resources available for research Rewards for research output Good salary Sufficient work time Clear research coordination goals Mentorship among colleagues Communication with professional networks library facilities (Books, journals other materials) Size of university Recruitment and selection of academic staff Positive group climate Research emphasis by university Access to relevant journals Teaching load Availability of Technology e.g. internet, computers Equipment for research Number of graduate students supervised 2 3 4

2. Kindly list possible factors hindering research productivity among lecturers in your department and suggest possible solutions to these problems. Problem Possible Solution

ccclxxxii

THANK YOU FOR YOUR PARTICIPATION

APPENDIX III

ccclxxxiii

Document Analysis Institutional research policies will be scrutinized to find out the following information;

1. Name of University

2. Year of research policy publication 3. Contents of research policy

4. Deliberate efforts to promote research productivity as outlined in the research

policy.

ccclxxxiv

APPENDIX IV

ccclxxxv

ccclxxxvi

ccclxxxvii

ccclxxxviii

ccclxxxix

cccxc

cccxci

cccxcii

cccxciii

cccxciv

cccxcv

cccxcvi

cccxcvii

cccxcviii

cccxcix

cd