Sie sind auf Seite 1von 75

Introduction to scientific

research 2.

Course Objectives
1. Research Methodology
1. Qualitative methods
2. Quantitative methods

2. How to do an interview
3. How to prepare a survey
4. How to prepare the presentation

Differences between
methods

Differences between quantitative and qualitative research

For a quantitative researcher, reality is objective;


It exists apart from the researcher and is capable of being
seen by all. In other words, it's out there.

For the qualitative researcher, there is no one single


reality.
Each observer creates reality as part of the research
process
It is subjective and exists only in reference to the observer.

The quantitative researcher believes that reality can


be divided into component parts, and he or she gains
knowledge of the whole by looking at these parts.

Differences between quantitative and qualitative research

The quantitative researcher believes all human


beings are basically similar and looks for general
categories to summarize their behaviors or feelings.
The average human!

The qualitative investigator believes that human


beings are all fundamentally different and cannot be
pigeonholed.

Differences between quantitative and qualitative research

Quantitative researchers aim to generate general


laws of behavior and explain many things across
many settings.
In contrast, qualitative scholars attempt to produce a
unique explanation about a given situation or
individual.
Whereas quantitative esearchers strive for breadth,
qualitative researchers strive for depth.
The practical differences between these approaches
are perhaps most apparent in the research process.

Five main differences in the research areas


Role of the researcher.
The quantitative researcher strives for objectivity and is
separated from the data.
The qualitative researcher is an integral part of the data; in
fact, without the active participation of the researcher, no
data exist.

Design.
In quantitative methods, the design of the study is
determined before it begins.
In qualitative research, the design evolves during the
research; it can be adjusted or changed as it progresses.

Five main differences in the research areas


Setting
Quantitative researchers try to control contaminating
and/or confounding variables by conducting their
investigations in laboratory settings.
Qualitative researchers conduct their studies in the field, in
natural surroundings.
They try to capture the normal flow of events, without
trying to control the extraneous variables.

Measurement instruments.
In quantitative research, these exist apart from the
researcher. In fact, another party could use the
instruments to collect data in the researcher's absence.
In qualitative research, the investigator is the instrument;
no other individual could fill in for the qualitative
researcher.

Five main differences in the research areas


Theory building.
In the quantitative area, research is used to test theory
and to ultimately support or reject it.
In the qualitative area, theory is "data driven" and
emerges as part of the research process, evolving from the
data as they are collected.

Qualitative methods

Types of qualitative research

field observations
focus groups
Intensive interviews
case studies.

Field observation
Field observation involves the study of a
phenomenon in natural settings.
The researcher may be a detached observer or a
participant in the process under study.
The main advantage of this technique is
its flexibility
it can be used to develop hypotheses, to gather
preliminary data
to study groups that would otherwise be inaccessible.

Its biggest disadvantage is the


difficulty in achieving external validity.

Focus group
The focus group, or group interviewing, is used to
gather preliminary information for a research study
or to gather qualitative data concerning a research
question.
The advantages of the focus group method are
the ease of data collection
The depth of information that can be gathered.

Among the disadvantages:


the quality of information gathered during focus groups
depends heavily on the group moderators' skill
focus groups can only complement other research because
they provide qualitative not quantitative data.

Focus groups on practice


The focus group, or group interviewing, is a research
strategy for understanding audience/ consumer
attitudes and behavior.
From 6 to 12 people are interviewed simultaneously,
with a moderator leading the respondents in a
relatively free discussion about the focal topic.
The identifying characteristic of the focus group is
controlled group discussion, which is employed to
gather preliminary information for a research
project, to help develop questionnaire items for
survey research, or to understand the reasons
behind a particular phenomenon.

The seven basic steps of focus groups


1.
2.
3.
4.
5.
6.
7.

Define the problem


Select a sample
Determine the number of groups necessary
Prepare the study mechanics
Prepare the focus group materials
Conduct the session.
Analyze the data and prepare a summary report

Interviewing
Intensive interviewing is used to gather extremely
detailed information from a small sample of
respondents.
Advantage
The wealth of data that can be gathered
Because intensive interviewing is usually done with small,
nonrandom samples

Disatvantage
generalizability
Interviewer bias

Case study
The case study method draws from as many data
sources as possible to investigate an event.
Case studies are particularly helpful when a
researcher desires to explain or understand some
phenomenon.
Some problems with case studies are that they can
lack scientific rigor
can be time-consuming to conduct
The data they provide can be difficult to generalize from
and to summarize.

Four essential characteristics of case study research:


1. Particularistic. This means that the case study focuses
on a particular situation, event, program, or
phenomenon, making it a good method for studying
practical real-life problems.
2. Descriptive. The final result of a case study is a detailed
description of the topic under study
3. Heuristic. A case study helps people to understand
what's being studied. New interpretations, new
perspectives, new meaning, and fresh insights are all
goals of a case study.
4. Inductive. Most case studies depend on inductive
reasoning. Principles and generalizations emerge from
an examination of the data. Many case studies attempt
to discover new relationships rather than verify existing
hypotheses.

Method of case study


The precise method of conducting a case study has
not been as well documented as the more traditional
techniques of the survey and the experiment.
Nonetheless, there appear to be five distinct stages
in carrying out a case study:

design
pilot study
data collection
data analysis
and report writing.

The Interview

The interview
The interview is one of the most common datagathering devices.
It can be employed to study a wide range of issues,
across widely varying samples of respondents.

Crano-Brewer, 2002

Rules of thumb in question construction


Keep the items as brief as possiblethe longer the item,
the more likely it is to succumb to one or another of the
problems listed below.
Avoid subtle shadingsif you want to know about
something, ask about it as directly as possible.
Avoid double-barreled questions, that is, questions that
logically allow for two (possibly opposed) answers, for
example, Do you like this years Fords or Chryslers?
Use language the respondents can understand (most
people are not social scientists, so to use the jargon of
the field is probably ill-advised).
If at all possible, pretest items on a small sample of
respondents drawn from the same population that will
constitute the ultimate data source.

Data to be collected
Sociodemographic Information
Questions concerned with descriptive personal
characteristics of the respondent (age, religion, sex, race,
income, etc.) are perhaps the most common of all items
included in the interview.

Reconstruction
The research interview is perhaps the most practical, and
certainly the most common, means of investigating
peoples reconstructions of past events.
Very often, events having important social implications
occur so rapidly or unexpectedly that researchers are
unable to observe behavior at the time the events occur.

Problems with reconstructions


There are three important factors that influence the
fidelity of recall of an event.
Uniqueness of the event
The magnitude of the events economic or social costs or
benefits
The long-term, continuing nature of the event

Data to be collected
Attitudes and Behavioral Intentions
We distinguish between these terms because an attitude,
an evaluative belief about a person or thing, may or may
not carry with it any clear behavioral implications, whereas
a behavioral intention is clearly an indication of an
individuals decision to act in a certain manner.

Private Beliefs and Actions


Sometimes, the focus of an investigation is on behavior
that is of a highly personal, secretive, or illegal nature.
People engaging in such actions usually are not willing to
be observed
However, surprisingly, they often are quite willing to
discuss their experiences, especially if they can be assured
of anonymity or confidentiality.

Interview structure (Types of interviews)


The Exploratory Interview
The exploratory interview resembles in general form a
participant, nonstructured, freeresponse observational
investigation.
In research of this type, neither the questions nor the
allowable responses are constrained.
There is no interview roadmap(structure) to guide the
interviewer, whose behavior is dictated by the responses
received
He or she is encouraged to follow any leads that appear
promising or informative.

The Structured-Nonscheduled Interview


Structured-nonscheduled interview techniques are
midway between the completely open,
nonstructured exploratory approaches and the
standardized, structured techniques.
The structured-nonscheduled interview imposes on
the researcher the necessity of obtaining certain
highly specified types of information (hence the
term, structured) but does not specify the manner in
which the information is to be obtained;
That is, no list of prespecified questions (the
interview schedule) is employed.

The Structured-Scheduled Interview


Before proceeding with a detailed discussion of this
technique, it is important to note a distinction between
the types of schedule that can be employed in this form
of interview.
The more common form employed in scheduled
interviews is the closed question.
Interviews making use of this type of question are analogous to
verbal multiple-choice tests.
Information is asked of a respondent, who is given a set of
allowable answers from which to choose
for example, Are you a Catholic, Protestant, Jew, or Muslim?
The open form of this question would be, Whatis your
religion?

Open vs. Closed questions


The most obvious administrative difference between
these approaches is that the use of open-ended
questions places somewhat greater demands on the
interviewer, who must transcribe the respondents
replies.
At the analysis phase, much greater costs can be
incurred, especially on questions allowing for a more
wide-ranging series of replies because with this form
of question, a system for classifying
respondentsanswers must be developed before any
analysis can proceed

Important: Avoid Leading questions!!

Quantitative Research

The experiment
1. Obtaining a pool of participants.
2. Pretesting them on the dependent variable of
interest.
3. Randomly assigning each participant to experimental
or control groups.
4. Carefully controlling for differences in the application
of the experimental treatment between the two
groups.
5. Remeasuring both groups on the dependent variable
at some time following the experimental treatment.

Pretest
The pretest in this basic design allows us to
demonstrate that those assigned to the two different
treatment conditions did not differ in their response
to the dependent variable at the outset of the
experiment.
Ideally, after random assignment the two groups are
essentially the same, on average, in their pretest
scores within the limits of chance variation.

Laboratory experiments

Prepare Materials
Instructions must be prepared, independent
variables and dependent measures planned and
constructed, and debriefings written.
The debriefing is the experimenters honest
explanation of what the study is about, and it is an
indispensable part of the study, especially if
voluntary participants are used in the researche.
In some research, participants are not aware of their being
under investigation, and in some instances, it is impractical
or impossible to debrief.

Submit to IRB
The instructions, dependent and independent
variables, and debriefing, must be submitted for
approval to a committee specifically constituted to
protect the welfare of research participants.
In most universities, this body is called the
Institutional Review Board (IRB) or, less formally, the
human participants committee.
It is imperative that no research involving human
participation ever be conducted without prior
approval of the IRB.

Set Up Environment
In considering all of the elements of an experimental
setting, we must distinguish between those features
of the context that are to be held constant, and
those that are to be systematically manipulated.
Experimental contexts are characterized by both a
physical environment and a social environment.
Apart from the specific features that are to be
manipulated as the independent variable (or
variables), it is critical to good experimental design
that these other features be defined and controlled
by the experimenter in such a way that they do not
interfere with the intended independent variable.

The experiment must be


Directed
Monitred
Repeatable/Reproducable

Environmentally
Tasks
Measures
Etc.

Accurate
Exact
Precise
Well documented

The survey/questionaire
method

Before the survey


Researchers must decide whether to

use a descriptive or an analytical approach


define the purpose of the study;
review the available literature in the area
select a survey approach
a questionnaire design, and a sample;
analyze and interpret the data;
and, finally, decide whether to publish or disseminate the
results

These steps are not necessarily taken in that order,


but all must be considered before a survey is
conducted.

Survey research
In survey research, the exact questions we ask are
our operationalizations.
The goal in questionnaire design is to avoid bias in
answers.
Question wording, length, style, and order may affect
a respondent's answers.

Why wording important?


Donald Rugg found that Americans' support for
freedom of speech was drastically altered by
different wordings of the following questions:
Do you think the United States should forbid public
speeches against democracy?
Do you think the United States should allow public
speeches against democracy?
(Schuman 2002, my emphasis)
The "forbid" question generated a much lower
agreement rate (54%) than the "allow" question
(75%)

Rules of questioning
The questions we pose should be dear in meaning and
free of ambiguity.
Do you exercise on a regular basis?
What is wrong with this question?

Survey questions should use common everyday language


the use of specialized language such as jargon, abbreviations, or
acronyms should be avoided.

Survey questions should use neutral language


emotional or leading language should be avoided.
Typically, emotional language is used to lead respondents to the
desired response.
Leading questions suggest to the respondent that one response
is better than or preferred over another.

Example of emotional leading


Social benefits for the unemployed is not a good
solution, because it makes them work less, and
spend it to alcohol.

Social benefits for the unemployed is good, because


it helps them to reactivate.

Example of emotional leading


University studies are helping the community,
because there will be more scientific capabilities.

University studies are just for making fun these days,


the students only want to party and high salary jobs!

Rules of questioning
Survey questions should be simple and easy for
respondents to answer
How do you rate police response time to emergency and
nonemergency calls? (wrong)

Avoid double negatives


Double negatives don't make no good sense in our writing
or in surveys.
Does it seem possible or does it seem impossible to you
that the Nazi extermination of the Jews never happened?
(22% - not happened)
Do you doubt that the Holocaust actually happened or
not? (3% - not happened)

Descriptive survey
A descriptive survey attempts to picture or
document current conditions or attitudes, that is, to
describe what exists at the moment.
For example, the Department of Labor regularly conducts
surveys on the amount of unemployment in the United
States.
Professional pollsters survey the electorate to learn its
opinions of candidates or issues.

In descriptive surveys of this type, researchers are


interested in discovering the current situation in a
given area.

Analytical survey
Analytical surveys attempt to describe and explain
why certain situations exist.
In this approach two or more variables are usually
examined to test research hypotheses.
The results allow researchers to examine the
interrelationships among variables and to draw
explanatory inferences.
For example, television station owners occasionally survey
the market to determine how lifestyles affect viewing
habits, or to determine whether viewers' lifestyles can be
used to predict the success of syndicated programming

Advantages
They can be used to investigate problems in realistic
settings.
The cost of surveys is reasonable considering the amount
of information gathered.
Researchers can control expenses by selecting from four
major types of surveys:
mail, telephone, personal interview, and group administration.

Large amounts of data can be collected with relative ease


from a variety of people.
Data helpful to survey research already exist.
Data archives, government documents, census materials, radio
and television rating books, and voter registration lists can be
used as primary sources (main sources of data) or as secondary
sources (supportive data) of information.

Disadvantages
Independent variables cannot be manipulated as in
laboratory experiments.
Without control of independent variable variation, the
researcher cannot be certain whether the relations
between independent and dependent variables are causal
or noncausal.

Inappropriate wording and placement of questions


within a questionnaire can bias results.
The potential problem of talking to the wrong
people.

Types of questions
Two basic types of questions
1. Open-ended
2. Close-ended

Open-ended questions
An open-ended question requires respondents to
generate their own answers. For example:
What do you like most about your local newspaper?
What type of television program do you prefer?
What are the three most important problems in your
community?

Open-ended questions allow respondents freedom in


answering questions and the chance to provide indepth responses.
Also, open-ended questions allow for answers that
researchers did not foresee in the construction of the
questionnaireanswers that may suggest possible
relationships with other answers or variables.

Open-ended questions
Open-ended questions are particularly useful in a
pilot version of a study.
Like at qualitative interview

From the list of responses provided by the subjects,


the researcher then selects the most-often
mentioned items and includes them in multiplechoice or forcedchoice questions.

Open ended questions


Disadvantages
The amount of time needed to collect and analyze the
responses.
Openended responses required interviewers to spend a lot
of time writing down or typing answers.

Open-ended questions
Problems with interpreting
In many cases, respondents' answers are bizarre.
Sometimes respondents don't understand a question and
provide answers that are not relevant.
Sometimes interviewers have difficulty understanding
respondents, or they may have problems with spelling
what the respondents say.

Once again, the main rules

Clear questions
As short as possible
Follow the purpose of the research
No double-barraled questions
Avoid biased words or terms
"In your free time, would you rather read a book or just
watch television?

Avoid leading questions


Do not use questions that ask for highly detailed
information
Avoid potentially embarrassing questions unless
absolutely necessary

Closed-ended questions
Dichotomous questions (Only two answers)
Do you agree with? 1 Yes ; 2 No
Are you a male or Female? 1-Male; 2-Female

Multiple choice questions


allows respondents to choose an answer from several
options.
In general, television commercials tell the truth. . .

All of the time


Most of the time
Some of the time
Rarely
Never

Multiple-choice questions
should include all possible responses.
A question that excludes any significant response usually creates
problems. For example:
What is your favorite television network?
Channel 1
Channel 2
Channel 3

Subjects who favor Channel 4 or 5 (although not networks in the


strictest sense of the word) cannot answer the question as
presented.

must be mutually exclusive:


there should be only one response option per question for each
respondent. For instance:
How many years have you been working in newspapers?
Less than one year
One to five years
Five to ten years (not good) six to ten years!!!

Rating scales
They can be arranged horizontally or vertically:
There are too many commercials on TV.

Strongly agree (translated as a 5 for analysis)


Agree (translated as a 4) Neutral (translated as a 3)
Disagree (translated as a 2)
Strongly Disagree (translated as a l)

Rating scales
Semantic differential scales
Frequently used to rate persons, concepts, or objects.
These scales use bipolar adjectives with seven scale points:

Rank ordering
The relative perception of several concepts or items.
Here are several common occupations. Please rank them in
terms of their prestige. Put a 1 next to the profession that has
the most prestige, a 2 next to the one with the second most,
and so on.

Police officer
Banker
Lawyer
Politician
TV reporter
Teacher
Dentist
Newspaper writer

Ranking of more than a dozen objects is not


recommended because the process can become tedious
and the discriminations exceedingly Fine.

Checklist question
What things do you look for in a new television set?
(Check as many as apply.)

Automatic fine tuning


Remote control
Large screen
Cable ready
Console model
Portable Stereo sound
Other _________

Forced-choice questions
Select one statement from each of the following
pairs of statements:

I enjoy attending parties with my friends.


I enjoy staying at home alone.
o Gun control is necessary to stop crime.
o Gun control can only increase crime.

Respondents generally complain that neither of the


responses to a forced-choice question is satisfactory,
but they have to select one or the other.

Fill-in-the-blank
Are used infrequently by survey researchers.
However, some studies are particularly suited for fillin-the- bla
For example,
"The senators from your state are _____ and _____." Or,
"The headline story on the front page was about
_____."nk questions.

Tables, graphs, figures


Some ingenious questioning devices have been
developed to help respondents more accurately
describe how they think and feel.

Designing the questionaire


Introduction
Who we are, why we are doing the research, how we will use
the answers.
Always make it clear, that responcing is VOLUNTARY!

Instructions
All instructions necessary to complete the questionnaire should
be clearly stated for respondents or interviewers.

Screener questions, or filter questions


used to eliminate unwanted respondents (or to include only
respondents who have specific characteristics or answer
questions in a specific manner), often require respondents or
interviewers to skip one or more questions.
Skips must be clearly specified.
In a typical week, do you listen to AM radio?
____ Yes
____ No [SKIP TO Q. 17]

Question order
All surveys flow better when the initial questions are
simple and easy to answer.
The questionnaire should be organized in a logical
sequence, proceeding from the general to the
specific.
Poor question order may bias a respondent's
answers.

Layout
BAD

GOOD

Layout
Format changes generally create confusion for both
respondents and interviewers.
Each question must have enough space for answers.
This is especially true for open-ended questions.

Questionaire lenght
There are no strict guidelines to help in deciding how
long a questionnaire should be. The length depends
on a variety of things.
Some of these include:
1. Purpose of the survey
2. Type of problems or questions investigated
3. Age of respondents involved in the survey
4. Type and complexity of questions in the questionnaire
5. Location in the country where the study is conducted
6. Specific setting of the testing situation
7. Time of year
8. Time of day
9. Type of interviewer used (professional or amateur)

General problems with survey research


Subjects or respondents are often unable to recall
information about themselves or their activities.
Due to a respondent's feelings of inadequacy or lack of
knowledge about a particular topic, they often provide
"prestigious" answers rather than admit they don't know
something.
Subjects may purposely deceive researchers by giving
incorrect answers to questions.
Respondents often give elaborate answers to simple
questions because they try to "figure out" the purpose of a
study, and what the researcher is doing.
Surveys are often complicated by the inability of respondents
to explain their true feelings, perceptions, and beliefs not
because they don't have any, but because they can't put
them into words.