Sie sind auf Seite 1von 4

This material is protected by U.S. copyright law. Unauthorized reproduction or online display is prohibited.

To purchase
quantity reprints, e-mail reprints@ons.org. For permission to reproduce multiple copies, e-mail pubpermissions@ons.org

Methods & Meanings


Diane G. Cope, PhD, ARNP, BC, AOCNP® • Associate Editor

The Use of Triangulation in Qualitative Research

Nancy Carter, RN, PhD, Denise Bryant-Lukosius, RN, PhD, Alba DiCenso, RN, PhD, Jennifer Blythe, PhD,
and Alan J. Neville, MBChB, MEd, MRCP, FRCP(c)

T
riangulation refers to the use of Data source triangulation involves the roso, 2003; Webb & Kevern, 2001; Zorn,
multiple methods or data sources collection of data from different types of Roper, Broadfoot, & Weaver, 2006). In
in qualitative research to develop people, including individuals, groups, terms of time, compared to IDI inter-
a comprehensive understanding of phe- families, and communities, to gain mul- views, FGs may initially be less demand-
nomena (Patton, 1999). Triangulation also tiple perspectives and validation of data. ing to researchers; however, the time and
has been viewed as a qualitative research effort required to analyze the complex
strategy to test validity through the con- data elicited from FGs might ultimately
Data Source Triangulation
vergence of information from different negate any time savings (Mansell, Ben-
sources. Denzin (1978) and Patton (1999) Most qualitative researchers studying nett, Northway, Mead, & Moseley, 2004).
identified four types of triangulation: (a) human phenomena collect data through The nature of data yielded by these
method triangulation, (b) investigator interviews with individuals or groups; two methods of collection differs. Brown
triangulation, (c) theory triangulation, their selection of the type of interview (1999) explained that FGs differ from IDI
and (d) data source triangulation. The depends on the purpose of the study interviews in that the “dynamic and in-
current article will present the four types and the resources available. Fontana and teractive exchange among participants”
of triangulation followed by a discussion Frey (2000) described the IDI interview in FGs lead them to produce “multiple
of the use of focus groups (FGs) and in- as one of the most powerful tools for stories and diverse experiences” (p.
depth individual (IDI) interviews as an gaining an understanding of human be- 115). Fern (1982) found that those who
example of data source triangulation in ings and exploring topics in depth. IDI participated in IDI interviews generated
qualitative inquiry. interviews, ranging from the structured more ideas than did those participating
and controlled to the unstructured and in either moderated or unmoderated
Types of Triangulation fluid, can elicit rich information about FGs. In a communications study, DeJong
personal experiences and perspectives and Schellens (1998) compared the use
The first type of triangulation is meth- (Russell, Gregory, Ploeg, DiCenso, & of IDIs and FGs to evaluate the text in
od triangulation. Method triangulation Guyatt, 2005). IDI interviews allow for a brochure about alcohol consumption
involves the use of multiple methods of spontaneity, flexibility, and responsive- and found that IDI participants focused
data collection about the same phenom- ness to individuals; however, conduct- on the finer details of the text, whereas
enon (Polit & Beck, 2012). This type of ing the interviews, transcribing the the interaction among FG participants
triangulation, frequently used in quali- discourse, and analyzing the text often identified potential problems with the
tative studies, may include interviews, require considerable time and effort. brochure. Kaplowitz (2000, 2001) found
observation, and field notes. In contrast, FGs elicit data from a that IDI interview participants were
Investigator triangulation involves group of participants who can hear more likely to discuss sensitive topics
the participation of two or more re- each other’s responses and provide ad- and stimulate discussion about dif-
searchers in the same study to provide ditional comments that they might not ferent topics when compared to FG
multiple observations and conclusions. have made individually. Researchers participants. Kaplowitz and Hoehn
This type of triangulation can bring who conduct FGs recognize that the (2001) found that using FGs and IDI
both confirmation of findings and dif- participant interaction, which stimulates interviews provided different perspec-
ferent perspectives, adding breadth to the identification and sharing of various tives on resources, values, and issues
the phenomenon of interest (Denzin, perspectives on the same topic, is central and concluded that one method was not
1978). to their success (Morgan, 1996). Several better than the other, but rather that the
Theory triangulation uses different authors have pointed out that research- two approaches were complementary. In
theories to analyze and interpret data. ers rarely evaluate or discuss this ap-
With this type of triangulation, differ- proach (Clayton, Butow, Arnold, & Tat-
ent theories or hypotheses can assist tersall, 2005; Duggleby, 2005; Kitzinger, ONF, 41(5), 545–547.
the researcher in supporting or refuting 1994; Lehoux, Poland, & Daudelin, 2006;
doi: 10.1188/14.ONF.545-547
findings. Sandelowski, 2000; Sandelowski & Bar-

Oncology Nursing Forum • Vol. 41, No. 5, September 2014 545


an ethnographic study exploring adoles- data via each approach, and comparison methodologic issues associated with
cent boys’ thoughts about sex, sequential of study data provided through the use combining FG and IDI interview data is
observations, FGs, and IDI interviews of each method. needed to better understand the impli-
were conducted with the same par- The second reason is to increase the cations of this approach and to further
ticipants (Wight, 1994). The adolescent validity of study findings through tri- explore the differences between FG and
boys expressed greater sensitivity and angulation and the collection of data IDI interview data.
were more open when participating in from all study participants using both
IDI interviews but displayed stronger methods, beginning with IDI interviews Nancy Carter, RN, PhD, is an assistant
professor in the School of Nursing at
expressions of masculinity and were and followed by FGs, or vice versa. The McMaster University in Hamilton, On-
more guarded when participating in FGs researcher must describe both methods tario; Denise Bryant-Lukosius, RN, PhD,
(Wight, 1994). of data collection, compare the study is an associate professor in the School of
Typically, researchers determine data results from each method, and describe Nursing and Department of Oncology at
collection methods based on the best fit how the data were integrated to arrive McMaster University, and clinician scien-
with the research question. Both FGs at study results. The strength of this tist and director at the Canadian Centre of
and IDI interviews may be intentionally consecutive method of data collection is Excellence in Oncology Advanced Practice
selected by researchers for the purpose the opportunity to triangulate the data Nursing at the Juravinski Cancer Centre;
and Alba DiCenso, RN, PhD, is a professor
of data triangulation or may be selected and to perform member checking. A emeritus in the School of Nursing and the
later in the research process as a result limitation of this approach is the restric- Department of Clinical Epidemiology and
of unanticipated challenges in data tion of study participants to only those Biostatistics, Jennifer Blythe, PhD, is an as-
collection. Lambert and Loiselle (2008) who can participate in both methods, sociate professor in the School of Nursing,
explored patterns of cancer information- therefore narrowing the spectrum of and Alan J. Neville, MBChB, MEd, MRCP,
seeking behavior and initially used both eligible patients. FRCP(c), is a professor in the Department
IDI interviews and FGs as a result of Several challenges exist when per- of Oncology and associate dean of educa-
some participants’ inability or unwill- forming data triangulation with the tion in the Faculty of Health Sciences, all
at McMaster University, all in Canada. No
ingness to participate in FGs. Purposeful use of both FGs and IDI interviews. financial relationships to disclose. Carter
use of these two methods, however, was Researchers must have a variety of can be reached at carternm@mcmaster
later performed once preliminary study strategies to ensure data dependabil- .ca, with copy to editor at ONFEditor@
findings revealed greater understanding ity and credibility, such as debriefing, ons.org.
of the phenomenon. Three methodologic member checking, triangulation, or use
Key words: triangulation; in-depth indi-
observations were made about the data of a reflexive journal. The assumption vidual interviews; focus groups; qualita-
derived: (a) comparing the data led to that more data are always better over- tive research
an iterative process, whereby phenom- shadows concerns about what to do
ena were explored more deeply, (b) with both types of data (Barbour, 1998).
References
the combined data led to an enhanced Questions about the analysis of the data
understanding of the context of the may arise. For example, if using the two Barbour, R.S. (1998). Mixing qualitative
phenomena, and (c) convergence of the methods, how are FG and IDI interview methods: Quality assurance or quagmire?
data enhanced trustworthiness of find- data analyzed together? Do concerns Qualitative Health Research, 8, 352–361.
ings. The authors suggested that further exist about the weighting of data? For Brown, J.B. (1999). The use of focus groups
research was needed to understand how example, does one FG with six par- for clinical research. In B.F. Crabtree &
W. L. Miller (Eds.), Doing qualitative re-
various types of data contribute to un- ticipants carry the same weight as one
search (2nd ed., pp. 109–124). Thousand
derstanding of phenomena (Lambert & IDI interview? Morse (2009) suggested
Oaks, CA: Sage.
Loiselle, 2008). that ad hoc combination of methods Clayton, J.M., Butow, P.N., Arnold, R.M.,
threatens trustworthiness. Therefore, & Tattersall, M.H. (2005). Discussing
Merits and Challenges the researcher performing data trian- life expectancy with terminally ill cancer
gulation must consider these issues and
of Combined Use analyze the data separately, synthesize
patients and their careers: A qualitative
study. Supportive Care in Cancer, 13,
Merits and challenges exist to us- and identify similarities and differences, 733–742.
ing both IDI interviews and FGs in a and conclude how the different methods DeJong, M. & Schellens, P.J. (1998). Focus
single study. Morse (2009) suggested affect the results. groups or individual interviews? A com-
that mixing qualitative methods allows parison of text evaluation approaches.
for different perspectives that may oth- Technical Communication, 45, 77–88.
Conclusion Denzin, N.K. (1978). Sociological methods: A
erwise be overlooked. Two important
reasons should be considered in using Data triangulation using FGs and IDI sourcebook. New York, NY: McGraw-Hill.
Duggleby, W. (2005). What about focus
both FGs and IDI interviews. The first interviews in qualitative inquiry may
group interaction data? Qualitative Health
is to increase participation of a broader result in a broader understanding of the
Research, 15, 832–840.
spectrum of eligible patients who might phenomenon of interest. Limiting data Fern, E.F. (1982). The use of focus groups
not otherwise be able to participate if collection to one of the two methods for idea generation: The effects of group
restricted to one method of data collec- may result in the exclusion of eligible size, acquaintanceship, and moderator on
tion (e.g., too ill to attend a FG). In that patients and may lessen the breadth response quantity and quality. Retrieved
scenario, the researcher must describe of results by only gaining partial in- from http://www.uta.edu/faculty/
both methods of data collection, the sight into the phenomenon of interest. richarme/MARK%205338/Articles/
number of participants who contributed Further examination of the potential Fern.pdf

546 Vol. 41, No. 5, September 2014 • Oncology Nursing Forum


Fontana, A. & Frey, J.H. (2000). The in- Mansell, I., Bennett, G., Northway, R., Sandelowski, M. (2000). Whatever hap-
terview. From structured questions to Mead, D., & Moseley, L. (2004). The pened to qualitative description? Research
negotiated text. In Y.S.Lincoln & N.K. learning curve: The advantages and in Nursing and Health, 23, 334–340.
Denzin (Eds.), Handbook of qualitative re- disadvantages in the use of focus groups Sandelowski, M., & Barroso, J. (2003).
search (2nd ed., pp. 645–672). Thousand as a method of data collection. Nurse Classifying the findings in qualitative
Oaks, CA: Sage. Researcher, 11, 79–88. studies. Qualitative Health Research, 13,
Kaplowitz, M.D. (2000). Statistical analysis Morgan, D.L. (1996). Focus groups. Annual 905–923. doi:10.1177/1049732303253488
of sensitive topics in group and indi- Review of Sociology, 22, 129–152. Webb, C. & Kevern, J. (2001). Focus groups
vidual interviews. Quality and Quantity, Morse, J.M. (2009). Mixing qualitative meth- as a research method: A critique of some
34, 419–431. ods. Qualitative Health Research, 19, 1523– aspects of their use in nursing research.
Kaplowitz, M.D. (2001). Assessing man- 1524. doi:10.1177/1049732309349360 Journal of Advanced Nursing, 33, 798–805.
grove products and services at the lo- Patton, M.Q. (1999). Enhancing the quality doi:10.1046/j.1365-2648.2001.01720.x
cal level: The use of focus groups and and credibility of qualitative analysis. Wight, D. (1994). Boys’ thoughts and
individual interviews. Landscape and Health Sciences Research, 34, 1189–1208. talk about sex in a working class local-
Urban Planning, 56, 53–60. Polit, D.F., & Beck, C.T. (2012). Nursing ity in Glasgow. Sociological Review, 42,
Kaplowitz, M.D., & Hoehn, J.P. (2001). Do research: Generating and assessing evidence 702–737.
focus groups and individual interviews for nursing practice. Philadelphia, PA: Zorn, T.E., Roper, J., Broadfoot, K., &
reveal the same information for natural Lippincott Williams and Wilkins. Weaver, C.K. (2006). Focus groups as
resource valuation? Ecological Economics, Russell, C., Gregory, D., Ploeg, J., DiCenso, sites of influential interaction: Building
36, 237–247. A., & Guyatt, G. (2005). Qualitative re- communicative self-efficacy and effect-
Kitzinger, J. (1994). The methodology of search. In A. DiCenso, G. Guyatt, & D. ing attitudinal change in discussing
focus groups: The importance of interac- Ciliska (Eds.), Evidence-based nursing: A controversial topics. Journal of Applied
tion between research participants. So- guide to clinical practice (pp. 120–136). St. Communication Research, 34, 115–140.
ciology of Health and Illness, 16, 103–121. Louis, MO: Elsevier Mosby. doi:10.1080/00909880600573965
Lambert, S.D. & Loiselle, C.G. (2008).
Combining individual interviews and
focus groups to enhance data richness. Methods & Meanings
Journal of Advanced Nursing, 62, 228–237.
doi:10.1111/j.1365-2648.2007.04559.x Methods & Meanings comments of Oncology Nursing Forum. For more
Lehoux, P., Poland, B., & Daudelin, G. and provides background on the information, contact Associate Editor
(2006). Focus group research and “the methodology used in one of the stud- Diane G. Cope, PhD, ARNP, BC,
patient’s view.” Social Science and Medi- ies reported in the that month’s issue AOCNP®, at dgcope@comcast.net.
cine, 63, 2091–2104.

Oncology Nursing Forum • Vol. 41, No. 5, September 2014 547


Copyright of Oncology Nursing Forum is the property of Oncology Nursing Society and its
content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.