Quality & Quantity 34: 419–431, 2000.
© 2000 Kluwer Academic Publishers. Printed in the Netherlands.
Statistical Analysis of Sensitive Topics in
Group and Individual Interviews
MICHAEL D. KAPLOWITZ
Department of Resource Development, Michigan State University, 311A Natural Resources, East
Lansing, MI 48824, U.S.A., e-mail: email@example.com
Abstract. The relative strengths of focus groups and individual interviews have been more the sub-
ject of speculation than systematic research. This study statistically tests the notion that participants
in focus groups and individual interviews equally raise sensitive topics for discussion. Ninety-seven
year-round residents from the Chelém Lagoon region in Yucatán, Mexico participated in 1 of 12
focus groups or 19 individual in-depth interviews. A professional moderator using the same dis-
cussion guide concerning the shared mangrove ecosystem conducted the sessions. The 31 sessions
resulted in more than 500 pages of transcripts which were systematically and iteratively coded using
a grounded theory approach. The coded qualitative data were transformed into summary variables
that allowed for statistical analysis and testing of the research hypothesis. The study illustrates that
the individual interviews were 18 times more likely to raise socially sensitive discussion topics than
the focus groups. Additionally, the study found the two qualitative methods to reveal complementary,
not substitute, sets of information.
Key words: qualitative methods, data analysis, chi-square test, focus groups, individual interviews,
statistical analysis, sensitive topics.
Increasingly, social scientists in diverse ﬁelds use qualitative methods as compre-
hensive research tools and as important components in designing and implementing
reliable research efforts (Carson et al., 1994; Hoehn and Krieger, 1994; Krueger,
1994; Weiss, 1994; Foddy, 1996; Sudman et al., 1996; Morgan, 1997; Schwarz,
1997; Kaplowitz, 2000). For some time, researchers have been advised to con-
sider using focus group interviews as well as tape-recorded individual interviews
for pretesting and development of survey based studies (Belson, 1981; Bryman,
1988; Mitchell and Carson, 1989; Oppenheim, 1992). Investigators, including
economists and sociologists, repeatedly assert the importance of qualitative ana-
lysis of people’s perceptions of the problem in framing their investigation (Smith,
1993; Foddy, 1996; Kaplowitz, 1999). However, survey researchers’ increasing
recognition of the value of qualitative research methods has apparently not been