SURVEY RESEARCH
EXPANSION BOX 15
Interviewing: Positivist and Feminist Approaches
respondents alike to control the communication sit-
uation and yield a uniform measure. Ordinary inter-
action contains built-in features to detect and correct
misinterpretation; it relies on nuance and give and
take. People achieve social meaning in ordinary con-
versation by relying on clues in the context, adjust-
ing the interaction flow to specific people involved,
and building on a cultural frame (often based on race,
class, gender, region, or religion). The fluid inter-
action of ordinary conversation is self-adjusting
because different people do not always assign the
same meaning to the same words, phases, and ques-
tions. For example, men and women report health
differently. A man saying he is in excellent health
In this chapter, we have mostly considered the posi-
tivist approach to survey research interviewing. In the
ideal survey interview, the interviewer withholds her
or his own feelings and beliefs. The interviewer
should be so objective and neutral that it should be
possible to substitute another interviewer and obtain
the same responses.
Feminist researchers approach interviewing very
differently. Feminist interviewing is similar to quali-
tative interviewing. Oakley (1981) criticized positivist
survey interviewing as being part of a masculine par-
adigm. It is a social situation in which the interviewer
exercises control and dominance while suppressing
the expression of personal feelings. The interview is
manipulative and instrumental. The interviewer and
the respondent become merely the vehicles for
obtaining the objective data.
The goals of feminist research vary, but two
common goals are to give greater visibility to the
subjective experience of women and to increase the
involvement of the respondent in the research
process. Features of feminist interviewing include the
following:
A preference for an unstructured and open-ended
format
A preference for interviewing a person more than
once
Creation of social connections and building a trust-
ing social relationship
Disclosure of personal experiences by the interviewer
Encouragement of female skills of being open, recep-
tive, and understanding
Avoidance of control and encouragement of equal-
ity by downplaying professional status
Careful listening; interviewers become emotionally
engaged with respondents
Respondent-oriented direction, not researcher ori-
ented or questionnaire oriented
Encouragement of respondents to express them-
selves in ways they are most comfortable—for exam-
ple, by telling stories or following digressions
Creation of a sense of empowerment and an esprit
de corps among women
means something different from a woman answer-
ing the same question with the same response. By
standardizing human interaction, the survey inter-
view strips away features in ordinary conversation
that provide self-correction, promote the construc-
tion of a shared meaning among different people,
and increase human mutual understanding.^73
Pilot Testing and Cognitive Interviews
It is important to pilot test survey interviews and
questionnaires prior to implementation. Systematic
study of pilot tests in the survey process and models
of cognitive processing has helped us better under-
stand the survey process. We see that the process of
answering survey questions has several steps: inter-
pret and comprehend the question, retrieve relevant
information, integrate and evaluate the information,
and select a response category. A recent area of study
is cognitive testing or cognitive interviewingin
which we study how respondents answer questions
in pilot test situations.^74
Cognitive interviewing A technique used in pilot
testing surveys in which researchers try to learn about
a questionnaire and improve it by interviewing respon-
dents about their thought processes or having respon-
dents “think out loud” as they answer survey questions.
