Social Research Methods: Qualitative and Quantitative Approaches

(Brent) #1
SURVEY RESEARCH

EXPANSION BOX 11

Naïve Assumption Model
of Survey Interviews


  1. Researchers have clearly conceptualized all variables
    being measured.

  2. Questionnaires have no wording, question order, or
    related effects.

  3. Respondents are motivated and willing to answer
    all questions asked.

  4. Respondents possess complete information and can
    accurately recall events.

  5. Respondents understand each question exactly as
    the reseacher intends it.

  6. Respondents give more truthful answers if they do
    not know the hypotheses.

  7. Respondents give more truthful answers if they
    receive no hints or suggestions.

  8. The interview situation and specific interviewers
    have no effects on answers.

  9. The process of the interview has no impact on the
    respondents’ beliefs or attitudes.

  10. Respondents’ behaviors match perfectly their verbal
    responses in an interview.


about which respondents have difficulty expressing
their thoughts.
Most professional survey researchers still rely
on standardized interviewing and question the valid-
ity of conversational interviewing. They believe
interviewer effects will distort or bias respondent
answers. However, both approaches to interview-
ing have their defenders. Advocates of a standard-
ized interview approach believe more refined survey
question wording can resolve any respondent mis-
interpretations. Advocates of conversational inter-
viewing emphasize the fluid nature of social
interactions and the different social realities or
understandings held by socially diverse respon-
dents. These advocates say that the goal is to create
a common interpretation of the survey researcher’s
intent behind a question, not to repeat the same
words in a question. To achieve a common inter-
pretation among diverse respondents, an interviewer
may have to ask some respondents the question in
different ways. Only a highly trained, socially adept

interviewer who has a deep understanding of the
researcher’s intent in each survey question may be
able to reach a shared understanding of that intent
with many diverse respondents. We can trace the
cause of the standard versus the conversational
interview disagreement to the assumptions of the
positivist versus interpretative approaches to social
science.^60

The Role of the Interviewer
Interviews to gather information occur in many set-
tings. Employers interview prospective employees,
medical personnel interview patients, mental health
professionals interview clients, social service work-
ers interview people who are needy, reporters inter-
view politicians and others, police officers interview
witnesses and crime victims, and talk show hosts
interview celebrities (see Expansion Box 12, Types
of Nonresearch Interviews). Survey research inter-
viewing is a specialized type of interviewing. As
with most interviewing, its goal is to obtain accurate
information from another person.^61
The interview is a short-term, secondary social
interaction between two strangers with the explicit
purpose of one person obtaining specific informa-
tion from the other. The social roles are those of the
interviewer and the interviewee or respondent.
Interaction takes the form of a structured conversa-
tion in which the interviewer asks prearranged ques-
tions and the respondent gives answers, which the
interviewer records. It differs in several ways from
ordinary conversation (see Table 5).
Interviewers often find that respondents are
unfamiliar with a survey respondent’s role and
“respondents often do not have a clear conception
of what is expected of them” (Turner and Martin,
1984:282). As a result, respondents may substitute
a role with which they are familiar (e.g., an intimate
conversation or therapy session, a bureaucratic
exercise in completing forms, a citizen referendum
on policy choices, a testing situation, or a form of
deceit in which interviewers are try to entrap
respondents). Even for a well-designed, profes-
sional survey, follow-up studies found that only
half of respondents understand questions exactly
as intended by researchers. Respondents often
Free download pdf