SURVEY RESEARCHquestionnaire while a person with disabilities was
in the same room, they reported higher levels of
happiness. Apparently, respondents felt compara-
tively better off due to the physical presence of the
person with the disability compared to situations
in which there was no immediate reminder of
the life situations of others.^69 A respondent who
answers identical questions differently depending
on features of an interviewer threatens representa-
tive reliability.
Cultural Meanings and Survey Interviews
Research into survey errors and interview bias has
advanced information about how people create
meaning and achieve cultural understanding.^70 We
are troubled when a word has different meanings
and implications depending on the social situation,
who speaks it, how it is spoken, and the social dis-
tance between the speaker and listener. Survey
research is complicated when respondents misin-
terpret the nature of survey research and seek clues
for how to answer in the wording of questions or
subtle actions of the interviewer. Moreover, “it is
important not to lose sight of the fact that the inter-
view setting is itself distinct from other settings in
which attitudes are expressed, and hence we should
not expect to find complete congruence between
attitudes expressed in interviews and in other social
contexts.”^71
We face a dilemma: An interviewer who strives
to act in a neutral and uniform way reduces the type
of bias that causes unreliability because of individual
interviewer behavior, yet such attempts cause other
problems according to interpretive or critical social
science researchers, including feminist researchers
(see Expansion Box 15, Interviewing: Positivist and
Feminist Approaches).^70
Nonpositivist researchers argue that meaning
is created in social context; therefore, standard sur-
vey question wording will not produce the same
meaning for all respondents. For example, some
respondents express feelings by telling stories
instead of answering straightforward questions with
fixed answers. Nonpositivist researchers advocate
the collaborative encounter modelof the survey
situation. This model views all human encounters as
highly dynamic, complex mutual interactions in
which even minor, unintended forms of feedback
(e.g., saying hmmm, laughing, smiling, nodding)
have an influence, and suggests conversational
interviewing. The collaborative encounter model
also allows interviewers to incorporate information
offered by respondents in response to fixed-choice
questions that the standardized interview prohibits
or treats as an error because it does not correspond
to a preset, standardized format.
According to the collaborative encounter
model, in complex human interactions, people add
interpretative meaning to simple questions. For
example, my neighbor asks me the simple question,
“How often do you mow your lawn?” I could inter-
pret his question in the following ways:How often do I personally mow the lawn (versus
having someone else mow it for me)?
How often do I mow it to cut grass (versus run my
lawnmower over it to chop up leaves)?
How often do I mow the entire lawn (versus cutting
the quick-growing parts only)?
How often do I mow it during an entire season,
a month, a week?
How often do I mow it most seasons (versus last
year when my lawnmower was broken several
times and it was very dry and the grass grew
less, so I did not mow it as frequently)?Within seconds, I make an interpretation and give
an answer, but the open-ended, ongoing interaction
between myself and the neighbor permits me to ask
for clarification and for several follow-up questions
that help us arrive at mutual understanding.
A survey interview interaction differs from
ordinary conversation. The standard survey research
interview is an artificial interaction that treats diverseCollaborative encounter model A particular survey
interview in which the respondent and interviewer
work together to reach the meaning of the survey
question as intended by the researcher and produce
an accurate response to it.