Full text: Brückner, Hannah: Surveys don't lie, people do?

-13- 
II. THEORETICAL AND EMPIRICAL APPROACHES 
Generally, from a constructivist perspective, social processes intervene between an account 
(read: data) and a presumably existing underlying reality. Woolgar (1983) stresses that the 
constructivist view of science operates under the assumption that accounts (i.e., data) are underde- 
fermined by facts of nature. Two or more alternative accounts can result from the same underlying 
reality (p. 247) depending on the mediating social interaction. Extending this notion to the survey 
setting, one might say that obtained sample means may be underdetermined by population means. 
From this perspective, answers, observations, records-in short, data-reflect outcomes of a complex 
social interaction rather than reflecting what was the case before the production of the data took 
place. " Based on varying levels of formulations of the problem of data, in the final analysis one can 
cast doubts on every finding. Without declaring themselves constructivists, survey researchers try to 
deal with this problem by holding social interaction constant. 
Standardizing questionnaires is only one step in this direction. Not only survey questions and 
their sequence but also probing, feed-back, and other parts of social interaction in the interview are 
prescribed. Another aspect of the process of standardization is the removal of interviewers' personality 
and involvement with the respondent from the interview situation. Neutral behavior is preferred over 
the earlier emphasis on maintaining the rapport' with the respondent (Martin 1983, Groves 1989). 
Similarly, a dispersion of interviews over a large staff of interviewers is preferred over extensive 
employment of interviewers within one survey. 
A great deal of attention has been devoted to the social interaction between respondents and 
interviewers in the survey. Sources of bias related to the interviewer have been identified. We have 
come a long way since the classic study of Rice (1929), who found that destitute respondents 
interviewed by prohibitionists tended to attribute their deprivation to alcohol abuse, while those 
interviewed by socialists tended to blame industrial society. While Rice interprets the large bias found 
in his study as an effect of social desirability, Hyman (1954) points to the critical role of interviewers 
expectations. He shows that early on in the interview interviewers tend to form an image of respon- 
Woolgar (1983) remarks that different 'schools of thought can be classified according 
to their use of words such as 'in addition too or 'instead of in the place of rather in this sentence. 
This paper is based on the assumption that there is a physical reality which-to varying extent- 
shows up in scientific and other accounts. The problem I address below concerns the degree to 
which the flow of events making up a life history is structured according to the 'rules' given by the 
questionnaire which are only one particular way to structure one's story.
	        
Waiting...

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.

powered by Goobi viewer