Collecting demographic data in Bangladesh : evidence from tape - recorded interviews

This report analyzes the transcripts of 220 such interviews - tape-recorded during the 1975-76 Bangladesh Fertility Survey (BFS) - in order to enhance understanding of some puzzling findings, describe interview dynamics, and illustrate the use of tape-recording in evaluating a demographic survey. After examining the use of tape-recording in demographic surveys and describing the rationale and methodology of the present study, the authors analyze BFS respondents" dating of their births, first marriages, and pregnancies. They then examine survey questioning about contraceptive knowledge and use and about fertility preferences. The authors find that the transcript material makes vividly clear the lack of knowledge of dates and ages in Bangladesh, and that there is little reason to believe that different interviewing techniques would have yielded more precise data. On the other hand, suspiciously low reported fertility and infant mortality for the 5 years preceding the BFS may be due in part to interviewers" failure to probe for unreported births (and subsequent infant deaths). Moreover, both interviewers and respondents were uncomfortable during questioning on knowledge and use of contraception, and respondents had difficulty supplying numeric responses to questions about family size desires. Recommendations are made regarding questionnaire design and interviewer strategies.