Blurred memory, deliberate misreporting, or “true tales”? How different survey methods affect respondents’ reports of partnership status at first birth
Michaela Kreyenfeld, Max Planck Institute for Demographic Research
Sonja Bastin, Max Planck Institute for Demographic Research
This paper examines the validity of biographical information gathered retrospectively. It draws on data from the German Family Panel (pairfam), which collected information on partnership status at first birth using two different methods. The first method is based on data on partnership and fertility histories collected retrospectively. The second method uses data gathered through the use of a “landmark question” on the respondents’ partnership status when their first child was born. We found that in almost 20 percent of the cases, the information collected using the first method did not correspond with the information collected using the second method. Partnership dissolution and “turbulence” in the partnership biography were strong predictors for discrepancies in the information gathered through the two different survey methods. We concluded by drawing attention to the limitations of the retrospective collection of partnership histories at a time when divorce and separation rates are increasing.
Presented in Session 43: Bias in demographic surveys