Effects on Survey Attrition from Increased Respondent Burden: Evidence from the Panel Study of Income Dynamics

Add to Calendar
Date and Time: 
Thursday, February 23, 2017
12:00 pm – 1:30 pm
Narayan Sastry
University of Michigan

A key issue for panel surveys is the relationship between respondent burden and subsequent resistance or attrition. Among the potentially most burdensome activities for panel respondents is participation in a supplemental study, which may, for instance, require a separate interview, a home visit, or other new activity. Knowledge of the potential negative effects of supplemental studies permits an informed assessment of the trade-off between the benefits and costs of these supplemental studies. Also, if the negative consequences are concentrated among certain sample segments, focused efforts could be used to minimize these effects. This study uses data from multiple waves of the biennial Panel Study of Income Dynamics (PSID) from 1997 to 2013 to examine the effects on attrition of selection into the original 1997 PSID Child Development Supplement (CDS). A major challenge in analyzing the effects of supplemental studies on core panel outcomes is that participation in the supplement—including the offer to participate—is rarely randomized to provide a clear comparison group. Although this situation presents a challenge for our analysis, we use several complementary approaches to create appropriate comparison groups to those selected for the supplement.