Brady T. West, Frauke Kreuter, Mark Trappmann
Is the Collection of Interviewer Observations Worthwhile in an Economic Panel Survey? New Evidence from the German Labor Market and Social Security (PASS) Study

Journal of Survey Statistics and Methodology, 2014: 2, Heft 2, S. 159-181
ISSN: 2325-0992

Although interviewer observations have potential as auxiliary sources of information on key survey variables, questions about their quality temper enthusiasm for their use in survey estimation and responsive survey design. This study considers the utility of two interviewer observations (household income and household receipt of unemployment benefits) collected in a panel survey: the German Labor Market and Social Security (PASS) study. We find that the ability of the interviewer observations to accurately identify these household features is not as high as that of prior-wave survey reports on these features, but the observations do tend to capture accurate information for households with changing socioeconomic status over time (where prior-wave reports may be inconsistent with current-wave reports). The observations add modest predictive power to models for key survey variables that also include survey reports on related variables in prior waves, but this predictive power may be limited by relatively high error rates and variance in observation quality among interviewers. Finally, estimates based on panel households only improve slightly when including the observations in nonresponse adjustments, which is likely due to the inability of the observations to predict response propensity (given a relatively low attrition rate for the panel households). Implications for practice and directions for future research in this area are discussed in conclusion.