Gauly, Britta, Jessica Daikeler, Tobias Gummer, and Beatrice Rammstedt. 2020. "What’s my wage again?: Comparing survey and administrative data to validate earning measures." International Journal of Social Research Methodology 23 (2): 215-228. doi: http://dx.doi.org/10.1080/13645579.2019.1657691.
Haven't you always wondered how valid respondents' income information is? Four authors of Gesis have also asked themselves this question and published an article on the presence of measurement errors in the income data from survey data. To quantify the error, they use administrative earnings data, which are assumed to be less error-prone, as a benchmark. The results reveal measurement error in survey data, which correlates with survey respondents' sociodemographic characteristics. To test how this may affect empirical findings, they regress the earnings information from both data sets on respondents' characteristics. As the results show statistically significant differences in the resulting regression coefficients, they conclude that measurement error in earnings information from surveys may result in biased results on the relationship between respondents' characteristics and earnings.