Feeling observed? : a field experiment on the effects of intense survey participation on job seekers' labour market outcomes

We ran a field experiment to causally identify the effects of intense survey participation on key labour market outcomes. We randomly excluded individuals willing to sign up for the German Job Search Panel, a high-frequency survey with a focus on job search and well-being. Using administrative data on labour market outcomes (e.g., employment, earnings), we find that, on average, survey participation had no effect on labour market outcomes during the year after signing up. Furthermore, there is no strong heterogeneity across subgroups. Overall, this is good news for the validity of survey-based research involving labour market outcomes. We also demonstrate that a comparison of individuals signing up for the survey with individuals not responding to the invitation could have been misleading. Even when controlling for a wide range of observable characteristics, survey participation and the subsequent take up of training programs correlate significantly. This speaks to the importance of experimental research designs in our context.