The Combined Effects of Prior-Wave Item Nonresponse and Perceived Burden on Subsequent-Wave Nonresponse in a Longitudinal Survey

Music City Center 

This paper examines the problem of wave nonresponse in longitudinal surveys. Wave nonresponse reduces the sample size needed to estimate trends and changes over time. As a result, it is critical to identify factors contributing to wave nonresponse. This study focuses on two factors related to wave nonresponse based on survey literature: item nonresponse and the perception of burden at a prior wave. According to the response continuum model (Yan and Curtin, 2010), item nonresponse at a prior wave is predictive of nonresponse to the subsequent wave. For example, panelists with a higher level of item nonresponse at Wave 1 will have a higher likelihood of not responding to Wave 2. The response burden framework (Yan and Williams, 2022) predicts a positive relationship between perceived burden and wave nonresponse; that is, panelists with a higher level of perceived burden are less likely to participate in the next round of interviews. I will examine the separate effects of prior-wave item nonresponse and perceived burden on nonresponse to the next wave of interviews. Additionally, I will investigate the combined effects of these two factors on wave nonresponse using data from a longitudinal web survey. The findings will have important practical implications and can be used to inform adaptive designs for longitudinal surveys to reduce wave nonresponse.

longitudinal survey

item nonresponse

response burden

wave nonresponse