The Performance of Multiple Imputation in Social Surveys with Missing Data from Planned Missingness and Item Nonresponse

This paper studies the quality of estimates from multiple imputation for the case of social survey data that combines planned missing data with missing data from conventional item nonresponse by survey participants. To this end, the paper uses a Monte Carlo simulation study on real data from the German Internet Panel. In this data, missingness is simulated based on item nonresponse with different mechanisms and proportions of item nonresponse as well as different proportions of planned missing data. Our results show that item nonresponse can jeopardize the quality of estimates after multiple imputation especially when the total amount of missing data from both sources is high or when there is a considerable proportion of item nonresponse that is missing not at random. Therefore, from an imputation perspective, survey designers should incorporate their expectations about item nonresponse on each variable when designing surveys with planned missing data.