Fast online surveys without sampling frames are becoming increasingly important in survey research. Their recruitment methods result in non-probability samples. As the mechanism of data generation is always unknown in such samples, the problem of non-ignorability arises making vgeneralisation of calculated statistics to the population of interest highly questionable. Sensitivity analyses provide a ...
We introduce a selection model-based imputation approach to be used within the Fully Conditional Specification (FCS) framework for the Multiple Imputation (MI) of incomplete ordinal variables that are supposed to be Missing Not at Random (MNAR). Thereby, we generalise previous work on this topic which involved binary single-level and multilevel data to ordinal variables. We apply an ordered probit ...
We introduce a selection modelâ€based multilevel imputation approach to be used within the fully conditional specification framework for multiple imputation. Concretely, we apply a censored bivariate probit model to describe binary variables assumed to be missing not at random. The first equation of the model defines the regression model for the missing data mechanism. The second equation ...
This article documents the number of target persons participating in the panel surveys of the National Educational Panel Study (NEPS) as well as the number of respondents who temporarily dropout and of those leaving the panel (attrition). NEPS comprises panel surveys with six mutually exclusive starting cohorts covering the complete life span. Sample sizes, numbers of participants and temporary as ...