Abstract

The IAB Establishment Panel is regularly supplemented by refresher samples.
At present, the refresher samples are taken with the help of specially developed software written in perl. In order to create greater transparency and simplify the maintenance and adaptation of the sampling software, the necessary routines are to be translated into the R programming language. Due to the relatively high distribution of R among young scientists and the comparatively low learning effort required to understand the language, a promising solution for drawing refresher samples from the IAB Establishment Panel is to be implemented in this way.
Furthermore, the sampling design of the refresher samples is to be optimised. The design of the refreshment sample includes a stratification by industry sector and company size. The aim is to improve the allocation of the sample size to the strata in terms of the quality of central statistics produced using the establishment panel data. In this context, the weighting strategy of the establishment panel will also be reviewed with regard to its methodological suitability and practicability and, if necessary, adapted. In sum, the measures described should help to further ensure and improve the quality of statistics based on the farm panel through a robust and sustainable methodology.