A widely applicable sampling method for spreading the survey burden in business surveys

Rotating Random Number Sampling is proposed as a flexible method to evenly distribute survey burden caused by a set of business surveys over enterprises without requiring aligned stratifications.

The sampling coordination methods used by Statistics Netherlands to spread the survey burden across businesses currently require that stratifications of business survey sampling designs are aligned. This report proposes Rotating Random Number Sampling (RRNS) to better support spreading of survey burden when different surveys use non-aligned stratifications.

Simulation results indicate that RRNS reduces "hotspots" of high burden in cross-sectional surveys, performing better in case of non-aligned stratifications and equally well in case of aligned stratifications compared to the system currently in use. However, unlike the current method RRNS cannot account for different burden weights and in particular cannot account for the cumulative burden of long-term panel membership.

Boonstra, H. J., M. Smeets (2026). A widely applicable sampling method for spreading the survey burden in business surveys. Discussion paper. Statistics Netherlands, The Hague/Heerlen.


Downloads