Weighting methodology for the Norwegian Labour Force Survey from 2021 onwards

Abstract

A new revision in the weighting procedure of the Norwegian Labour Force Survey since the last one made in 2018 was required due to the change in the target age group for employment statistics, the introduction of the new sampling design in 2021 and the new weighting requirements with the new EU legislation for labour force surveys within the EU. Compared to the revision in 2018, the main weighting method via model-calibration has remained unchanged for the age group 15 − 74. How ever, significant changes occurred in the calculation of selection probabilities, which is not straight forward, and initial weights to calibration due to the new sampling design, changes in the data struc ture and the new weighting requirements by the new EU legislation. A linear calibration method was introduced for those within age groups 0 − 14 and 75+. In order to fulfil the individual- and household-level requirements simultaneously and to ensure same weights for all individuals within any given household, an integrative calibration approach was introduced to be used for the calcula tion of yearly household weights. Several models were evaluated in this respect, and a final model was chosen by considering robustness and gain in precision.