Abstract
Statistics Norway established the Labor Force Survey (LFS) in 1972, and it has been conducted quarterly ever since. The LFS measures the population's participation in the labor market and provides comprehensive information on unemployment, employment, people outside the labor force, temporary employees, underemployed and other subgroups that are not captured by register-based statistics. This makes the LFS one of the most important sources of information about conditions in the Norwegian labor market.
This module collects information on receiving a pension, whether individuals combine work and pension, as well as the reasons for working or not. The purpose of the module is to map labour market participation among individuals aged 50 to 74.
The annual variables in the LFS are asked to individuals during the 2nd and 6th interview rounds.
In the LFS, individuals can be exempted from interviews starting from the 3rd round if, for example, they are retirees and no longer working. This is because the primary purpose of the LFS is to analyse the active labour market. Responses from the 2nd interview for those exempted are carried forward into subsequent rounds (passive sample). The consequence of this approach, particularly for this pensions module, is an undesirably high sample loss among retirees. Therefore, Statistics Norway impute for this module.
Stratified nearest neighbour imputation was used. For 2023, the stratification was specially tailored to impute responses to questions about pensions and labour market participation for individuals aged 50 to 74. The 2023 stratification is based on sex, age groups (15-49, 50-69, 70-89), marital status, and registered disability.
The imputation process involves copying all responses from the 2nd interview for 2-year and 8-year variables from a donor within the same stratum, with the smallest age difference (nearest neighbour), to a recipient in the passive sample when they would have been interviewed for the 6th time.