Changes in sex ratios observed in census and population estimates have drawn the attention of demographers and media in recent years and are articulated in the media as a problem with ‘missing men’. The underlying causes of these changes have been hotly debated in a number of forums with mixed results. Consequently, alternative data sources have been sought to throw further light on the problem. One such potential alternative data source is the Linked Employer-Employee Dataset (LEED). This data source has very good coverage of the age groups of greatest interest – those aged 20–49 years – and may contribute to an understanding of the underlying patterns, trends, and sources of the perceived problem.
The paper discusses the relationship between census, estimated resident, and LEED populations. The paper concludes that LEED exhibits sex ratios that are consistent with expectations. Changes due to annual migration flows are reflected similarly in both LEED and estimated resident populations (ERPs). However, further work would be required to fully understand the relationship between the LEED population and census populations, in particular the biases that are introduced by the different coverage of the various data sources