Weights to produce consistent time-series between the new Spanish Migration and Change of Residence Statistics and previous migration statistics

Background: Spanish migration sources are widely used by researchers and international organisations. Thus, consistent time-series are crucial. In 2021, the Spanish National Statistics Institute (INE) replaced the Residential Variation Statistics (EVR) and the Migration Statistics (EM) with the new Change of Residence and Migration Statistics, which included methodological changes. Objective: I evaluate differences between the EMCR and previous sources and produce weights to make the EMCR consistent with the EVR and EM. I publish the weights in an open repository (https://github.com/MiguelGonzalezLeonardo/Weights_EMCR_Spain). Methods: I analyse the percentage differences for internal and international migration between the EMCR and the previous sources at NUTS 2 and NUTS 3 levels in 2021, the only year when the three sources were published simultaneously. I then produce weights by dividing the migration counts from the EVR and EM by those from the EMCR for each spatial unit. Results: I found strong differences between the new source and the previous sources for international migration, especially for emigration, which needs to be calibrated with the weights. The EMCR and EVR register similar numbers of internal migrants. Conclusions: Methodological changes in statistical sources can lead to large differences in the measurement of a social phenomenon. This suggests the need to assess time-series consistency and produce weights for calibration. Contribution: I provide open data with weights to calibrate internal and international migration in the new Spanish migration source and produce consistent time-series with previous sources at different scales. The weights can be used by researchers from different countries and international organisations.