Updated methods for the use of official personnel statistics

The Swedish Research Council has a mandate to conduct research policy analyses and give advice to the Government on research policy issues. It is therefore crucial that the personnel statistics we use in the analyses are as accurate as possible. But the education register, where the official statistics on doctoral degrees are shown, does not capture all doctoral degrees of researchers who were awarded their degrees outside Sweden. As ever more researchers in Sweden have doctoral degrees awarded in another country, this has led to one third of the postdocs being shown in the education register as lacking doctoral degrees in 2022.

“It is positive to see increasing international mobility, but we need to address the consequence that the education register has worse coverage for doctoral degrees”, says Linnea Wickström Östervall, Analyst at the Swedish Research Council.

Foreign degrees are not automatically reported

She and her colleagues at the Swedish Research Council realised that the trends of employee numbers within the different employment categories – primarily postdocs, but also other career appointments, and to some extent also other categories – were impacted by the fact that foreign doctoral degrees, as opposed to Swedish ones, were not reported automatically in the education register. If these degrees are to be included in the register, then active measures are needed from the parties involved.

To achieve a more accurate picture of the research and teaching personnel, the Swedish Research Council has now worked out two changes to how we use the personnel statistics. The revised methods are reported in the publication “Vetenskapsrådets användning av den officiella personalstatistiken” (“The Swedish Research Council’s use of official personnel statistics”), and will as far as possible be used in future analyses and statistics summaries.

“It is important for us to be transparent and explicit about the methods we apply when reporting personnel statistics in our publications. We also hope that the new methods can be useful for others who analyse the official personnel statistics,” says Anders Weström, Head of the Swedish Research Council’s Research Policy Analysis and Evaluation Unit.

The Swedish Research Council will not be revising previously published analyses and summaries. For presentations of time series, we will have to make an assessment in each case whether a break in the time series is problematic. Any deviation from the new methods will in that case be shown.

Read the report on the Swedish Research Council’s use of official personnel statistics (in Swedish)