The analysis of how available official statistics can be used to measure the pay gap between women and men for work of equal value involved a number of different activities which are summarised in this section.
Expert dialogue: As part of the information gathering phase and discussion on the implementation of the project, four interviews were conducted in June 2024 with experts, researchers and investigators in the field: two researchers from Finland, two researchers from Norway and one investigator from Sweden. These were supplemented by information gathering and recommendations by a researcher in Iceland. The aims were to gain insights into relevant statistical sources on pay outside of those available through the websites of the national statistical offices, develop an understanding of how the issue of equal pay for work of equal value is discussed in each country, as well as whether it is perceived as an issue on their respective political agendas, and identify any actors driving the issue that the researchers should be aware of for the purposes of the project. Notes were taken during the interviews. It was later determined that more than enough material was available through the national statistical authorities for the purposes of this pilot project and the idea of using other databases was abandoned.
Literature review: Searches and reviews of previous research and grey literature (reports) were carried out continuously during the project. Previous research comparing pay between occupations of equal value with different gender compositions was found to be almost non-existent. Although several different reports, mainly produced by government agencies, mentioned the need to make pay comparisons between jobs of equal value in addition to conducting pay audits that look at pay structures, almost none have dealt with the issue in more detail.
Review of the official occupational classifications of each country: The occupational classifications AML 2010 in Finland, STYRK-08 in Norway and SSYK 2012 in Sweden were reviewed and compared against the DISCO-08 and Ístarf21 classifications in Denmark and Iceland respectively. Comparison of similarities and differences between the occupational classifications in Finland, Norway and Sweden was carried out and documented in Excel, occupation code by occupation code. The occupational classifications of the three countries were compared with the international standard ISCO-08.
The studies of national official statistics were conducted using a stepwise approach. Statistics from 2022 are used throughout, as well as some earlier comparison years. Statistics from 2022 were in many cases the most recently available at the start of the project.
Orientation in existing statistics: The available tables in each country’s statistical databases from the national statistical offices were mapped. One of the authors concentrated on Sweden and Finland, the other on Norway.
The orientation involved examining available statistics by extracting various tables from the databases to ascertain the availability of suitable data on, for example, gender distributions by occupation, average pay in different occupations, pay gaps in different occupations, pay in different sectors, pay by working hours (full/part time) and pay by level of education. Through this process, several gaps were identified, although they differed between the statistical databases of the different countries. As the statistical databases are in many cases not fully comparable, it was decided following the orientation process to mostly use data from only one country to illustrate results.
Pair comparisons: Categorisations by the Swedish organisation Lönelotsarna (Harriman et al., 2023) were used when comparing the pay situation between pairs of male- and female-dominated occupations in which, according to the Lönelotsarna model, work of equal value is performed. These pairs were compared across all three countries. Harriman et al. guided the choice of occupational pairs; the remit for the pilot study did not include a valuation of occupations, but illustrating the shortcomings and applicability of the statistics required some kind of comparative material. While the evaluations by Harriman et al. are not indisputable, they are based on solid work. The pairwise comparisons illustrated mostly similar pay gaps between male- and female-dominated occupations in all three countries. As pay gaps within occupations were also identified, the interaction between intra- and inter-group pay gaps could be analysed.
Special tabulations: As the publicly available statistics in the statistical databases could not address all questions on age, education and part-time employment, specific tables were commissioned, which the statistical offices can supply at a cost. Three tables were ordered from Finland, one on age, one on education – both on levels and fields of education – and one on pay for full-time and part-time work. All tables were gender-disaggregated and based on the four-digit occupational classification.
In Sweden, age could be extracted from publicly available statistics. Data on education related to pay is also publicly available, although not based on occupational classifications, only the classification of education. Thus, only statistics on full-time and part-time pay for women and men in different occupations were requested.
The request for special tables on the Norwegian statistics was cancelled because no information was received as to whether the requested tables could be produced and within what time frame. After a period of waiting, it was decided that the requested Norwegian statistics could not be delivered to enable analysis within the time frame of the project.
Analysing existing and commissioned statistics: The available tables from the statistical databases of the national statistical offices from the respective countries were analysed from a variety of perspectives, alongside the commissioned statistical tables, focusing on comparisons of pay in different (equal) occupations at the three- and four-digit levels of the occupational classifications, pay in different sectors, pay by working hours (full/part time), pay by level of education, pay by age and pay gaps measured using mean and median values.