Abstract

The paper presents a multiple linear regression and hierarchical cluster analysis investigating the gender gap in tertiary education within the selected 30 European countries. The gender gap (GG) within a particular age group is defined as the percentage of men who have attained tertiary education minus that of women. Within 2023, 31% of EU citizens between 15 and 64 had finished tertiary education. In 2023 the females’ share was 34% and for men in the same age group, it was 28%. The proportion of men with tertiary education increased over the last ten years but at a slower skip than for women, widening the gender gap in education.
The main variable under study performs the dependent variable in the regression analysis, and that is Gender gap in tertiary educational attainment, 15-64 years population, 2023. Based on the most recent data, the independent variables considered were Gross Domestic Product (GDP) per capita in Purchasing Power Standards (PPS) (Indices, EU27_2020 = 100), Percentage of females by tertiary educational attainment (levels 5-8), 15-64 years, Employment rates (%) for females, and tertiary educational attainment (level 5-8), Gender gap in employment, 2023 (% males minus % females), and General government expenditure, as percentage of gross domestic product (GDP) for Education, 2022. Two valid representative multiple regression models were estimated. No model assumptions were violated, as the conducted diagnostics showed. After exploratory data analysis, several mild outliers were found, but there were no need to remove any country data from further analysis. However, several leverage points and influential values were found. Luxembourg and Ireland stand out due to their high GDP per capita and relatively strong education/employment indicators. Romania and Italy are outliers because of their low female tertiary education attainment. Greece has lower economic performance and employment rates, which make it unique in the model.
Finally, hierarchical clustering was completed using Ward linkage and Euclidean distances on standardized data for two combinations of variables associated with the main variable under study, and clusters of similar countries were recognized. Cluster 1 of Leading Gender Gap-Reducing Performers, gathers countries that have excelled in creating environments that support women in education and employment, with small gender gaps reflecting progressive societal attitudes and effective policy interventions. These are top performers in gender equality, especially in terms of educational attainment and employment for women. Cluster 2 of Moderate Gender Gap-Reducing Performers, collects countries that are making noticeable strides but still face obstacles in achieving full gender equality, requiring continued focus on education and labor reforms. Cluster 3 of Lagging Gender Gap-Reducing Performers, gathers countries, that are struggling to close the gender gap, despite some being economically advanced potentially due to ingrained cultural norms or insufficient policy measures targeting gender equality. These countries may need more targeted interventions to support women’s access to education and employment opportunities.

Keywords: Gender Gap in Education, Gender Gap in Employment, Multiple Linear Regression Analysis, Euclidean distances, Ward Linkage, Hierarchical Clustering.