Spatial-temporal evolution of digital skills in the EU countries

Abstract

Digital skills, integral to the functioning of the digital economy and information society, show temporal and spatial variations measured by various indicators. In this article, we assess the spatial and temporal evolution of digital skills under the influence of key factors and domains in the EU countries from 2015 to 2021. Applying spatial autocorrelation analysis, robust geographical heterogeneity and consistent spatial patterns in digital skills are outlined, resulting in two ‘high–high’ and ‘high–low’ clusters in the North and Center, and a ‘low–low’ cluster in the South. Using feature importance selection, key indicators within aggregate domains driving digital skills policy are identified. Spatial lag regression analysis highlights the significance of all domains, revealing spatial and spillover effects on digital skills, with the primary influence observed in the social sphere, technology and innovations, and demography domains. Although the ICT infrastructure domain is statistically more significant in our spatial model along with the economy and technology and innovations, its spillover effects appear relatively modest, indicating a corresponding degree of within-country localization. This study contributes to the understanding of the evolution of digital skills by revealing both spatial relationships and temporal dynamics and strengthening spatial digital policy measures in the EU. The spatial coherence of digital policies, the spatial network of technological and innovation centers in both ‘high–low’ clusters and cross-border locations, and improving the social, demographic, and economic profiles of citizens are critical among other measures to improve digital skills in EU countries.