Analysing the labour market through a web data source: the skill dynamic based on online job advertisements

Abstract

Online advertisements in job portals and company sites have great potentials for job market analyses, providing detailed information on the jobs and skills required by employers. This new source of data – available thanks to advances in web crawling technologies, machine learning and big data techniques – represents an important use case of the ESSnet Web Intelligence Network project, whose main objective is to foster the integration of web data in official statistics. At European level, Online Job Advertisements (OJAs) are centrally collected, and stored in a Data Lab jointly developed by European Centre for the Development of Vocational Training (CEDEFOP) and Eurostat for producing new statistics on job vacancies. OJAs data provide a lot of information on job position characteristics and requirements by enterprises not collected through the official surveys on job vacancies. The objective of this work is to present the potentialities of OJA data in the analysis of the labour demand. In particular, the paper shows some analyses on Italian skill demand, over a period of four years (2019-2022), focussing on relevant changes at national and macro regional level, by economic activity sectors and major occupational groups.