Towards a shared infrastructure for online job advertisement data

摘要

Following the increasing penetration of the internet, the number of websites that advertise jobs is growing. The European Centre for the Development of Vocational Training (Cedefop) and the ESSnet Big Data have engaged in parallel projects to assess the feasibility of using online job advertisements (OJA) for labour market analysis and job vacancy statistics. After an initial feasibility study finalised in 2016, Cedefop is developing a Pan-EU system providing information on skills demand present in OJA, which will be operational by 2020. The ESSnet has focussed on statistics that can be derived from OJA and entered into a second phase in November 2018 aiming at creating the conditions for a larger scale implementation of the use of OJA in official statistics. This paper builds on experiences gathered in both projects and identifies opportunities and limitations of using OJA for the above-mentioned purposes. In addition, it discusses the feasibility of creating a joint system for processing and analysing OJA data based on discussions that have taken place in the past two years between Cedefop and the ESSnet Big Data on both projects. In this respect, this paper outlines a possible partnership between Cedefop and the European Statistical System to create and manage a unique source of OJA data that would serve multiple uses in the domain of labour market analysis and official statistics. It presents potential types of (statistical) data and variables based on the information contained in OJAs at European, national and regional levels. Data limitations linked to OJAs nature and specificities will be stressed, too. The paper concludes that there is high potential for combining institutional efforts and creating a joint data collection and processing system on OJA and intends to feed a discussion on the feasibility and the implications of creating a European system for OJA, which can serve European and national needs.