A dream job? Skill demand and skill mismatch in ICT

ABSTRACT

We analyse labour demand and mismatch for educational traits in information and communication technology (ICT) occupations, taking into account the areas of education, occupations and skills in demand, and the supply of labour. We based our analysis on almost 40 million online job offers and a CAWI survey of people aged 18–65 in Poland. The analysis uses official classifications, and considers both job-related and transversal skills. To analyse an objective measure of mismatch, we compare job requirements stated in job offers to the declared characteristics of potential labour supply, and present demand and mismatch for skills across occupations. We also model a declared educational mismatch and its determinants, including educational profiles of surveyed participants. In order to further investigate the mismatches in ICT fields of education, occupations and skills, we compare them with the mismatches in other occupational groups. The analysis concludes with direct educational policy recommendations, including the Sectoral Qualifications Framework for ICT occupations in Poland. Our approach shows the complete educational characteristics of skill demands and skill mismatches.

KEYWORDS:

• ICT
• online job offer
• skill demand
• skill mismatch

Disclosure statement

No potential conflict of interest was reported by the author(s).

Contribution

RP directed the research and supervised the project within which the research were conducted. He planned the research and wrote the theoretical part of the article. He wrote the introduction, discussion and conclusion. He conducted the modelling of the declared mismatch and described its results. He was also involved in the process of correcting the article. HC collected the literature. He calculated and wrote descriptive statistics. He prepared databases, including data collection, transformation and calculation of the demand and supply side indicators, as well as mismatch indices. He described the results. He prepared graphical presentation of results. MK provided advice during the analytical process, formulated recommendations at each of its stage, and initiated discussions within the team of authors as to the use of optimal methods and solutions. He corrected the whole article and ensured that the analyses have been done properly.

Notes

1. Gradzewicz et al. (2016) show that the most important source of job vacancies in Poland was the media, mostly Internet (approx. 50% of all vacancies). The second-most frequently used way of disseminating information about vacancies was recommendation (approx. 30%). Data for 2018–2019 from the Balance of Human Capital Study (2020) reveal that 64% of companies advertised their openings through public employment offices in 2018, compared to 51% in 2019; 47% used online job boards in 2018 and 42% in 2019; 41% relied on recommendations from current employees in 2018 and also in 2019; and 33% of companies used internal job postings in 2018, compared to 35% in 2019.

2. The survey was conducted by the research panel of Ariadna (https://panelariadna.pl/), using the questionnaire we delivered. We chose Ariadna since it was determined to ensure the highest quality of research in Poland, based on the following criteria: having a PKJPA Polish certificate of the quality of CAWI surveys, the size of its Internet panel of respondents, its share of registered users, and experience (in terms of time and the number of large surveys conducted). Ariadna also fulfilled the ‘28 Questions to Help Buyers of Online Samples’ (ESOMAR 2012).

3. The novelty of our approach is a method of continuous collection of information on labour demand (job offers). In order to calculate mismatch indices, we also needed data on potential job seekers (via CAWI survey). The CAWI survey was conducted in the beginning of the period of job offer collection. The set of information collected with the CAWI survey was unique. No continuously collected database contained such information. Because the sample and scope of information collected within it was vast, and the survey was very costly, it was impossible to repeat the survey. Hence, our data consists of a time series for labour demand and static (one-period data) information on job seekers. While this might be a limitation, we think that the characteristics of people do not change profoundly over two years. That is why we believe that the comparisons we make represent the true mismatch between the two sides of the labour market.

4. A ‘learned occupation’ means a profession that a person acquired during formal education.

5. https://ec.europa.eu/eurostat/cache/metadata/en/jvs_esms.htm

Additional information

Funding

This work was supported by the Polish Ministry of Science and Higher Education within the Programme DIALOG under Grant Horizontal educational mismatch: a new method of measurement with application to Poland [number DIALOG 0127/2016].

Notes on contributors

Robert Pater

Robert Pater (https://orcid.org/0000-0001-7619-9843) holds the position of the Head of Department of Economics and Finance and Associate Professor of the University of Information Technology and Management in Rzeszow, Poland. He also works as a team leader and expert in forecasting the labour demand and skill demand in the Educational Research Institute in Warsaw, Poland. He specialises in macroeconomics and econometrics. He has been conducting regular research on the vacancy market in Poland with the use of job advertisements, developing this methodology since 2004, and participated in several international projects about job vacancy measurement. His research interests include the cyclical and structural change in the labour market, the vacancy market measurement with big data, and skill demand.

Herman Cherniaiev

Herman Cherniaiev (http://orcid.org/0000-0001-9286-6490) is a Researcher at Department of Economics and Finance of the University of Information Technology and Management in Rzeszow, Poland. He also works as an expert in forecasting the labour demand and skill demand in the Educational Research Institute in Warsaw, Poland. His research interests include machine and deep learning, big data, econometrics and labour market with particular attention to online job vacancies.

Marcin Kozak

Marcin Kozak (https://orcid.org/0000-0001-9653-3108) is a Full Professor at the University of Information Technology and Management in Rzeszow, Poland. A multidisciplinary researcher, he has published in the fields of statistics, agriculture, biology, economics, scientometrics, information sciences, and data analysis and visualisation.