Gender inequalities at work in Southern Europe

Lara Maestripieri, Yijun Ren, Alessandra Guglielmi

Research output: Other contribution


Despite a long-term trend towards reduction, the gender gap in employ-
ment keeps standing in Southern Europe. Numerous potential causes
have been individuated, such as the household configuration, the human
capital of the women, or the institutions that regulate the labour mar-
ket. Less is know about the role of the locality. This paper explores what
covariates influence women’s access to labour markets, and whether it is
unevenly distributed across different countries and regions in the South-
ern Europe. The analysis is based on the dataset round 9 (2018) from
the European Social Survey. We focus on the following countries avail-
able in the dataset: Cyprus, Italy, Spain and Portugal. Italy and Spain
are further differentiated into vulnerable and affluent regions accord-
ing to the regional GDP in 2018. We apply a regression model for the
binary response that is the indicator of having been doing paid work
for the last seven days of each individual in the sample. We adopt the
Bayesian approach, in order to derive conclusions via a whole proba-
bility distribution, i.e., the posterior of all parameters, given data. The
statistical goal is the selection of the most important covariates for
access to labour market, focusing on gender differences. Our analysis
finds out that the individual characteristics are mediated by house-
hold composition. Even though a higher education increases women’s
employment, the presence of children and having an employed part-
ner reduce such involvement. Moreover, a larger gender gap is detected
in vulnerable regions rather than in affluent ones, especially in Italy.
Original languageEnglish
Number of pages28
Publication statusPublished - 3 Nov 2022


  • Bayesian approach
  • Covariate selection
  • Gender gap
  • Labour market
  • Regression models


Dive into the research topics of 'Gender inequalities at work in Southern Europe'. Together they form a unique fingerprint.

Cite this