Social Networks, Female Unemployment, and the Urban-Rural Divide in Turkey: Evidence from Tree-Based Machine Learning Algorithms

M.G. Celbis*

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

1 Citation (Web of Science)

Abstract

This study takes a novel, algorithmic approach for understanding the underlying mechanisms related to the employment status of individuals. Using the data from the most recent survey of the International Social Survey Programme (ISSP) on Turkey, the present study examines how social connectivity and location play a role in the prediction of employment status through the use of two tree-based modern machine learning techniques, namely random forest, and extreme gradient boosting. We obtain a wide array of observations, with gender being the most prominent finding when periphery and rural locations are considered.

Original languageEnglish
Pages (from-to)73-93
Number of pages21
JournalSosyoekonomi
Volume29
Issue number50
DOIs
Publication statusPublished - 1 Oct 2021

JEL classifications

  • j68 - Mobility, Unemployment, and Vacancies: Public Policy

Keywords

  • Machine Learning
  • Unemployment
  • Turkey
  • Rural
  • Urban
  • gender inequality

Cite this