Evaluating socioeconomic factors for crime against women in developing countries: A data-centric statistical learning approach

Esther Jose, Sayanti Mukherjee and Jose Swaminathan

Socio-Economic Planning Sciences, 2025, vol. 101, issue C

Abstract: Women are often targeted in crimes of sexual violence, trafficking, and domestic abuse, especially in developing countries. There are two types of risk factors for women being victims of such violence. Personal risk factors include attributes or features of the woman’s self or identity, such as how old she is, how educated she is, and whether she is married. There is a second set of factors that we call “regional” risk factors, which include the attributes or characteristics of a region (defined as a state or union territory) such as how electrified it is, how many colleges it has, or how many roads it has. We offer insights on regional risk factors and how they influence rates of crime against women in that region. We also address the challenge of under-reporting and present insights into factors that could reduce under-reporting. We use a suite of advanced machine learning techniques to identify and evaluate the socio-economic and political risk factors for high rates of both reported and adjusted crime against women in a region. We establish our research framework with a case study conducted in India, using data from different states and union territories from 2004–2020. We consider 23 factors, including the financial condition of the state, the ruling political party, access to electricity, access to education, employment rate, and birth rate. Our results show that high access to education, low gender disparity in education, low poverty, and increased household access to electricity are positively correlated with reduced crime against women. We also observe that under-reporting is more often a problem in poorer regions, regions where higher percentages of women are illiterate than men, and regions where household access to electricity is low. While policymakers cannot easily change personal risk factors, these regional risk factors can be addressed explicitly by government agencies, institutions, or leaders.

Keywords: Crime against women; Gender-based violence; Risk analysis; Machine learning; Crime rate under-reporting (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0038012125001041
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:soceps:v:101:y:2025:i:c:s0038012125001041

DOI: 10.1016/j.seps.2025.102255

Access Statistics for this article

Socio-Economic Planning Sciences is currently edited by Barnett R. Parker

More articles in Socio-Economic Planning Sciences from Elsevier
Bibliographic data for series maintained by Catherine Liu ().