Women’s unemployment in urban Ethiopia: bayesian multilevel model approach
Abstract Introduction Unemployment is a significant socio-economic problem affecting virtually all countries worldwide, with approximately 172 million people unemployed globally. This staggering figure underscores the urgency for effective policies and strategies to address unemployment and promote...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-07-01
|
| Series: | Discover Sustainability |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s43621-025-01249-y |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract Introduction Unemployment is a significant socio-economic problem affecting virtually all countries worldwide, with approximately 172 million people unemployed globally. This staggering figure underscores the urgency for effective policies and strategies to address unemployment and promote economic stability across nations. In Ethiopia, unemployment is one of the most pressing concerns, particularly as women's potential contribution to economic activity, growth, and well-being remains greatly underutilized, despite women comprising over half of the population. The fundamental root causes of women's unemployment in Ethiopia's urban areas are an intricate web of interdependent issues that require thorough understanding. The objective of this study was to investigate the determinants of women’s unemployment in urban Ethiopia using a Multilevel Bayesian logistic regression model. Methodology This study is based on the Ethiopian Urban Employment Unemployment Survey (UEUS) conducted by the Central Statistical Agency (CSA) of Ethiopia, using a multistage stratified cluster sampling technique. The multilevel Bayesian model incorporates various levels of variation and employs Bayesian inference for estimation and prediction, allowing for the integration of prior knowledge and uncertainty. Parameters were estimated using the Markov Chain Monte Carlo (MCMC) method in MLwiN. Result The analysis of 29,649 women indicated that 26.6% were unemployed, while 73.4% were employed. The Bayesian multilevel random coefficients model showed that women aged 24–34[AOR = 0.3(95% CI − 1.311, − 1.1)], age above 34 [AOR = 0.4 (95% CI − 1.018, − 0.85)], primary education level [AOR = 1.25 (95% CI 1.031, 1.404)], secondary education [AOR = 1.24 (95% CI 0.006, 0.40)], higher education [AOR = 0.38 (95% CI − 1.258, − 0.71)], marital status married [AOR = 1.56 (95% CI 0.367, 0.515)], widowed [AOR = 0.54 (95% CI − 0.78, -0.45)],divorced [AOR = 0.88 (95% CI − 0.256, 0.008)], household size 4-6 [AOR = 1.32 (95% CI 0.197, 0.357)] above 6 family size [AOR = 8.15 (95% CI 0.453, 3.516)], relationship with household head spouse [AOR = 1.43 (95% CI 0.242, 0.477)], children/relative[AOR = 4.34 (95% CI 1.379, 1.559)], non-relative [AOR = 6.2 (95% CI 1.728, 1.926)], training social [AOR = 0.4 (95% CI 0.01, 0.4)],, natural [AOR = 0.55 (95% CI − 0.958, − 0.24)], Engineering [AOR = 0.53 (95% CI − 1.064, − 0.188)], health [AOR = 0.28 (95% CI − 1.736, − 0.81)], Agriculture [AOR = 0.44 (95% CI − 1.629, − 0.046)], other social service [AOR = 0.73 (95% CI − 0.565, -0.058)], duration of unemployment 13–24 month [AOR = 0.24 (95% CI − 1.525, − 1.31)], 25–95 month[AOR = 0.3 (95% CI − 1.302, − 1.083)], above 95 month [AOR = 0.43 (95% CI − 0918, − 0.762)], were statistically significant for womens unemployment in Urban Ethiopia. Conclusion Women in Dire Dawa and the Somali region face higher unemployment rates than those in other urban areas of Ethiopia. The study employed Bayesian multilevel binary logistic regression with random coefficient models, selecting models based on the lowest DIC values. Key factors identified included age, education, marital status, household size, relationship with the household head, training, job search efforts, duration of unemployment, economic sector, challenges of joblessness, reasons for unemployment, and barriers to starting a business. The analysis highlighted significant variances in women’s unemployment across regions, indicating a need for tailored interventions. Increasing access to education and vocational training aligned with women’s interests and labor market demands, partnering with private businesses, and addressing regional challenges through literacy improvement, financial support for startups, and tackling employment disparities. |
|---|---|
| ISSN: | 2662-9984 |