[Journal:] CBN Journal of Applied Statistics [ISSN:] 2476-8472 [Publisher:] The Central Bank of Nigeria [Place:] Abuja [Volume:] 6 [Year:] 2015 [Issue:] 2 [Pages:] 1-17
This study considered the nature of employment that women engage in as a multi-categorical response. A multinomial logistic model with geo-additive predictors was used to examine the determinants and geographical variations using data from the 2008 Nigeria Demographic and Health Survey. Diffuse priors were assumed for modelling fixed effects, Bayesian p-spline for the nonlinear smooth functions and intrinsic conditional autoregressive prior for the spatial effects. Results showed that while a north-south divide existed in the likelihood of women engaging in all-year employment against not working, an east-west divide was found in seasonal/occasional jobs. Other important factors found to be significantly associated with employment status included women's age, educational level, marital status, sex of household head, and type of place of residence. Policymakers need to develop appropriate strategies to address the observed imbalance in the spatial distributions of women employment status in the country.
Geo-additive model spatial analysis Bayesian women employment