Linear Discriminant Modeling of Wage Nonfarm Employment between Women and Men in Rwanda

Anatole Mulindwa, Joseph K. Mungatu, Marcel Ndengo

Abstract: The ultimate goal of the study was to find out statistical linear model which predictor variables exactly discriminate or separate women and men groups in wage non-farm employment sector in Rwanda. Linear discriminant method was used with numerical data given by third EICV conducted from 2010-2011 published by NISR in 2012. Discriminant analysis assigns observations to one of the pre-defined groups based on the knowledge of the multi-attributes. For this study I had a single classification variable as sex (male and female) that were divided into two groups of male workers and female workers in non-farm works and the distribution with each group was multivariate normal. The research’s sample was limited to the age between 18 and 65 years old by which women and men who are engaged in the wage nonfarm employment sector. This implied that 7,353 individuals belonged to the actual sample size with 3,772 (51.3%) women and 3,581 (48.7%) men. Majority of respondents were between 18 and 32 years old. 80% of the respondents had been to school and the level of Diploma is at 1%, Bachelor with 0.7%. 55% of the NFE workers are in the trade businesses. In the Non-Farm Employment sector. The SPSS was used to perform tests including the ANOVA test, test of variance, test of equality of group means, the Box’s M test, the Wilks’ Lambda test and Canonical discriminant analysis. Dependent variable was the sex type of male and female which was categorical and independent variables were: Type of non-farm activity (enterprise group), Education level, Income, Income-Unit of time, Expenditure, Expenditure-Unit of time, Duration, Urban/Rural location, and Poverty. The best predictors variables of the discrimination between women and men in the NFE sector were: Education level, Income, Income-Unit of time, Expenditure, Expenditure-Unit of time, Urban/Rural location, and Poverty and the weak predictor variables were: Duration in the business and Industry group of jobs. Keywords: discrimination, women, men, nonfarm employment sector and statistical linear model. Title: Linear Discriminant Modeling of Wage Nonfarm Employment between Women and Men in Rwanda Author: Anatole Mulindwa, Joseph K. Mungatu Phd, Marcel Ndengo Phd, International Journal of Mathematics and Physical Sciences Research ISSN 2348-5736 (Online) Research Publish Journals

Vol. 4, Issue 2, October 2016 – March 2017

Citation
Share : Facebook Twitter Linked In

Citation
Linear Discriminant Modeling of Wage Nonfarm Employment between Women and Men in Rwanda by Anatole Mulindwa, Joseph K. Mungatu, Marcel Ndengo