Using Multinomial Logistic Regression to Examine the Relationship Between Children’s Work Status and Demographic Characteristics

Authors : Erkan ARI

Pages : 77-93

View : 23 | Download : 19

Publication Date : 2016-01-01

Article Type : Research Paper

Abstract :This study aims to examine the relationship between children’s work status and their demographic characteristics using multinomial logistic regression. To this end, data collected by TUIK’s Turkish Statistical Institute “Child Labor Force Survey, 2012,” conducted with the participation of 27,118 children, were used. At the first stage of the analysis, eight independent variables on the demographic characteristics of the participants were examined using the chi-square test of independence, the variable that was not significant was removed, and the subsequent analyses were conducted using the remaining seven variables. The validity of the model in the study was examined using maximum likelihood estimation, and the model was significant. Odds ratios of the variables in the model were calculated, and twocategory comparisons were made on the basis of the reference category using odds ratio coefficients. In comparisons 1 and 2, odds ratio coefficients for the variables rural/urban, gender, age group, household size, literacy, school attendance, and level of education of the head of household were significant
Keywords : Children, Working Children, Multinomial logistic regression