Multilevel modeling of determinants of fertility status of married women in Ethiopia

摘要

The main objective of this study is to investigate the determinant factors of fertility status of married women in Ethiopia and to examine the reasons for variations of fertility across regions of Ethiopia based on data on 7052 married women obtained from Ethiopian Demographic and Health Survey (EDHS, 2011). Multilevel binary logistic regression models on fertility status of married women were employed. This study revealed that the random intercept and fixed slope model fits the data significantly better than the other multilevel logistic regression models. The results confirmed that woman's education level, sex of household head, being visited by family planning worker last twelve months, child loss experience, woman's occupation, religion and age of woman at first birth were found to be significant determinants and also contributing factors for variation in fertility status of married women among the regions of Ethiopia. In random intercept model the overall variance of constant term was found to be statistically significant implies that women with the same characteristics in two different regions have different fertility status: that is, there is a clear region effect. In this study multilevel model best fit the data as compared to single level model.