Spatial distribution and factors of ideal number of children among Ethiopian women: The Case of Ethiopia Demographic and Health Survey

Background: Fertility is one of the important subjects in public health studies which affects populationgrowth.Ideal number of children is one of the main indexes showing the fertility preferences of couples. The main objective of this study was to analyze the spatial distribution and factors of an ideal number of children among Ethiopian women using the multilevel count regression model. Methods The study design was a cross sectional study. About 14003 women’s ages of 15–49 who fulfill the inclusion criterion were considered in this study. The study was evaluated based on different statistical model such as NB regression model and cross level interaction between level-2 and level-1 predictors (model5). In the meantime, a model with the best fit of the data using AIC and BIC was found to be a Multilevel Negative Binomial regression model. Results From this study, 33% of the women had need more INC and the variance of the INC was higher than its mean, this indicates that there is over dispersion so NB regression models were considered to select the best fit for the data. The LRT suggested that, INC visits vary among zones Furthermore, the Ideal number of children was spatial clustered (Global Moran’s I = 0.1439, p < 0.00043). Significant hotspot clusters were found in the Somali region such as in Afder, Shabelle, Korahe, Doolo zone. Conclusion Our study investigated the spatial heterogeneity of an ideal number of children, and its association with women’s age, region, place of residence, women's education, contraception use, religion, family size, spatial variable (Si) and age at first birth were significant predictors of the ideal number of children by women at reproductive age. Significant hotspot clusters were found in the Somali region such as in Afder, Shabelle, Korahe, Doolo zone. Therefore, Spatial distribution of INC different across each zone of Ethiopia.