A Cross Sectional Study of Iranian Women and Sex Preference for Children

Children sex preferences may have significant effects in fertility behavior which is an influential component of population dynamics and could control the population size, structure, and composition. One of the core concerns of researches and policy makers is to study these determinants. The main objective of this study is to investigate Iranian women’s sex preference and its highly influential factors through applying Classification & Regression Trees algorithm as a practical classification method.Materials: A cross-sectional study during 2014 was conducted to collect demographical data of 1250 Iranian women in age 15-49. To classify sex preferences for children, age, educational level, place of residence, and difference number of siblings for women, were nominated as predictors. 71 percentages of women’s sex preference have been classified correctly (the accuracy of the model=0.71). In extracted decision tree, women's age, educational level and difference number of siblings were significant. One of the vital results indicated that educated Iranian women in different age cohorts are in favor of having girls. Classification & Regression Trees is an effective and easy to interpret non-parametric method which builds classification tree for predicting dependent variables.