Introduction and Application of CART Model to Classify Ideal Number of Children for 15-49 Years-Old Women Semnan providence

The main objective of this paper is introducing and applying a tree model to classify demographic events which has been used for classifying ideal number of children from a project entiteld "Study of the Marriage and Fertility behaviors of Married 15-49 Years Old Women in Semnan Province in Iran, 2012" survey. The results showed that Classification and Regression Trees (CART) model by using Gini coefficient with estimated prior probability had sufficient accuracy to predict the ideal number of children class. The results of fitted model to classify ideal number of children for 405 married women according to predictor variables of age at first marriage, marriage type, education level, job status, place of birth and birth cohort stated that the job status had different role among birth cohorts. Similarity of fertility opinions and intentions of the first and second birth cohorts was as another result of this article. Employed women comparing to unemployed women in first birth cohort had more ideal number of children. Also, age at first marriage among first and second birth cohorts for those women whose birth place were rural areas had important effect in their ideal number of children.