|Modelling Childbearing Desire: Comparison of Logistic Regression and Classification Tree Approaches|
|Arezoo Bagheri1, Mahsa Saadati1|
|1National Population Studies & Comprehensive Management Institute, Tehran, Iran|
Viewed : 541 times
Downloaded : times.
Keywords : Fertility Preferences, Child, Women, Decision Trees, Logistic Regression
|| Related Articles|
Objectives: According to health surveys, population growth and Total Fertility Rate (TFR) in Iran are declining. Economic and social factors, in addition to changing values and attitudes in Iranian society, have had a major impact on fertility decisions and the actions of families, especially women, towards childbearing. This is an important issue for policy makers and many researchers in demography and public health, so factors that affect low TFR need to be reviewed and investigated.
Materials and methods:One of the most applicable classification trees, the Classification and Regression Trees (CART) algorithm, and logistic regression were applied to model the tendency of 4898 women for childbearing in provinces with a TFR lower than the replacement level in Iran. The secondary data has been analysed by Spss 24.
Results:By applying these two approaches, it could be concluded thatlogistic regression despite the CART algorithm suffers from some shortcomings, including the difficult interpretation of three levels of interactions and no specific method for handling outliers. According to the CART results, women’s Children Ever Born (CEB), age and opinion had significant impacts on their desire to have a child. The “10-39-year-old women with CEB 2” and “40-49-year-old women with positive attitudes towards childbearing” groups desired to have more children. However, “women with CEB 3” did not show any tendency for childbearing.
Conclusion: According to the results of this study, adopting policies for changing women’s views on childbearing and creating the necessary resources to prevent delays in marriage could be important steps in altering fertility rates. Another important conclusion is applying CART algorithm as a convenient method to classify demographical data.
Cite By, Google Scholar