Deduction of Correlated Variables As The Determinants of Contraceptive Use in Bangladesh: An Application of Factor Analysis
Bangladesh is facing various problems for high and rapid population growth. The high population growth is a great obstacle to the socio-economic progress of the country. The government has given importance to reduce the rapid growth of population and employing all its efforts for the success of population control and family planning program. This paper has been made an attempt to investigate the determinants of contraceptive use in Bangladesh adopting factor analysis and logistic regression model using the nationwide Bangladesh Demographic and Health Survey data. Specifically, factor analysis is performed as a data reduction technique to extract a small number of factors instead of a large number of correlated variables and then multiple logistic regression method is applied to identify the significant determinants of contraceptive use. The factor analysis extracted two factors viz., current status of marriage, age, & birth (F1)- consisted of the variables respondent’s age, partner’s age, marital duration, and number of living children; and inception status of marriage & birth (F2)- consisted of the variables age at marriage and age at 1st birth. These two factors were used in the subsequent analysis instead of the six demographic variables. The logistic regression analysis indicates that region, type of place of residence, religion, respondent’s working status, partner’s occupation, exposure to media, involvement with NGO, source of FP, ‘current status of marriage, age, & birth’ and ‘inception status of marriage, & birth’ have had significant influence on contraceptive use in Bangladesh.
Keywords- Contraceptive use, Data reduction, Factor Analysis,