Age at first marriage for women in Matlab Bangladesh: the role of biological and social determinants.

The purpose of this study was to examine the determinants of age at marriage in Bangladesh. Determinants considered important were age at menarche nutritional status participation in the informal labor market vital statistics and socioeconomic factors. The average age at menarche in Bangladesh is 15.8 years due to poor nutritional status and health. The average age at marriage is young at 17.3 years. Background information is given on cultural and demographic marriage patterns. Data were obtained from the International Center for Diarrheal Disease Research in Matlab on 382 women in 1976 and 1989-90. Generalizability of findings is limited because of sample selectivity due to migration and exclusion of subjects without an age at menarche. The data set is unique however in providing data on nutritional status in adolescence age at menarche and ages at marriage in a developing country. Time to marriage is measured in fractions of months because of the short time frame. Z-scores are used for anthropometric measurement. Socioeconomic variables include religion education and household wealth (land radios wristwatches traditional quilt kerosene lamp). Descriptive statistics and Cox proportional hazards models were provided in the analysis. The results confirm that both social and biological factors determine the timing of marriage in rural Bangladesh. Menarche is an important predictor and does not affect the waiting time to marriage but does act as a step function. Weight is positively related to the hazard of marriage. The "U" shaped relationship between education and age at marriage was not observed in the multivariate models. Household wealth was not important in determining marriage age. There is the suggestion that womens income generation may delay age at marriage. When menarche and nutritional status are added to the socioeconomic models there is not a great reduction in the hazard. Even though increased nutritional status may increase fertility policy should still promote health gains for females; physical development may be a more important factor in maternal and child outcomes than maternal age.