Factors associated with female age at first marriage: An analysis using all waves of the Pakistan Demographic and Health Survey

Abstract

In conventional Muslim societies, fertility occurs within the marital union. Therefore, fertility outcomes depend on females’ age at first marriage (AFM). The present study explores the pattern of AFM in Pakistan, analyses of which are scarce in the literature. We aim to identify the factors associated with the AFM among currently married women in Pakistan. Demographic factors like birth cohort, and socioeconomic and cultural factors such as province and place of residence, education level, whether or not a woman had ever worked before marriage, ethnicity, and husband’s education were studied to explore the pattern of female AFM. Data were taken from the Pakistan Demographic and Health Surveys (PDHSs) of 2012–13 and 2017–18, and a comparison was made with the findings from the earlier PDHSs of 1990–91 and 2006–07. The analysis concentrates on women born between 1941 and 1992, aged between 25 and 49 years during the data collection periods. One-way analysis of variance (ANOVA) was used to assess the difference between the mean AFM for different sub-groups of the population. To identify the covariates that are associated with AFM multiple linear regression models were estimated. We observed a gradually increasing trend in female AFM over time among women born after 1950. The ANOVA results revealed that birth cohort, province, and place of residence, female education level, whether or not a woman had ever worked before marriage, ethnicity, and husband’s education were significantly associated with AFM ( p -values < 0.05). In a multiple regression model, we found that the birth cohort significantly affects the AFM ( p -value < 0.05). Having worked before marriage is associated with a statistically significant one-year rise in the AFM. Interestingly, all other ethnic groups have lower AFM compared with women whose mother language was Punjabi. Education has a highly significant effect on the AFM: the regression results revealed that uneducated females have a mean AFM 4 to 5 years lower than highly educated women. The results also revealed that educated men marry older women as compared to uneducated men. We conclude that the education of females and even males in Pakistan could lead to a rise in the female AFM.