Intermediate variables and fertility: a case study of Ghana (West Africa).

Using data obtained from the Danfa Comprehensive Rural Health and Family Planning Project in Ghana this study assesses the relative importance of various "intermediate" variables in explaining the variations in fertility among the different segments of rural Ghanian society. The intermediate variables are 1) age of entry into unions 2) proportion of women never entering unions 3) reproductive period spent after or between unions 4) voluntary abstinence 5) involuntary abstinence 6) coital frequency 7) involuntary fecundity and infecundity 8) use or nonuse of contraception 9) voluntary fetal mortality and 10) involuntary fetal mortality. These 10 variables fall under 3 categories: 1) intercourse variables (numbers 1-6) 2) contraception variables (78) and 3) gestation variables (9-10). The author uses stepwise multiple regression to examine the relationship between fertility and the intermediate variables. Results show that fertility of wives who married at a younger age is significantly higher than those whose 1st marriage took place at a late age. The larger the proportion of pregnancies resulting in fetal deaths the higher is the mean birth interval. Wives having a history of previous marriages had a higher mean birth interval than those with relatively stable marriages. Mean birth intervals are longer for polyganous wives. Fertility performance of wives who had ever used contraception is higher than for those who had never used contraception. The type of conjugal relationship is not statistically significant. Of all the intermediate variables the inverse relationship between age at marriage and fertility is most pronounced and consistent.