Infant-child mortality and son preference as factors influencing fertility in Pakistan.

This study uses data from the Pakistan National Impact Survey of 1968-69 to explore the influence of sex composition of live births and surviving children and sex differentials in infant and child mortality through parity 4 on the subsequent fertility behavior of couples. Separate multiple classification analyses were performed using 3 dependent variables: total number of live births, birth intervals, and parity progression ratios. Among the findings that achieved statistical significance, the proportion of female children among survivors was associated with a decreasing number of live births to the date of the survey and was positively associated with the parity progression ratio for parity 2 and 4. The "total number of deaths prior to specific parity" was negatively related to total number of live births at parity 2 but positively related at parity 3 and 4, and at parity 2 it increased the subsequent birth interval. The total number of deaths prior to specific parity was associated with an increasing parity progression ratio for parities 2 and 4 but a decrease at parity 3. The association of the "total number of male deaths prior to specific parity" to total live births and to parity progression ratios was positive for all parities and its association with the birth interval was negative for parity 2. The association of male infant deaths to the total number of live births was positive for parities 2 and 3, its association with birth interval was positive for all parities, and its association with parity progression ratio was positive for parity 4. The major finding of the study was that infant and child mortality has a direct bearing on fertility levels, which is especially pronounced because of the universal practice of prolonged breastfeeding. Couples who lost male children compensated for the loss comparatively more than those losing female children.