On the Validity of Fertility Analysis for Jordan.

The principal concerns of this study are the analysis of fertility levels, trends and differentials in Jordan's 1976 Fertility Survey. In addition, the study examines the association between infant mortality and fertility behavior and investigates the effect of reporting errors on this relationship. The analysis demonstrates that carefully conducted household surveys can achieve as good estimates of current fertility as detailed maternity history surveys. Estimates from both types of surveys are similar, and indicate exceptionally high levels of current fertility. An analysis of marital fertility by birth cohorts provides convincing evidence that there are omissions of live births and that there is event misplacement (earlier events are pushed toward the interview date). However, for the decade preceding the survey, events seem to be correctly reported and the analysis shows that marital fertility remains fairly constant except for a modest decline among older women. Duration-specific fertility rates by marriage cohorts show an insignificant fertility differential before and after the 1967 War between the populations of the East Bank and the occupied territories (West Bank and Gaza Strip). The study reveals a significant fertility differential by age at marriage, educational level, place of residence and contraceptive use; at longer durations of marriage the rates are lower for women marrying late, women not using contraception, women with more education and with urban residence. The differential fertility by age at marriage holds true for all durations and all marriage cohorts. For the total sample, marriage cohorts since 1962 show no significant fertility differential by cohort; however, among earlier marriage cohorts a firm judgement is not always possible due to omission and misplacement of events occurring in the distant past. After controlling for socioeconomic and demographic variables through multiple classification analysis we found a strong positive association between infant mortality and fertility behavior; women with experience of infant deaths tend to make up for the loss of children. Reporting or recording deaths as having occurred after infancy tends to underestimate this association. However, the bias is significant only when considerable number of deaths are so misreported.Ph.D.DemographyUniversity of Michiganhttp://deepblue.lib.umich.edu/bitstream/2027.42/158709/1/8204732.pd