|Trinidad and Tobago||-0.1||-0.4|
|Sources: Calculations based on Tables 12.2 and 12.3.|
We now turn to examining socio-economic differentials in wanted fertility levels. The averages in Table 12.3 and Fig. 12.3 show that at the time of the WFS the least educated women wanted 3.6 births compared to 2.4 births among the most educated women, a differential of 1.2 births. Eight to ten years later this differential had narrowed, with the least educated women wanting 2.7 births, as compared with 2.1 births among the most educated women, a differential of only 0.6 wanted births. A similar narrowing in differentials in wanted fertility can be observed in the comparison between rural and urban women. At the time of the WFS, the NWTFR for urban women was 2.6 births compared with 3.4 for rural women, a difference of 0.8 births. By the time of the DHS, however, the differential in wanted fertility had declined to 0.4 births. A similar reduction in differentials in the reported desired family size also occurred (data not shown). These findings indicate that in the countries examined here, women of different social strata are converging toward a narrow range of fertility preferences.
Perhaps the most striking result in Fig. 12.3 is the steep gradient in unwanted fertility according to socio-economic group. The unwanted TFR of women with no education exceeded by more than two births the rate among women with more than primary education in both survey rounds. Similarly, rural women had more than twice the unwanted fertility of urban women. Clearly, women in the higher social strata have much better control over their fertility and are more successful at implementing their fertility preferences than rural women or women with low levels of education.
A general conclusion to be drawn from this analysis is that most of the variation in actual fertility at the time of both WFS and DHS is not due to variation in the number of children wanted but instead due to differential success in controlling fertility to desired levels, which is consistent with prior findings by Lightbourne