The TFR is probably the most commonly used demographic indicator. It is closely associated with contraceptive prevalence and other indicators of reproductive health such as the maternal mortality ratio. It is a useful indicator of population momentum and a good proxy measure for the success (or failure) of family planning services. The TFR may also be used as a measure of poor physical reproductive health, since high parity (>5 births) represents a high risk of maternal morbidity and mortality.
The main strength of the TFR is that it is a single summary measure that is independent of age structure, unlike the general fertility rate that only partially controls for age structure and the crude birth rate that does not do so at all. It is thus useful for international comparisons and for monitoring trends over time. It should be emphasized, however, that the TFR is a hypothetical measure of completed fertility; in cases of rapid fertility transition its value is primarily illustrative.
As mentioned above, disaggregation of the ASFRs is useful in reflecting the age pattern of fertility, especially in high-risk groups such as adolescents and older women. TFRs are not useful in gauging the direct impact or success of family planning programmes. Family planning programmes can reduce total fertility only by reducing unintended as opposed to intended fertility. Nevertheless, there is strong empirical evidence that high contraceptive prevalence is associated with a low TFR and that increasing contraceptive prevalence is related to lowering the TFR (2).
In general, the TFR is a good summary figure for comparing countries, major population subgroups or trends over time. Nevertheless, distinguishing between real and artificial changes in the TFR can be complicated. Observed differences or changes are not necessarily specific to changes in fertility behaviour. They could be due to numerous factors largely related to the data sources used, data quality, or shifts in the age-specific fertility distribution or incidence of early pregnancy loss.
It is very important that data quality is assessed before ASFRs and TFRs are calculated and interpreted. An awareness of biases resulting from common reporting errors in censuses or surveys and their impact on calculating ASFRs and TFRs is critical for their appropriate interpretation.
Underreporting of births is typically greater for older women and for births that occurred a relatively long time ago. This is a minor problem, however, if information only on births during the last three years is used to estimate ASFRs and TFRs.
A more serious error commonly found in survey data is displacement of births. The typical pattern is a peak in the period 4-9 years prior to the survey and a trough in the five-year period immediately preceding the survey, showing a spurious decline in fertility. Displacement can also occur in the year prior to the survey. It is therefore recommended that births in the last three years be used to estimate ASFRs and TFRs. Census data are also prone to such biases.
Misreporting of women's ages It is advisable to examine the possibility of misreporting of ages by survey or census respondents.
Estimates derived from surveys are prone to large sampling errors. It is therefore essential to provide sampling errors and confidence intervals for the estimated TFRs.
Was this article helpful?