Evaluation of the Mexican fertility survey 1976-77

The report presents detailed findings on age reporting, nuptiality, fertility, and mortality. Age reporting was as accurate as possible for such surveys. Nuptiality data - which are significantly less affected by response error than in comparable surveys in other countries - show an increase in the mean age at first marriage occurring among women born since the mid-1940"s. Marriage history data clearly warrant further and more refined analysis. EMF maternity histories provide reasonably reliable results (a fertility rate of 5.55) which significantly contribute to documenting a decline in fertility. However, data on changing fertility among older ages may be exaggerated as a result of event displacement by less educated respondents. The data regarding infant and child mortality are remarkably consistent and provide convincing evidence that infant mortality is substantially higher than was previously thought. Evaluation of EMF data uncovered interesting weaknesses in the traditional data bases. There seems to be a significant bias in the census results on marital status and number of children ever born to women of childbearing age. Vital registration statistics seem to be affected by severe underreporting of infant deaths in the first month of life and by marked distortions in the shape, if not in the level, of fertility. Finally, the extra effort made by EMF to collect full marital and pregnancy histories has provided more reliable results than those found in the Contraceptive Prevalence Survey. The report includes 25 tables, five figures, and a list of 14 references (1966-80).