Census/Survey Consistency of Reports of Childbearing: The Case of Taiwan

Demographers are perpetually challenged by the use of less-than-perfect data. Experience has shown that different sources of data and methods of collection may make comparisons over time and across sources difficult. Taiwan is often considered the "gold standard" among developing countries in terms of the quality of its demographic data: the population tends to report ages and dates accurately and the resulting data reports tend to be comprehensive and useful. This study uses data from two of Taiwan's censuses--1966 and 1980--and a variety of survey data collected between 1967 and 1989 to examine the consistency and validity of fertility reporting among the Taiwanese population. The authors found overall high consistency in reports of children born across censuses and surveys, but preferred surveys for estimating child mortality. Results indicate that the censuses provide useful information for estimating fertility, even without supplemental survey data. Unfortunately, the question on the number of children born was removed from the 1990 Census of Taiwan.