Academic journal article Demographic Research

Reverse Survival Method of Fertility Estimation: An Evaluation

Academic journal article Demographic Research

Reverse Survival Method of Fertility Estimation: An Evaluation

Article excerpt

(ProQuest: ... denotes formulae omitted.)

1. Introduction

In order to study changes in fertility levels and trends in less statistically developed countries, demographers have developed a series of estimation techniques based on data from census counts and household surveys (Brass 1975, Moultrie et al. 2012, United Nations 1983). Since the launch of the World Fertility Surveys (WFS) program in the late 1970s (and especially since the implementation of the Demographic and Health Surveys (DHS) program), population specialists have mostly relied on the ever-growing body of household surveys that have collected full birth history to derive fertility estimates in these countries. Yet, as a recent study has shown (Avery et al. 2013), alternative methods of fertility estimation can return very consistent fertility estimates using only basic demographic information.

Among the existing methods of fertility estimation the reverse survival method is one of the most parsimonious. Based on population data by age and sex collected in one census or single-round survey, the method consists in 'reverse surviving' those no longer present in the population of a given age in order to derive the number of births that occurred n years ago, using a set of probabilities of child and adult survivorship and age-specific fertility rates (ASFRs). The reverse survival method of fertility estimation is very similar to the own-children method of fertility estimation (Cho et al. 1986), but its data requirement is even lower.

Notwithstanding its simplicity, the reverse survival method of fertility estimation has seldom been applied. However, the method can be applied to a large body of existing and easily available population data, which has remained largely under- exploited. In contexts where limited demographic data are available the sole reliance of the method on age and sex distribution makes it of prime interest. Furthermore, the comparison of reverse survival fertility estimates with alternative fertility figures from multiple data sources (vital registration, sample survey) provides an additional tool to evaluate the quality of population data. At a time of increasing reliance on sample survey data to estimate fertility for less statistically developed countries, the reverse survival method allows revisiting and highlighting the contribution of both historical and contemporary population census data to the study of fertility. Finally, given its coverage (the whole national population), the population census remains a unique source for deriving fertility estimates for different geographic levels and/or socio- economic and cultural groups. However, given that the reverse survival method of fertility estimation is sensitive to migration, it should be stressed that its use at the local level may be limited (unless detailed migration data are available).

The present analysis is inspired by a series of studies of the own-children method of fertility estimation, the aim of which was to determine the effect of different factors on the fertility estimates (Cho 1973; Retherford, Chamaratrithirong, and Wanglee 1980; Abbasi-Shavazi 1997). The analysis pursued two main objectives. Using the Excel template FE_reverse_4.xlsx, provided with Timæus and Moultrie (2012),2 first the consistency of the reverse survival method of fertility estimation to estimate fertility levels and trends was investigated. Second, the sensitivity of the method to the effect of different types of data quality issues as well as erroneous assumptions (i.e., age patterns of fertility, levels and age patterns of mortality, effect of international migration, and age and sex population distribution) on the reverse survival estimates of fertility was assessed. In their presentation of the method, Timæus and Moultrie (2012) discussed briefly the use of erroneous mortality estimates, the importance of the quality of the age distribution, and the potential effect of international migration. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.