Bayesian population reconstruction of female populations for less developed and more developed countries

We show that Bayesian population reconstruction, a recent method for estimating past populations by age, works for data of widely varying quality. Bayesian reconstruction simultaneously estimates age-specific population counts, fertility rates, mortality rates, and net international migration flows from fragmentary data, while formally accounting for measurement error. As inputs, Bayesian reconstruction uses initial bias-reduced estimates of standard demographic variables. We reconstruct the female populations of three countries: Laos, a country with little vital registration data where population estimation depends largely on surveys; Sri Lanka, a country with some vital registration data; and New Zealand, a country with a highly developed statistical system and good quality vital registration data. In addition, we extend the method to countries without censuses at regular intervals. We also use it to assess the consistency of results between model life tables and available census data, and hence to compare different model life table systems.     

Empirical bayes estimation of demographic schedules for small areas

In this article, we analyze empirical Bayes (EB) methods for estimating small-area rate schedules. We develop EB methods that treat schedules as vectors and use adaptive neighborhoods to keep estimates appropriately local. This method estimates demographic rates for local subpopulations by borrowing strength not only from similar individuals elsewhere but also from other groups in the same area and from regularities in schedules across locations. EB is substantially better than standard methods when rates have strong spatial and age patterns. We illustrate this method with estimates of age-specific fertility schedules for over 3,800 Brazilian municipalities.     

The Level and Age Pattern of Mortality in Bandafassi (Eastern Senegal): Results From a Small-Scale and Intensive Multi-Round Survey

The data collected in the Bandafassi demographic study in Eastern Senegal, a small-scale intensive and experimental follow-up survey of a population of about 7,000 in 1983, were analysed to derive an estimated life table. The use of multi-round surveys, combined with anthropological methods to estimate ages and collect genealogies, has resulted in unusually reliable data. Taking into account the uncertainty of the estimates due to the small size of the population, mortality was high, with life-expectancy at birth close to 31 years; a pattern of infant and child mortality close to that observed in other rural areas of Senegal, with a very high level or mortality between ages six months and three years; a seasonal pattern in child mortality with two high-risk periods, the rainy season and the end of the dry season; an adult mortality pattern similar to that described in model life tables for developed countries; no significant difference by sex or ethnic group. The Bandafassi population study and a few similar studies suggest that one possible way to improve demographic estimates in countries where vital registration systems are defective would be to set up a sample of population laboratories where intensive methods of data collection would continue for extended periods.     

An Application of Bayesian Methods to Small Area Poverty Rate Estimates

Efforts to estimate various sociodemographic variables in small geographic areas are proving difficult with the replacement of the Census long-form with the American Community Survey (ACS). Researchers interested in subnational demographic processes have previously relied on Census 2000 long-form data products in order to answer research questions. ACS data products promise to begin providing up-to-date profiles of the nation’s population and economy; however, unit- and item-level nonresponse in the ACS have left researchers with gaps in subnational coverage resulting in unstable and unreliable estimates for basic demographic measures. Borrowing information from neighboring areas and across time with a spatiotemporal smoothing process based on Bayesian statistical methods, it is possible to generate more stable and accurate estimates of rates for geographic areas not represented in the ACS. This research evaluates this spatiotemporal smoothing process in its ability to derive estimates of poverty rates at the county level for the contiguous United States. These estimates are then compared to more traditional estimates produced by the US Census Bureau, and comparisons between the two methods of estimation are carried out to evaluate the practical application of this smoothing method. Our findings suggest that by using available data from the ACS only, we are able to recreate temporal and spatial patterns of poverty in US counties even in years where data are sparse. Results show that the Bayesian methodology strongly agrees with the estimates produced by the SAIPE program, even in years with little data. This methodology can be expanded to other demographic and socioeconomic data with ease.     

A Flexible Bayesian Model for Estimating Subnational Mortality

Reliable subnational mortality estimates are essential in the study of health inequalities within a country. One of the difficulties in producing such estimates is the presence of small populations among which the stochastic variation in death counts is relatively high, and thus the underlying mortality levels are unclear. We present a Bayesian hierarchical model to estimate mortality at the subnational level. The model builds on characteristic age patterns in mortality curves, which are constructed using principal components from a set of reference mortality curves. Information on mortality rates are pooled across geographic space and are smoothed over time. Testing of the model shows reasonable estimates and uncertainty levels when it is applied both to simulated data that mimic U.S. counties and to real data for French départements. The model estimates have direct applications to the study of subregional health patterns and disparities.     

Making the Housing Unit Method Work: An Evaluation of 2010 Population Estimates in Florida

The housing unit (HU) method—in which population estimates are derived from estimates of occupied HUs—is the most commonly used method for making small-area population estimates in the United States. It is widely used because it is conceptually simple, can utilize a wide variety of data sources, can be applied at virtually any level of geography, and often produces reliable estimates. Yet the HU method is more nearly a general approach to population estimation than it is a specific methodology. In this paper, we describe and evaluate several data sources and estimation techniques that can be used in applying the HU method. Using a set of county and subcounty estimates produced by the Bureau of Economic and Business Research (BEBR) at the University of Florida for April 1, 2010, we analyze errors by population size and growth rate, calculate errors for each of the three components of the HU method, and investigate the accuracy of estimates based on several different data sources and estimation techniques. We compare the accuracy of the 2010 estimates with previous BEBR estimates and estimates produced by the U.S. Census Bureau. We conclude that although some data sources and estimation techniques work better than others, the HU method can be tailored to produce reliable population estimates for a wide variety of geographic areas.     

Revising Infant Mortality Rates for the Early Twentieth Century United States

Accurate vital statistics are required to understand the evolution of racial disparities in infant health and the causes of rapid secular decline in infant mortality during the early twentieth century. Unfortunately, U.S. infant mortality rates prior to 1950 suffer from an upward bias stemming from a severe underregistration of births. At one extreme, African American births in southern states went unregistered at the rate of 15 % to 25 %. In this study, we construct improved estimates of births and infant mortality in the United States for 1915–1940 using recently released complete count decennial census microdata combined with the counts of infant deaths from published sources. We check the veracity of our estimates with a major birth registration study completed in conjunction with the 1940 decennial census and find that the largest adjustments occur in states with less-complete birth registration systems. An additional advantage of our census-based estimation method is the extension backward of the birth and infant mortality series for years prior to published estimates of registered births, enabling previously impossible comparisons and estimations. Finally, we show that underregistration can bias effect estimates even in a panel setting with specifications that include location fixed effects and place-specific linear time trends.     

Estimation of local demographic variation in a flexible framework for population projections

A cohort component projection of local populations based on sex and single year of age offers great value for planning local services, but demands data beyond the detail available. Local fertility, mortality and migration schedules by age and sex must be estimated sensitively to local variation if the results are to be of greater value than simpler methods of projection. Two approaches are compared, using data for the recent past: (a) direct estimation of local area age-specific schedules of fertility, mortality and migration based on data available to the national statistical agency; (b) graduation of national schedules using only local area population estimates by age, total numbers of births, and total numbers of deaths; age-specific migration is indirectly estimated from successive population estimates. These two approaches are compared with a projection using the same rates for each area. The three projections have been implemented for electoral wards in the Fife local government area of Scotland, using the flexible framework provided by POPGROUP software. Persuasive local population projections based on standard data for standard areas are feasible without the regular publication of migration flows.     

Infant mortality in the Marshall Islands

Levy and Booth present previously unpublished infant mortality rates for the Marshall Islands. They use an indirect method to estimate infant mortality from the 1973 and 1980 censuses, then apply indirect and direct methods of estimation to data from the Marshall Islands Women's Health Survey of 1985. Comparing the results with estimates of infant mortality obtained from vital registration data enables them to estimate the extent of underregistration of infant deaths. The authors conclude that 1973 census appears to be the most valid information source. Direct estimates from the Women's Health Survey data suggest that infant mortality has increased since 1970-1974, whereas the indirect estimates indicate a decreasing trend in infant mortality rates, converging with the direct estimates in more recent years. In view of increased efforts to improve maternal and child health in the mid-1970s, the decreasing trend is plausible. It is impossible to estimate accurately infant mortality in the Marshall Islands during 1980-1984 from the available data. Estimates based on registration data for 1975-1979 are at least 40% too low. The authors speculate that the estimate of 33 deaths per 1000 live births obtained from registration data for 1984 is 40-50% too low. In round figures, a value of 60 deaths per 1000 may be taken as the final estimate for 1980-1984.     

Estimating Mortality Level for Small Populations: An Evaluation of a Pair of Two-Census Methods

A pair of two-census methods of estimating mortality levels are tested with simulated census data. The populations considered range in size from 250 to 1500 individuals of each sex; censuses were taken at intervals of five and ten years. In general, the methods are resistant to bias, and yield variances similar in magnitude to those obtained using vital registration data and life table techniques for censored data. The two-census methods represent a substantial improvement over the techniques of mortality estimation previously available for small populations, since two reliable censuses are more likely to be available for these populations than complete vital registration.     

Estimating infant mortality trends from child survivorship data

Summary Brass's procedure for estimating mortality from census or survey data on numbers of children born to women by age group and numbers of children surviving is generalized to allow the estimation of mortality trends. The new procedure is applied to data for Costa Rica and peninsular Malaysia. The resulting infant mortality rate estimates are compared with rates calculated from vital registration figures. The comparisons suggest, surprisingly, that the estimates derived from statistics for women aged 30-50 are not noticeably inferior to those derived from those for women aged 20-30. This suggests that the common practice of disregarding statistics for women aged over 30 or 35 years may be a mistake. Figures are presented which suggest that estimates based on women aged less than 20 are likely to be very seriously biased because of differential infant mortality by age of mother at birth.     

The impact of incomplete geocoding on small area population estimates

Small-area population estimates are often made using geocoded address data in conjunction with the housing-unit method. Previous research, however, suggests that these data are subject to systematic incompleteness that biases estimates of race, ethnicity, and other important demographic characteristics. This incompleteness is driven largely by an inability to complete georeference address-based datasets. Given these challenges, small-area demographers need further, and to date largely unavailable, information on the amount of error typically introduced by using incompletely geocoded data to estimate population. More specifically, we argue that applied demographers should like to know if these errors are statistically significant, spatially patterned, or systematically related to specific population characteristics. This paper evaluates the impact of incomplete geocoding on accuracy in small-area population estimates, using a Vintage 2000 set of block-group estimates of the household population for the Albuquerque, NM metro area. Precise estimates of the impact of incomplete geocoding on the accuracy of estimates are made, associations with specific demographic characteristics are considered, and a simple potential remediation based on Horvitz-Thompson theory is presented. The implications of these results for the practice of applied demography are reviewed.     

An analysis of indirect mortality estimation

Abstract In this paper the robustness of Brass's child-survivorship indirect mortality estimation technique is investigated. An analytical method is developed for studying the error or bias caused in indirect mortality estimates by poor data, badly chosen model functions, and specific demographic assumptions that are often violated in practice. The resulting analytical expressions give insight into the rationale of indirect methods, the conditions under which they are robust, and the magnitude of errors that occur when specific assumptions are violated.     

An Evaluation of Population Estimates in Florida: April 1, 2000

The housing unit (HU) method is the most commonly used method for making small-area population estimates in the United States. These estimates are used for a wide variety of budgeting, planning, and analytical purposes. Given their importance, periodic evaluations of their accuracy are essential. In this article, we evaluate the accuracy of a set of HU population estimates for counties and subcounty areas in Florida, as of April 1, 2000. We investigate the influence of differences in population size and growth rate on estimation errors; compare the accuracy of several alternative techniques for estimating each of the major components of the HU method; compare the accuracy of 2000 estimates with that of estimates produced in 1980 and 1990; compare the accuracy of HU population estimates with that of estimates derived from other estimation methods; consider the role of professional judgment and the use of averaging in the construction of population estimates; and explore the impact of controlling one set of estimates to another. Our results confirm a number of findings that have been reported before and provide empirical evidence on several issues that have received little attention in the literature. We conclude with several observations regarding future directions in population estimation research.     

An analysis of indirect mortality estimation

In this paper the robustness of Brass's child-survivorship indirect mortality estimation technique is investigated. An analytical method is developed for studying the error or bias caused in indirect mortality estimates by poor data, badly chosen model functions, and specific demographic assumptions that are often violated in practice. The resulting analytical expressions give insight into the rationale of indirect methods, the conditions under which they are robust, and the magnitude of errors that occur when specific assumptions are violated.     

Estimating adult mortality levels from information on widowhood

Summary A range of indirect techniques has been developed for mortality estimation in societies lacking adequate vital registration records. Information on orphanhood has been widely used as an estimator of adult mortality, with generally plausible results. Doubts have remained, however, about potential biases, and the method is less satisfactory for the estimation of male mortality. Information on widowhood, or more strictly the survival of first spouse, has several possible advantages over information on orphanhood. Model first marriage functions and model life tables are used to calculate proportions widowed of first spouse, for both females and males, by marital duration and by age. These proportions widowed are then related to life table survivorship probabilities to provide weighting factors for the conversion of observed proportions widowed into estimates of survivorship probabilities. The application of the method is illustrated with data collected by the 1974 post-enumeration survey of Bangladesh, with apparently encouraging results.     

Maternal mortality in South Africa: lessons from a case study in the use of deaths reported by households in censuses and surveys

South Africa is unique in being a developing country which has asked questions on pregnancy-related deaths in both its 2001 census and 2007 household survey, and monitors maternal and pregnancy-related mortality through vital registration and a confidential enquiry into maternal deaths. These sources of data provide a wide range of estimates of maternal mortality for the country. This paper examines these estimates to assess to what extent the differences between them are due to data deficiencies, methodological deficiencies or definitional differences. The results show that since maternal deaths are relatively rare it is fairly difficult to establish the maternal mortality rate with a great degree of accuracy in a setting where data are less than perfect. They also show that to some extent the differences are due to differences and errors in processing of data but that pregnancy-related mortality should not be treated as synonymous with maternal mortality. However, after adjustment, pregnancy-related mortality from vital registration was comparable with the level that may be expected using several alternative approaches, while the rate reported by households in census and surveys was about double that from vital registration. Nonetheless, all the data indicate an upward trend in maternal mortality that is in keeping with the impact of the HIV/AIDS epidemic, which is likely to have contributed to the discrepancies.     

Models for the estimation of the probability of dying between birth and exact ages of early childhood

Abstract This paper develops two models, each of which is designed to estimate the probability of surviving from birth to selected exact ages of early childhood: namely ages two, three and five. The models are designed for use in areas with deficient registration systems. They require, as input, statistics which can be derived from retrospective data supplied by census or survey respondents. The first model, the age model, converts statistics on the proportion dead of children ever born to women in age groups 20-24, 25-29 and 30-34 into estimates of q2, q3 and q5. The second model, the marriage model, converts statistics on the proportion dead of children ever born to women of five-year marriage duration intervals into these estimates. The models can be used independently or simultaneously. These models were developed from data generated by a large number of empirical fertility and mortality schedules. Regression analysis was used to determine the parameter values of the relationships specified, and several sets of equations for estimating values of qa, for a = 2, 3 and 5 comprise the final product of the paper. It should be noted that the conceptual basis for the models was first derived by William Brass. The data generated for the regression analysis provided an opportunity to test the original Brass estimated model. We are able to report that the model performed well over the wide range of fertility and mortality conditions included in the test.     

Estimates of vital rates for the Canadian Indians, 1960–1970

The present paper is an attempt to evaluate the registered data on Canadian Indians collected by the Department of Indian Affairs and Northern Development and to prepare vital rates for 1960–1970 using the adjusted data. A cursory examination of registered data for the purpose of developing various demographic indices and for making future estimates of population indicates certain anomalies that call for a careful appraisal of the data. The main problem is the inconsistency in the reporting of births, due largely to the late registration of births. One plausible reason for late registration may be the increased outward movement of Indians from their reserves. Indirect methods are used to adjust the number of births and infant deaths reported annually since 1960. On the basis of the adjusted data, vital rates for the Canadian Indians are calculated for the period 1960–1970. The crude death rate decreased from 10.9 in 1960 to 7.5 in 1970. The infant mortality rate registered a drastic decline, from 81.5 deaths per 1,000 births in 1960 to 34.9 in 1970. During this same time period the birth rate also declined, from 46.5 to 37.2.