Estimating a frequency distribution when the sampling is biased

A simple random sample on a random variable A allows its density to be consistently estimated, by a histogram or preferably a kernel density estimate. When the sampling is biased towards certain x-values these methods instead estimate a weighted version of the density function. This article proposes a method for estimating both the density and the sampling bias simultaneously. The technique requires two independent samples and utilises ideas from mark-recapture experiments. An estimator of the size of the sampled population also follows simply from this density estimate.     

Profile upper Confidence Limits from Discrete Data

When the data are discrete, standard approximate confidence limits often have coverage well below nominal for some parameter values. While ad hoc adjustments may largely solve this problem for particular cases, Kabaila & Lloyd (1997) gave a more systematic method of adjustment which leads to tight upper limits, which have coverage which is never below nominal and are as small as possible within a particular class. However, their computation for all but the simplest models is infeasible. This paper suggests modifying tight upper limits by an initial replacement of the unknown nuisance parameter vector by its profile maximum likelihood estimator. While the resulting limits no longer possess the optimal properties of tight limits exactly, the paper presents both numerical and theoretical evidence that the resulting coverage function is close to optimal. Moreover these profile upper limits are much (possibly many orders of magnitude) easier to compute than tight upper limits.     

Residential and Racial Mortality Differentials in the South by Cause of Death1

Abstract Mortality levels and differences by metropolitan-nonmetropolitan residence for blacks and whites are examined in the South using life-table techniques. Life tables are constructed for Southern subpopulations using U.S. vital statistics and census data for 1980, and differences in the life expectancies are decomposed by cause of death. Results suggest that life expectancy is higher for metropolitan and white populations. Racial mortality differences are slightly greater in metropolitan areas of the South when compared to the nonmetropolitan South. Residential mortality differences are attributable mainly to differences in the effects of heart disease and accidents. Most of the racial mortality difference is due to differences in effects of malignant neoplasms, cardiovascular disease, perinatal conditions, and homicide. Policy efforts aimed at specific causes of death in specific populations are suggested as a means of reducing racial or residential life expectancy differentials.     

Community councils as a medium for public participation: a case study in Grampian region

Planning authorities in Scotland are obliged to consult the public before finalising policies and proposals to be included in their development plans. Community councils are intended to operate as ‘the voice of a neighbourhood’ and this paper analyses an attempt by a regional planning authority to use community councils as a vehicle for consulting the public in the preparation of a structure plan for a rural area. The paper concludes that the attempt was unsuccessful because the community councils were not clear about the role they were expected to play; nor were they adequately prepared for their role. The authors believe, however, that the involvement of community councils has potential for promoting greater public participation in the preparation of development plans.     

Marriage and childbirth as factors in dropping out from school: an analysis of DHS data from sub-Saharan Africa

Leaving school prematurely is often claimed to be among the most negative consequences of early marriage and pregnancy for girls in less developed countries. However, an analysis of the relative frequency with which these events actually occur or are named as reasons for leaving school reveals that, at least in the case of francophone Africa, they explain no more than 20 per cent of dropouts. To the extent that demographic events trump school or family factors as determinants of school-leaving, our data indicate that it is union formation--defined by the DHS as first marriage or cohabitation--rather than childbirth that is more likely to have this effect. 'Schoolgirl pregnancy' typically accounts for only between 5 and 10 per cent of girls' departures from school. Furthermore, the risks of leaving school because of pregnancy or marriage have declined over time with the decline in rates of early marriage and childbearing.     

Socioeconomic Determinants of White and Black Males’ Life Expectancy Differentials, 1980

Epidemiological transition theory suggests that two population existing under disparate socioeconomic conditions would have different life expectancies as the result of cause-of-death differences. The effect of racial socioeconomic differentials on the total racial life expectancy differential are examined as they act through specific cause-of-death differentials. Results suggest that residential isolation of blacks has a strong effect on the total life expectancy differential as it acts through the racial homicide differential. The racial unemployment difference also has a strong effect on the total differential as it acts through the racial heart disease differential. Implications of the findings for reducing life expectancy differentials are discussed.     

The demographic evolution of human populations: The role of selection and environmental factors

Selection (genetic and cultural) and environmental variation are the principal mechanisms determining patterns of demographic change in human populations. Conditions exist under which the nature and intensity of these forces can be inferred from temporal trends in the demographic variables. These conditions, which can be expressed in terms of relations between the Malthusian parameter and population entropy, provide a means for evaluating the effect of selective and nonselective factors on demographic trends in human populations. The distinction between the roles of selection and environmental factors is illustrated by a study of the demographic transition in Sweden (1778-1965). This study shows that demographic changes during the pre- and posttransitional phases are determined mainly by environmental factors, whereas the changes during the transitional phase are mainly due to cultural selection. This analysis provides, for all three phases of the demographic transition, quantitative measures of the intensity of the forces (selective and nonselective) acting on both mortality and fecundity distributions.