No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Cross-Population Comparability and PPPs: Using Micro-Data on Indicators of Consumer Durables
Abstract
This paper addresses the issue of cross-population comparability of economic status indices estimated from household-level indicator variables on consumer durables using latent variable statistical models. The problem is similar to the one faced in estimation of PPPs. The same basket of consumer durables need not imply the same level of economic status across countries. This is because, even for the same level of the true economic status, the likelihood of owership of a given consumer durables is not the same across countries. This may be due to different prices structures, preferences, or other enviornmental factors. The problem we face is similar to that faced by psychometricians in intelligence testing using question banks: even for the same level of intelligence some questions exhibit differential item functioning (DIF) in that the likelihood of responding correctly for these questions varies significantly across socio-demographic groups. We propose one way of making the estimates of economic status comparable across countries which relies on assuming that the entire set of consumer durables not exhibit DIF. This enables us to estimate economic status on ac ommon scale across countries. The results seem promising meriting further research for testing and validating using better-quality data on consumer durables.
... The full equation system thus comprises the schooling equation (1) and equations (2) for the living-standards indicators. In setting out the model in this way, with latent factors embedded in structural equations, we follow an approach that has been recommended by several researchers (notably Sahn and Stifel 2000; McDade and Adair 2001; Tandon et al. 2002; Ferguson et al. 2003). Filmer and Pritchett (1999, 2001) have developed an alternative approach based on the method of principal components. ...
... The full equation system thus comprises the schooling equation (1) and equations (2) for the living-standards indicators. In setting out the model in this way, with latent factors embedded in structural equations, we follow an approach that has been recommended by several researchers (notably Sahn and Stifel 2000; McDade and Adair 2001; Tandon et al. 2002; Ferguson et al. 2003). Pritchett (1999, 2001 ) have developed an alternative approach based on the method of principal components. ...
This paper presents findings of an investigation into the effects of living-standards and relative poverty on children’s schooling in urban and rural areas of Senegal. To measure living standards, we apply a multiple-indicator, multiple-cause (MIMIC) factor-analytic model to a set of proxy variables collected in the 2000 Multiple Indicator Cluster Survey and extract an estimate of the relative standard of living for each household. Using this estimate, we find that in Senegal’s urban areas, living standards exert substantial influence on three measures of schooling: Whether a child has ever attended school; whether he or she has completed at least four grades of primary school; and whether he or she is currently enrolled. In rural areas of Senegal, however, the effects are weaker and achieve statistical significance only for the wealthiest fifth of rural households. Two educational inequalities persist with living standards held constant. First, the advantages enjoyed by urban families in Senegal remain considerable: Even the poorest fifth of urban children are more likely than rural children to have attended school, to have completed four years or more of primary education, and to be currently enrolled. Second, gender gaps in schooling are pervasive and are only modestly influenced by standards of living. In both urban and rural areas of Senegal, girls suffer from marked disadvantages relative to boys in all three measures of schooling. In wealthier urban households, girls’ disadvantages are smaller, but not completely eliminated. Furthermore, no systematic reduction in female disadvantage is apparent in rural Senegal, even in the uppermost stratum of households. To judge from these findings, in Senegal income growth alone is unlikely to close the schooling gap between urban and rural areas or between boys and girls.
... Because the DHS program gathers no information on incomes or expenditures as such, measures of poverty are limited to those that can be fashioned from a small set of proxy variables, including ownership of consumer durables and crude assessments of the quality of housing. A lively literature has emerged in the past few years in which the merits of various statistical techniques that use such indicators are debated (Filmer andPritchett 1999, 2001;Montgomery, Gragnolati, Burke, and Paredes 2000;Hewett and Montgomery 2001;Sahn and Stifel 2000;McDade and Adair 2001;Tandon, Gakidou, Murray, and Ferguson 2002;Ferguson, Tandon, Gakidou, and Murray 2003). We explore two of the most promising approaches for distilling the proxies into a single living standards index: confirmatory and structural factor analysis. ...
For the foreseeable future, world population growth will be mainly concen-trated in the cities of developing countries. According to the United Nations (2000), by the year 2030 the world's population will exceed today's total by 2.06 billion persons, and some 1.94 billion of these will be added to the urban areas of low-income countries. By 2030, it is believed, all developing regions will have crossed the half-way point to become more urban than rural. Even if a generous allowance is made for the variety of urban definitions and the uncertainties of demographic forecasts, the central message remains clear: soon it will no longer be possible to speak of low-income countries as being mainly rural. If these are the prospects in view, then researchers concerned with poverty and opportunity must increasingly set their concerns in urban contexts. What might city life imply for levels of health, and for health inequalities? In its new book, the Panel on Urban Population Dynamics (2003) finds substantial evidence, across the full range of developing countries surveyed by the Demographic and Health Surveys program, that the urban poor face health risks that resemble those of rural populations. In these analyses, the urban poor were defined as the lowest quartile of urban households, a group sizable enough to merit policy attention. Furthermore, relying on evidence from micro-studies of selected slum populations, the Panel on Urban Population Dynamics (2003) found that the risks facing the residents of slums can rival and even exceed rural risks.
... The full equation system thus comprises the schooling equation (1) and equations (2) for the living standards indicators. In setting out the model in this way, with latent factors embedded in structural equations, we follow an approach that has been recommended by several researchers (notably Sahn and Stifel 2000; McDade and Adair 2001; Tandon et al. 2002; Ferguson et al. 2003). Pritchett (1999, 2001) have developed an alternative approach based on the method of principal components. ...
This paper explores the implications of urban poverty for children’s educational attainment, a central measure of human capital that has a well-documented and pervasive influence on later-life demographic and labor force behavior. We compare levels of children’s schooling in Cairo and urban Egypt with those of Allahabad, India, a rapidly growing city of some 1.1 million persons in the northern state of Uttar Pradesh, looking for poverty effects at both household and neighborhood levels.
... McDade and Adair 2001; Tandon et al. 2002; Ferguson et al. 2003). Pritchett (1999, 2001) have developed an alternative approach based on the method of principal components. ...
This paper investigates the effects of living standards and relative poverty on infant and child mortality in the urban areas of Egypt, with a special fo-cus on Greater Cairo. To measure living standards, we apply a multiple-indicator, multiple-cause (MIMIC) factor-analytic model to a set of proxy variables collected in the 2003 Interim Demographic and Health Survey for Egypt, and extract an estimate of the relative standard of living for each household. Using this estimate, we find that living standards exert substan-tial influence on early childhood chances of survival. Moreover, there is ev-idence that household living standards make a difference to child survival, much more than that of neighborhood.
... The full equation system thus comprises the health equation (1) and equations (2) for the living standards indicators. In setting out the model in this way, with latent factors embedded in structural equations, we follow an approach that has been recommended by several researchers (notably Sahn and Stifel 2000; McDade and Adair 2001; Tandon et al. 2002; Ferguson et al. 2003). Pritchett (1999, 2001) have developed an alternative approach based on the method of principal components. ...
In the United States and other high-income countries, there is intense scholarly and programmatic interest in the effects of household and neighborhood living standards on health. Yet few studies of developing-country cities have explored these issues. We investigated whether the health of urban women and children in poor countries is influenced by both household and neighborhood standards of living. Using data from the urban samples of 85 Demographic and Health Surveys and modeling living standards using factor-analytic MIMIC methods, we found that the neighborhoods of relatively poor households are more heterogeneous than is often asserted. Our results indicated that poor urban households do not tend to live in uniformly poor neighborhoods: about 1 in 10 of a poor household's neighbors is relatively affluent, belonging to the upper quartile of the urban distribution of living standards. Do household and neighborhood living standards influence health? Using multivariate models, we found that household living standards are closely associated with three health measures: unmet need for modern contraception, attendance of a trained health care provider at childbirth, and young children's height for age. Neighborhood living standards exert a significant additional influence in many of the surveys we examined, especially for birth attendance.
ResearchGate has not been able to resolve any references for this publication.