Literaturnachweis - Detailanzeige
Autor/inn/en | Ashenfelter, Orley; und weitere |
---|---|
Institution | World Bank, Washington, DC. |
Titel | Collecting Panel Data in Developing Countries: Does It Make Sense? Living Standards Measurement Study Working Paper No. 23. |
Quelle | (1986), (63 Seiten) |
Sprache | englisch |
Dokumenttyp | gedruckt; Monographie |
ISBN | 0-8213-0673-1 |
Schlagwörter | Data Collection; Demography; Developed Nations; Developing Nations; Economic Factors; Economic Research; Economics; Living Standards; Observation; Research Design; Research Methodology; Social Indicators; Social Science Research Data capture; Datensammlung; Demografie; Developed countries; Industriestaat; Industrieland; Developing country; Developing countries; Entwicklungsland; Ökonomischer Faktor; Wirtschaftsforschung; Volkswirtschaftslehre; Lebensstandard; Beobachtung; Forschungsdesign; Research method; Forschungsmethode; Social indicator; Sozialer Indikator; Social scientific research; Sozialwissenschaftliche Forschung |
Abstract | There is no difference in principle between developing and developed countries in deciding whether it is desirable to collect panel data. However, there are many relevant circumstantial differences in developing countries. A number of aspects of the collection and use of panel data from developing country households are reviewed. Sampling issues are discussed in Section 1. The authors conclude that there are likely to be real, if modest, benefits from incorporating some panel element into household survey data collection in developing countries. The recognition that panel data are likely to be subject to substantial errors of measurement does not invalidate this conclusion. Section 2 discusses the measurement of income dynamics, an issue that cannot be addressed without panel data. Recent research using U.S. data is reviewed to show that comparable work for developing countries would add an important dimension to discussions of poverty, inequality, and development. It is in the third area of review, that of econometric analysis, that the real benefits of panel data appear most fragile. These issues are discussed in Section 3. While it is true that panel data offer the unique ability to deal with the contamination of econometric relationships by unobservable fixed effects, the presence of measurement error can compromise the quality of the estimates to the point where it is unclear whether cross-section or panel estimators are superior. This situation is in sharp contrast to that for sampling where errors of measurement typically cannot reverse the superiority of panel over cross-section estimators. The document concludes with the argument that panel data should be collected in both developing and developed countries. (Author/BZ) |
Erfasst von | ERIC (Education Resources Information Center), Washington, DC |
Update | 2004/1/01 |