Øystein Kravdal

http://www.demographic-research.org/volumes/vol15/1/

doi:10.4054/DemRes.2006.15.1

	Click the icon to view and/or download the PDF file. Once you are in the PDF file, use your browser back button to return to this page.

Abstract
There is much interest these days in the importance of community institutions and resources for individual mortality and fertility. DHS data may seem to be a valuable source for such multilevel analysis. For example, researchers may consider including in their models the average education within the sample (cluster) of approximately 25 women interviewed in each primary sampling unit (PSU). However, this is only a proxy for the theoretically more interesting average among all women in the PSU, and, in principle, the estimated effect of the sample mean may differ markedly from the effect of the latter variable. Fortunately, simulation experiments show that the bias actually is fairly small - less than 14% - when education effects on first birth timing are estimated from DHS surveys in sub-Saharan Africa. If other data are used, or if the focus is turned to other independent variables than education, the bias may, of course, be very different. In some situations, it may be even smaller; in others, it may be unacceptably large. That depends on the size of the clusters, and on how the independent variables are distributed within and across communities. Some general advice is provided.

Author's affiliation
Øystein Kravdal
University of Oslo, Norway

Keywords
average, bias, clustering, contextual, DHS, measurement error, multilevel, simulation, size, small

Word count (Main text)
4522

Other articles by the same author/authors (in Demographic Research)

	[18-7] Does income inequality really influence individual mortality?: Results from a ‘fixed-effects analysis’ where constant unobserved municipality characteristics are controlled
	[17-9] Effects of current education on second- and third-birth rates among Norwegian women and men born in 1964: Substantive interpretations and methodological issues
	[16-15] Does cancer affect the divorce rate?
	[10-1] Educational differentials in male mortality in Russia and northern Europe: A comparison of an epidemiological cohort from Moscow and St. Petersburg with the male populations of Helsinki and Oslo
	[9-2] The problematic estimation of "imitation effects" in multilevel models
	[6-10] The impact of individual and aggregate unemployment on fertility in Norway
	[6-9] Is the Previously Reported Increase in Second- and Higher-order Birth Rates in Norway and Sweden from the mid-1970s Real or a Result of Inadequate Estimation Methods?
	[5-6] The High Fertility of College Educated Women in Norway: An Artefact of the Separate Modelling of Each Parity Transition
	[3-3] A search for aggregate-level effects of education on fertility, using data from Zimbabwe
	[S3-6] An Illustration of the Problems Caused by Incomplete Education Histories in Fertility Analyses

Similar articles in Demographic Research

[9-2] The problematic estimation of "imitation effects" in multilevel models (multilevel, bias)

[ Back to previous page ]