Research Advances (Archive)

RESEARCH ADVANCES: ARCHIVE

MPC researchers are making important advances in a broad range of population-related fields. The following paragraphs highlight a few of the most striking research advances of the past year, focusing especially on measurement issues.

John Robert Warren and Andrew Halpern-Manners (2012), writing in Demography, used linked data from the Current Population Survey (CPS) to investigate panel conditioning effects and concluded that such effect downwardly bias CPS-based measures of unemployment.

Sheela Kennedy and Catherine Fitch (2012), also writing in Demography, examined the measurement of cohabitation through new questions in the CPS compared with the older measure based on "unmarried partner" relationships.

J. Trent Alexander's NBER working paper, "Inaccurate age and sex data in the Census PUMS files: Evidence and Implications" identified significant errors in age and sex values in more than 200 public use datasets produced by the Census Bureau. The paper was profiled in the New York Times, the Wall Street Journal, and the Washington Post. The Wall Street Journal said that the paper was "sending researchers inside and outside government scrambling to check whether some key findings need to be reassessed." The Washington Post reported that the paper led the Director of the Census Bureau to order a review of internal procedures relating to data quality and confidentiality. Because of the paper's findings, the Census Bureau is now considering releasing corrected versions of more than 200 erroneous datasets. The working paper is regularly cited in the introduction to Social Security Administration reports and is forthcoming in Public Opinion Quarterly.

Ragui Assaad and Deborah Levison have published their work on the effect of child labor on schooling in Egypt. They argue that in Egypt, girls' work primarily takes the form of domestic tasks, which are not considered in many studies of child labor. Their paper investigates the effect of girls' work on their school attendance. It uses a modified bivariate probit approach to estimate the effect of work on schooling while allowing for the simultaneous determination of the two outcomes. It presents evidence that the substantial burden of girls' domestic work leads to lower rates of school attendance. Policies that attempt to ban the labor-force work of children will have practically no effect on girls' education in Egypt, while interventions reducing the drudgery of household labor through, for example, improved water and sanitation infrastructure, have better prospects for success.

In their 2009 paper in Demography, Ruggles, Alexander, Oakes and others describe the appropriate use of historical census data for drawing statistical inferences. Virtually all quantitative microdata used by social scientists derive from samples that incorporate clustering, stratification, and weighting adjustments. Such data can yield standard error estimates that differ dramatically from those derived from a simple random sample of the same size. Researchers using historical U.S. census microdata, however, usually apply methods designed for simple random samples. The resulting p values and confidence intervals could be inaccurate and could lead to erroneous research conclusions. Because U.S. census microdata samples are among the most widely used sources for social science and policy research, the need for reliable standard error estimation is critical. In this paper, MPC researchers evaluate the historical microdata samples of the Integrated Public Use Microdata Series (IPUMS) project from 1850 to 1950 in order to determine (1) the impact of sample design on standard error estimates, and (2) how to apply modern standard error estimation software to historical census samples. They exploit a unique new data source from the 1880 census to validate our methods for standard error estimation, and then apply this approach to the 1850–1870 and 1900–1950 decennial censuses. They conclude that Taylor series estimation can be used effectively with the historical decennial census microdata samples and should be applied in research analyses that have the potential for substantial clustering effects.

Using data from the American National Election Study linked to income data from the U.S. census, Matthew Schroeder and Glenn Firebaugh published the 2009 paper "Does Your Neighbor's Income Affect Your Happiness?" in the American Journal of Sociology. They found that Americans tend to be happier when they reside in richer neighborhoods (consistent with neighborhood studies) in poorer counties (as predicted by the relative income hypothesis). The relative income or income status hypothesis implies that people should be happier when they live among the poor. Findings on neighborhood effects suggest, however, that living in a poorer neighborhood reduces, not enhances, a person's happiness. Thus it appears that individuals in fact are happier when they live among the poor, as long as the poor do not live too close.

Kathleen Call, Lynn Blewett and Pamela Jo Johnson sought to determine whether aggregate national data for American Indians/Alaska Natives (AIANs) mask geographic variation and substantial subnational disparities in prenatal care utilization. In this study, published in the American Journal of Public Health, they used data for US births from 1995 to 1997 and from 2000 to 2002 to examine prenatal care utilization among AIAN and non-Hispanic White mothers. The indicators studied were late entry into prenatal care and inadequate utilization of prenatal care. Rates and disparities were calculated for each indicator at the national, regional, and state levels, and examined as to whether estimates for regions and states differed significantly from national estimates. State-specific changes in prevalence rates and disparity rates over time were then estimated. They found that prenatal care utilization varied by region and state for AIANs and non-Hispanic Whites. In the 12 states with the largest AIAN birth populations, disparities varied dramatically. In addition, some states demonstrated substantial reductions in disparities over time, and other states showed significant increases in disparities. They concluded that substantive determinations about AIAN health care disparities should be geographically specific, and that conclusions drawn at the national level may be unsuitable for policymaking and intervention at state and local levels.