Research
The three central themes by which we organize the research activities of the Population Research Center are: (1) Children, Families, and Marriage; (2) Healthy Behavior at all Ages; and (3) Employment, Earnings, and Poverty. Each theme expresses several hallmarks of our program: its multidisciplinary nature; its dual emphasis on theory and evidence and on concepts and measurement; and its attention to social policies that are demographic in nature. The three are neither wholly distinct from one another nor exhaustive in capturing the interests and contributions of our Center's research associates.
Children, Families, and Marriage
No subject better exemplifies the importance of demographic research than that of understanding the factors affecting the fertility of a nation and its investments in children. Children are nurtured in families and they are affected by the forces that influence the size, structure, and social arrangements within families and households. Understanding the changes in those structures and how they influence children are key topics in demographic research. These social institutions also affect adults in many ways, so even aside from our interests in children, the behavior of families remains an important subject for study. We need to better understand the incentives to form and sustain family structures, and how and why the functions of families change. Marriage, as a distinct social institution, has for years been a focus of study by several PRC research associates, who analyze its functions, assess its limits, compare its attributes to cohabitational unions, and study its stability and its value to spouses, children, and society. Marriage is currently a topic of great policy interest because of rights extended to "marriage partners" but not enjoyed by less formal partnerships and because of the key role it plays in welfare reform. Dispute reverberates through many contexts regarding what's right from the various perspectives of religion, ethics, morality, efficiency, and law, among others, so this research domain has been and will continue to be one of our primary areas of scholarly contribution.
We identify three sub-themes within the area of "children, families, and marriage": (a) fertility patterns, (b) investments in children, and (c) marriage, divorce, and cohabitation as social institutions through which family activities are organized. Below, we provide a brief description of the work underway and planned by PRC research associates, by each of these sub-themes.
Fertility Patterns
Regarding fertility patterns, ongoing projects are global in focus, looking at issues of the demographic transition as they play out today in developing countries, and reconsidering the Malthusian model of global growth in population and real incomes. Other researachers are studying fertility patterns and their relationship to economic growth, motivated by the observation that negligible changes in population size and economic well-being over many centuries have been followed by quite extraordinary growth in both over the past two hundred years. This contrasts dramatically with the Malthusian steady-state equilibrium patterns of a level population and an iron law of wages and no growth in per capita incomes implied by the diminishing returns of that Malthusian system. The circumstances of agricultural societies with low population density and low rates of growth of economic produce have given way to the highly dense urban communities characterized by economies of scale in research and development, specialization, and investments in human capital. We are exploring the conditions under which these scale phenomena induce greater, not smaller, investments in human capital as populations increase in size and density. Don Bogue is studying why Hispanic fertility in the U.S. is about 50 percent higher than that of the general U.S. population.
Parental Investments in their Children
The issues of expenditures on children, of the stages of their growth best suited for efficient investments in their human capital, of family influences on their healthy development, and a host of related topics of considerable intellectual and policy import are under study by research associates of the PRC. Ariel Kalil focuses her research agenda here and has recently co-edited a book on Family Investments in Children's Potential. The over-arching question of her research is how children develop and how parents function in stressful or disadvantageous social environments. Thus, she writes on the links between economic hardship and family well-being, looking primarily at adolescents. In her recent work Kalil focuses on the impact on adolescents of growing up in a cohabiting family in contrast to a married family structure, and the psychological impact on the adolescent of a single mother's employment transitions. Similarly, other colleagues in the PRC have produced several assessments of family changes from the perspective of children, documenting in particular the increasingly complex sequence of childhood living arrangements in a number of Western countries. James Heckman's essays and analysis of the efficacy of investments in skills at various ages and life stages is among the most influential of the newer work from our Center. He has documented the facts that the estimated rates of return to skill investments are dramatically higher for pre-school children than for school-aged children, but lower yet for post-school-aged mid-life adults in whom remedial schooling and job training programs are embedded.
Marriage, Family Structure, and the Family
In this third sub-theme, studies examine the nature of adult partnerships, including the factors that determine their quality and how sexual patterns within the partnership affect the relationship. Among our senior research associates, Linda Waite has focused much of her recent research attention on the case for marriage as a viable and efficacious social institution. She emphasizes evidence on the health benefits of marriage.
Rob Townsend's Thailand survey project investigates the role of the social institution of the family, including the extended family, in offering insurance and credit to its members. He places these functions in a context of a village and a community that also enjoys more formal financial institutions and the structures of certain markets. The relationship of economic development in the transference of functions from the informal social institutions to the more formal market institutions is but one of the focuses of his work that includes the creation of a major new data resource, funded through NICHD and NSF, and publicly available through the web as described below. In the context of contemporary U.S., Derek Neal studies aspects of the marriage market and how they intersect with patterns in the labor market, explaining the growth in parenthood among people who have never married, and considering the role of social welfare policies.
Healthy Behavior at all Ages
Over the past 15 years, two institutional developments within the University of Chicago have strengthened the linkages between the social science research and medical research on health. One was the creation of the Harris School of Public Policy Studies with its focus on health policy; the other was the creation of the Department of Health Studies within the Pritzker School of Medicine/Biological Sciences. Each houses several social scientists conducting research on health at the interface between the physician and the family. Currently the ties are strongest between the social science/policy and medical groups, but as this partnership continues to develop, we expect to see stronger linkages between the social science/policy work and the rapidly developing population-relevant work in the biological sciences, especially research on human genetics and neurological development in utero and during infancy.
We characterize our work on the theme of healthy behavior at all ages in two sub-themes: (a) health-promoting strategies, and (b) sexual behavior and health including studies of HIV/AIDS. We note as well a third nascent area of considerable interest to several of our research associates; while it is not a full-fledged area of research, it constitutes a growing interest that we intend to develop within the next few years. It is the relationship between environmental exposures and investments at very young ages, including in utero, and later health and skill levels in adulthood: "early origins of adult well-being."
Health-Promoting Strategies
We have several research associates, both junior and senior, studying issues that involve health-promoting strategies. Kate Cagney studies a range of factors that impact health outcomes, especially how neighborhoods affect health strategies. Among our senior researchers studying aspects of healthy behavior at all ages, Robert Fogel's work on nutrition and its effects on health offers one fascinating example. Fogel has published extensively on the effects of economic development on human physiology. He is now exploring the implications of what he calls "physiological capital." He argues that throughout most of history the available food supply was too small to allow for proper physiological development - the food supply was barely sufficient to provide the needed energy the body requires for baseline maintenance - i.e., to maintain body temperature and keep the heart and other vial organs functioning plus the energy needed to digest food and engage in vital hygiene. (He points out that the typical American male today is about 70 inches tall and weighs about 172 pounds, and so he needs about 2,300 calories for baseline maintenance. If, he points out, 18th century British or French males had been that large, "virtually all the energy produced by their food supplies would have been required for maintenance, and hardly any would have been available to sustain work" or any other activities.)
Fogel uses this perspective to study infant death rates, among other things. He estimates that nearly 80 percent of the births to impoverished English workers around 1800 were under 2,500 grams in weight, and using the evidence on the relevant relation of birth weight to neonatal deaths in the U.S., he contends that if the contemporary U.S. population had the distribution of birth weights of 1800 England, the neonatal death rate would be about 173-per-1000 births instead of the actual 27-per-1000. The larger size of U.S. babies at birth (3,128 grams in his data from 1950) can explain most all of the observed decline in neonatal death rates! He takes the argument a step further: malnourished mothers produce small babies, he says, not only because of deficiencies in their diet and exposure to disease during pregnancy, but because they have small pelvic cavities, implying that the birth weight that minimized perinatal death was about 700 grams below that of U.S. women in 1950. Thus, "a condition for surviving the birth process was such a low birth weight that the neonate was at very high risk of dying shortly after birth." Economic growth has increased the physiological capital of poor women sufficiently that now the birth weight of their children is 1.7 times what it was two centuries ago, meaning that less than 8 percent of all births in OECD nations are now below 2,500 grams. Fogel is at work investigating how much of the improvement in physiological capital is attributable to improvements in the environment, medical technology, or the development of effective systems for delivering health services.
PRC senior researcher Tomas Philipson also researches weight, with a focus on obesity and why it is that, instead of spending major portions of income producing energy by caloric intake, today we spend a growing amount on high-income efforts to diet and exercise in order to be able to burn off all the calories we intake. The decline in the cost of producing nutritious foods is a key factor in Philipson's work as in Fogel's. Willard Manning studies aspects of the cost-effectiveness of medical treatments and brings his considerable expertise in econometrics and evaluation research to this domain. He is currently working on studies of depression among diabetic patients, treatment of substance abuse and cardiovascular disease, as well as the new hospital specialty of "hospitalists". Manning and colleagues look at the price sensitivity of alcohol dependency, a topic of much relevance to the policy issue of discouraging high rates of alcohol consumption especially among youths by taxing the product. Working with Donald Kenkel, Manning discusses how to evaluate the economic consequences of improvements in nutrition- and exercise-related health.
David Meltzer has undertaken a major inquiry of the new medical specialty of hospitalists, those physicians whose practices are more than one-quarter in hospital inpatient care. He has underway a multi-million dollar AHRQ-funded, multi-site study of their effectiveness, the cost implications of their handling of patients, and their influence on both the culture of hospitals and the interplay between the doctor and the patient/family as managers of health care. Meltzer's work on hospitalists is undertaken, in part, with Manning. Marshall Chin also investigates strategies that are health promoting, focusing on the advisory role of the physician in guiding the individual's choices about those strategies. Dr. Chin studies issues of diabetes, especially differences in white and African-American populations, how different physician-specialists treat older diabetic patients differently, and how community health centers handle health care for elderly diabetics. Bruce Meyer, whose primary research interests lie in assessment of other federal welfare programs, has recently completed a provocative assessment of the crowd-out effects on private health insurance of programs such as the free health care provided by Federally Qualified Health Centers. He and a colleague show that matched CPS data indicates that while hospital care does not have a crowd-out effect, clinic care does substantially crowd out private health insurance coverage.
The work of psychologist John Cacioppo also falls within our research theme of health at all ages. He works in the field of social neuroscience, bridging social science and biological insights about complex social behaviors and their influence on health. His work on loneliness is most salient to our Center's research.
Sexual Behavior
The second sub-theme in healthy behavior at all ages is sexual behavior. Among the senior researchers, Tomas Philipson had published a book and several articles on risks of AIDS, and has a policy interest expressed in his essay with R. Posner on "The Optimal Regulation of AIDS." The NICHD-supported "Chicago Health and Social Life Survey" undertaken by Edward Laumann and his students has resulted in several papers and a book that includes investigations of the behaviors and social structures that influence transmission of HIV in Chicago. Laumann undertook an innovative study of sexual practices and attitudes in China, again with NICHD support, and again with a strong focus on the behaviors and strategies that affect the risk of transmission of HIV. Don Bogue has undertaken a study of HIV/AIDS as well, focusing on how women in some 44 developing countries assess their own risks of contracting the disease, and how well informed they are about ways of preventing it.
Publications reporting research on other negative outcomes from sexual behaviors are also widespread among our faculty. Sexual dysfunction has been a topic of several papers by Laumann who has also been a principal investigator on an 11-country comparative analysis of sexual dysfunction, funded by Pfizer, Inc. Moreover, our researchers are also strongly committed to research on the positive outcomes of sexual behavior. Much of the book by Laumann and his students looks at the social dimensions of sexuality and its expression. In the survey project overseen by Linda Waite, with an emphasis on sexual issues among the elderly, there is much attention to the positive aspects of sexual behavior for the health and well-being of older men and women. Robert Michael has published on the sexual behaviors of U.S. adolescents, based on the NSLY97 data set on which he was P.I. His project has a major focus on intimacy, on skills in sexual behaviors, and on the positive products from sex, arguing that too much of our attention has focused on the negative side of sex because negative externalities from sex prompt public discourse and create public expense. Yet another, very different project by Martha McClintock, reported in the J Sex Research, studies sexual experiences of women as their interest in sex varies over the menstrual cycle.
Adult Consequences of Fetal and Infant Experiences
In Fogel's work on birth-weight and the role of nutrition, he has embarked on one project that explores linkages to birth weight and will investigate how small-area (ward and census track level) environments of water quality, milk distribution and quality, and sanitation, over the years 1898 to 1935, link to birth records in three major city hospitals. He is drawing explicitly on the work of David Barker and colleagues. Of special interest to Fogel and his team is the impact of prenatal environment and maternal health on later-life mortality and disability. In this work, Fogel collaborated with Dr. Kwang-Sun Lee, a Professor of Pediatrics and Section Chief of Neonatology at our medical school, who has written extensively on neonatal morality and very low birthweight. If we develop this area of research into a center theme, Dr. Lee will be an important addition to the team. In his new work on the parental investments in children, Heckman stresses that "skill begets skill" and uses that fact in explaining the accumulation of evidence that the rate of return on human capital investments are typically much higher when those investments are made at very young ages, since young children then benefit in the form of greater skill at acquiring more skill as they pass through each subsequent level of education. Given the evidence about the critically important development in the fetus in the weeks, not months, following conception, Michael notes the important influence of a mother's behavior during those early weeks of her pregnancy on the healthy development of her child. These are shown to relate to the cognitive development of the child at age 11. While the phenomena that have been most extensively studied are aspects of adult health, the same behavioral mechanisms apply to investments in cognitive capability, skill capital, and lifetime productivity. While skill begets skill, healthy neurological development enables skill development. The nature of these efforts by parents and the exposures to fetal insult have major impact on subsequent investments in the children's human capital.
The work of Fogel, Heckman, and Michael suggest one line of work on this subject. Martha McClintock, has documented in animal studies that social factors like isolation are linked to mammary tumors, and she speculates that this may be so in human populations as well. Addressing how the social behaviors and exposures interact with and affect the regulation of genes is much associated with this area of childhood precursors of adult health.
Another reason for our growing interest in these questions lies in the potential of NICHD's National Children's Study. With no claim here to a proprietary role, we note that the longitudinal NCS which plans to follow 100,000 births, has as one of its several areas of focus a working group on "early origins of adult health." We quote that working group's description of its interest, suggesting that they parallel our own. They describe the "lifecourse approach to chronic diseases," saying "The overarching principles are that susceptibility to adult chronic disease is determined by a dynamic process that occurs over the lifespan. Perturbations (insults) that determine adult health states may occur anywhere from pre-conception to embryonic, fetal, infant, childhood, adolescent (and adult) life. These insults can affect both somatic growth and maturation of metabolic systems, and they include a range of determinants, including societal, lifestyle and biological. These determinants act in concert with each other." Since the NCS plans to follow these children to age 21, and to include a broad array of social and behavioral factors, there will be an exceptional opportunity to investigate the familial and community social investments in these children. Our PRC faculty hope to be able to study the family responses to the chemical, biological, and social environmental exposures experienced by these children.
Employment, Earnings, and Poverty
In the earliest years of the PRC (1980 to 1983), ours was a P-50 center focused exclusively on employment and earnings from an economic perspective. As the Center transformed into a broader multi-discipline center, issues in sociology of education, measurement of cognitive achievement of children, and the dynamics of the work-and-family or career-and-parenting behaviors came to play an increasingly important role. Issues of social welfare policy have subsequently became more salient for our Center as our policy school faculty has grown, and four of the five new faculty members focus their research in this area.
In this broad signature theme we identify three sub-areas: (a) youth transitions to the labor market; (b) skill investments and schooling; and (c) federal policies focused on inequality and poverty, including anti-discrimination policies and employment, welfare (TANF) employment, and family preservation.
Youth Transitions to the Labor Market
Adolescent behavior is one of the cross-cutting interests within our Center, with several faculty looking at transitions to marriage. A key transition, of course, is the transition from school to work, from dependence to relative economic independence and a career in the labor market. Here, several of our faculty members devote much attention. For example, Derek Neal looks at the complexity of job mobility. He distinguishes between job changes among firms while performing the same basic tasks and job changes that involve changing tasks or careers. He offers a formal model suggesting that young people first move among jobs to get the career and its tasks right, and then at a second stage of their search, they select the "right" firm or job. He shows that about two-thirds of men's lifetime job changes occur during the first ten years of work experience and about one-third of real wage growth occurs during that interval of the working lifetime. Thus, Neal suggests both theoretically and empirically using the NLSY, much of what is going on in those early years is finding a good career match. Thereafter, the worker experiences some mobility seeking the best job with a narrowly defined set of tasks. He suggests that apprenticeship systems are attractive to employers but he notes that they may have the negative side effect of severely reducing allocative efficiency: it is important that a young man find, by experience, what he does well before settling into selecting the optimal firm in which to do it. (Neal's paper on this subject won the 2000 Lewis Prize for best paper in the JoLE.)
Susan Mayer's work looks at the influence of inequality on children's well-being, their schooling, their aspirations, and their opportunities for training that pays well in the labor market. The NLSY97, on which Robert Michael was P.I., contains considerable detail about the schooling behaviors and job experiences of the youths, and nine of the dozen chapters in the book were authors by PRC research associates, their students, or PRC pre- or postdoctoral fellows.
Skill Investments and Schooling
Continuing the long tradition of research in both economics of education and the sociology of education, Center faculty currently study both the nature of schooling and its effects on subsequent earnings and demographic behaviors. Ofer Malamud investigates different strategies by school systems for causing youths to make choices about their occupational specialization early or late in their training - he contrasts the policies in Scotland and England that differ substantially in when they require students to specialize. James Heckman studies the importance of parental background and family environment, as well as credit constraints as determinants of human capital investments. Robert LaLonde has focused on two labor market issues: the retraining of displaced workers in formal classroom settings and on the job, and the assimilation of immigrants. Also, LaLonde has also recently published several literature-summaries pertaining to social policy and the labor market. Kaz Yamaguchi studies gender role attitudes, both comparatively across the sexes and across nations, focusing on U.S.-Japanese commonalities and differences. Jeff Grogger has investigated several distinct features of U.S. education policy and their impact on post-school earnings; he has studied the effects of greater expenditures per pupil on subsequent earnings. Derek Neal, as well, has investigated the reasons for the differential student test scores and other outcomes from Catholic and public schools.
Federal Policies Focused on Inequality and Poverty
Our faculty's work on social policy covers both anti-discrimination and welfare policies. Regarding the former, Jonathan Guryan's assessed the 1954 Supreme Court ruling that separate schools for blacks and whites were "inherently unequal" to determine if desegregation plans of the subsequent decades benefited black students for whom those plans were designed. The patterns of black-white employment and earnings over the past half century have captured the attention of several of the PRC senior researchers. Jim Heckman has approached this topic from several perspectives, including the evaluation of social welfare programs like job training and anti-discrimination laws, the measurement of returns on investments in post-school-aged workers, the basic functioning of the labor market, and the labor supply decisions of workers in the context of differential family structures. Kevin Murphy has also been a student of black-white earnings differentials throughout his career. Derek Neal found evidence that suggests that racial gaps in potential wages are much larger among men than among women. However, when Neal estimates this gap in potential wages using the NLSY data, he finds that the gap in 1990 for women is about two-thirds as large as for men - considerably larger than other studies have suggested. He tracks the trends on labor force participation of women and shows how selection issues into the market might have resulted in the impression of greater equality among women than was in fact the case. In Neal's other research he suggests the subtle ways in which the federal government's economic policies in taxes (so-called "marriage penalties") and transfer programs alter the attractiveness of marriage.
Bruce Meyer has studied the effects of Unemployment Insurance on labor market patterns and how welfare policy influences the behavior and well-being of working mothers. Meyer has undertaken an assessment of the use of income or consumption measures of the well-being of single mother families, concluding that consumption offers a better indicator. Ariel Kalil studies the effects of parent's employment and welfare transitions on child development. Other faculty members have projects assessing aspects of TANF, employment, and family preservation. In assessing some of the changes in the recent welfare "reforms", Diane Whitmore has written on the issues of food stamps. She has also been investigating the nutritional quality of school lunch programs and their effects on obesity. In her work on food stamps she investigated the spending habits, consumption patterns, and nutrient intake of families on an experimental program that replaced the food stamps with an equivalent amount of cash (called a "food stamp cash-out" experiment), finding that most recipients do in fact treat their food stamps in much the same way they expend the actual cash. Prior to the experimental evidence, studies had found that recipients purchased more food when the stamps were in the form of welfare support, while her work with the experimental data indicates this is not so. She suggests that the two findings are reconciled by the nature of the linear functional form typically assumed in the earlier studies. As described above, the research of Judith Levine, Larry Hedges, Robert LaLonde, and Barbara Schneider also address the policy topics related to labor markets, employment, and family preservation.
Methods and Tools
Throughout the past decades, PRC research associates have frequently generated new statistical tools and concepts as well as new public-domain survey data. Jim Heckman (analysis of censored data using survival analysis and other techniques) and Larry Hedges (meta-analysis) have made seminal contributions to the statistical toolkit used by demographers and have been acknowledged for their leadership in statistical methods. Many other PRC research associates are continuing to contribute statistical tools and guidance in the uses of new techniques. Heckman addresses the issue of how to deal with the fact that the coefficient and the random variable may be correlated, as when the rate of return to schooling is in fact correlated with the level of schooling attained. Will Manning examines how well alternative estimators behave in terms of bias and precision in log models designed to deal with skewed outcomes. Manning compares the results one would get using a variety of commonly used statistical methods, applied to data on patient's length of stay in hospitals. Several statistical models, based on alternative assumptions of the underlying data generating mechanisms are formulated and compared using data on length of stay, in particular, the OLS, gamma, Weibull, and Cox proportional hazard models. No single alternative model performs best under all the conditions examined, and guidelines for their uses are offered.
Kaz Yamaguchi describes a method of using multinomial logit latent-class regression models in sociological research, the categorical analogue of latent-variable regression for continuous latent variables like those used in LISREL. The models are, he points out, regression extensions of log-linear latent-class models with group variables.
Colm O'Muircheartaigh, with colleagues, addresses the methodology of the quality of attitude data when the respondent is offered only two or three choices on a rating scale. Arguments have favored both strategies - whether the third alternative offers an "easy out" for fence-sitters. Based on their structural equation modeling of responses about attitudes toward science and technology, O'Muircheartaigh and colleagues find that the middle alternative reduces random measurement error in responses, increases reliability, and does not affect the validity of the measurement. They did find some evidence of "acquiescence response bias," stronger among the older, and the less-educated, and the female respondents.