In an examination of the impact of black race on earnings in the US, comparing African blacks (immigrants) with native-born blacks & whites & white immigrants, it was theorized that ethnic stratification, vs assimilation, better explains the performance of African immigrants in the US labor market. Ordinary least squares multiple regression models were fitted to data from the 1980 & 1990 US population censuses. Results showed that, in both years, African immigrants had the highest educational attainment among the four groups. Despite this educational advantage, analysis by race & immigration status showed that, in 1980, earnings returns to education for Africans were the lowest for African immigrants followed by native born blacks, foreign-born whites, & native-born whites. By 1990, African immigrants had achieved earnings parity with their native-born black counterparts. Earnings returns to education for Africans were, however, less than those among native & foreign-born whites. A racial hierarchy in earnings is observed, with whites at the top, followed by African Americans & Africans. Findings in both 1980 & 1990 held even after taking human capital characteristics into account, & controlling for possible confounders, eg, duration of US residence, English language proficiency, sex, marital status, fertility, health, & labor supply. It is concluded that African immigrants & native-born blacks are discriminated against in the labor market, & the two groups are uniquely disadvantaged by their race.

This paper reports on first-year results for an evaluation of the New York School Choice Scholarships Foundation program, which was designed to provide scholarships for children from low-income families currently attending public schools to transfer to private schools. Families won scholarships through a lottery. Researchers examined data from scholarship and nonscholarship students' scores on the Iowa Test in Basic Skills in reading and mathematics and from parent/caretaker surveys regarding their children's school experiences. Results indicated that after 1 year, students who received scholarships scored higher in math and reading; parents of scholarship users were much more satisfied with their children's education; scholarship students were being educated in smaller schools and classes and were being asked to do more homework; parents of scholarship students reported more frequent school communications; 75 percent of those offered scholarships used them; and families of scholarship applicants were similar to those eligible for participation in terms of income, but were more likely to be dependent on governmental assistance and to be African American. (SM) Reproductions supplied by EDRS are the best that can be made from the original document.

During the four decades extending from 1950 to 1960, the ratio of the farm to nonfarm wage fluctuated between 0.45 and 0.55. We employ the Oaxaca technique and the Public Use Microdata Sample of the 1990 Census to decompose the farm/nonfarm wage differential. We also use the American Chamber of Commerce Research Association (ACCRA) cost-of-living data to measure the impact of the farm/nonfarm cost-of-living index on the wage gap. First, we find that the observed wage gap is reduced by half when nominal wages are deflated by the (ACCRA) cost-of-living index. Second, by applying the Oaxaca method separately to a subsample of young employees and a subsample of old employees we find that while for the subsample of employees younger than thirty-five years the real wage gap practically vanishes, the gap for the subsample of employees who are thirty-five years of age and over is higher than that of the total sample. This finding lends strong support to the hypothesis that between 1950 and 1970 the process of farm out-migration was positively selective.

Rapid advances in computing are changing virtually every aspect of scholarly work in sociology. Most of these advances offer opportunities for sociologists to improve the quality of their work and bring new insights and approaches to bear on important problems. This discussion looks at the overall state of progress in applying computing technology to sociology, identifying areas or underdevelopment, and suggesting some future trends.

Challenges the conventional wisdom that the majority of African Americans in the postbellum South were sharecroppers or tenant farmers working family-sized farms. Compares the agricultural and population manuscript census schedules for two counties in the Mississippi-Yazoo Delta - Washington and Tunica - for evidence that discrepancies between expected and enumerated numbers of tenant farmers are the result of enumerator error or misinterpretation. Enumerator interpretations show bias toward both overcounting and undercounting African-American farms. Even with the most generous interpretations of enumerator designations of African-American farmers, fewer than half of African Americans could be identified as workers of family-sized farms; most were agricultural and nonagricultural laborers.

Children take considerable time and effort to produce and their production is overseen by theirfamilies. As a consequence, family type may have a significant effect on child outcomes. One wouldexpect that the relative disadvantages of having unmarried parents would have diminished over the past few decades. The expansion of social welfare programs and greater social acceptance of alternate lifestyles should have reduced the burdens faced by non-married couple families; improved control over fertility should have reduced the likelihood of children being born into disadvantaged environments.I present evidence that the opposite is true: the average difference between children whose parents are married and those whose parents are not has increased. This increase is consistent with an asymmetric information model of marriage (Murphy 1999). The apparent increase in the penalty faced by children growing up outside of married couple families rejects a composition effect: the pool of surviving marriages has changed. Expanded AFDC, lower divorce costs, and smaller penalties associated with out-of-wedlock births allow women to more easily avoid or escape bad marriages; as a consequence, surviving marriages are better on average.

This dissertation is a sociological and statistical analysis of the institutional and social meaning of race as a concept and an identity. It uses an interdisciplinary approach which integrates census-based socio-historical research and demographic methods to argue that American institutions have helped to construct the idea of racial difference in ways that reinforce a system of social inequality. This thesis is illustrated through a historical analysis of the measurement of racial categories used by the U.S. Census Office of 1880, and how it was embedded in the social institutions that produced the idea of racial difference and linked the privileges of citizenship and American identity to 'whiteness.' This dissertation identifies the social and ideological components of racial classification in the United States, and analyzes the significance of non-biological factors in the perception of an individual's racial status by other persons. The concept of social race developed by Charles Wagley and the theory of racial formation developed by sociologists Michael Omi and Howard Winant are used to analyze the influence of social and economic indicators in the process of racial categorization. The longstanding use of color and/or racial classification systems in the United States Census of the Population is evidence of the importance of perceived and legislated racial differences to the institutional and social processes of the construction of American identity and statehood. Current popular, academic, and governmental debates about attempts to include a separate 'biracial' or 'multi-racial' category in the 2000 Census are but the latest version of the ongoing struggle to affect the significance of race by changing its boundaries. In 1880, the Census Office used five racial categories: 'White,' 'Mulatto,' and 'Black,' 'Indian' and 'Chinese.' I hypothesize that Census enumerators used not only biological characteristics but also social characteristics in classifying individuals. More specifically, I argue that lower status individuals would be more likely to be classified as 'mulatto' than 'white' and that, because women have lower status than men, 'mulatto' category would have an excess of women. Moreover, statistical evidence shows that other characteristics such as a low status occupation and illiteracy are associated with being classified as 'mulatto' rather than 'white.' In sum, I demonstrate that a category that most people take for granted as 'natural' and unproblematic was maintained in a way that reinforced the privileged social position of persons classified as 'white.'

Historians interested in the historical experiences of ordinary Canadians face a considerable challenge. Letters and diaries, company and association reports and minutes, government records, and contemporary newspapers, magazines and literature shed light on prescribed ideals, social discourse, and, to a certain extent, actual behaviour. However, since the 1960s, historians have drawn upon routinely-generated sources such as parish registers and census enumerations to gain a long-term and systematic perspective on individual and family behaviour.2 The major challenge faced by these historians has been to transform such extensive and detailed sources into the more manageable form of machine-readable data. The creation of historical microdata allows researchers to take advantage of computer processing and statistical analysis to study the interrelationships among individual and family characteristics. The purpose of this paper is not to assert the validity and usefulness of such projects, which has been debated elsewhere, nor to . . .

I propose and estimate an alternative measure of [U.S.] old-age coresidence with an adult child: the number of life-years lived in coresidence with an adult child while age 65 or older. In addition, I measure the share of lifetime lived in this state.... After I estimate the measure, I decompose it to determine the degree to which changes in the number of life-years lived in old age are due to changes in the rate of coresidence, changes in mortality, and changes in fertility.

In the past 50 years, structural changes have occurred in the social institutions of work and the family, changing our perceptions of work and retirement. At the same time, increases in life expectancy and better health at older ages suggest that adults currently at mid-life will live longer and healthier lives, a growing proportion of which will be without child- or job-related commitments. Data from the Health and Retirement Survey (HRS) and Integrated Public Use Microdata Series (IPUMS) are used to describe the ways in which the family and work characteristics of those in later mid-life have changed over the past 50 years and to examine the relationship between family change work change, and mobility in later mid-life. This research extends the literature on mobility, family, and work in two ways. First, while previous studies have generally focused on the differences between stayers and local movers (or stayers and long distance migrants), this study examines all three possibilities (stayer, local mover, long distance migrant). Second, this research examines the relationship between two later mid-life transitions--specifically, changes in the presence of adult children in the home and changes in work status--and mobility behavior. The results suggest that both family and work play important roles in shaping mobility in later mid-life. Although we tend to assume that changes in the family life cycle are associated with local mobility, this relationship appears to be more complex for adults in later mid-life who have adult children leaving home: Changes in the presence of adult children in the home and work status change are both significant predictors of long-distance mobility. These results extend the existing literature on mobility by demonstrating differences in the effects adult children at home and work status change on local and long-distance mobility. The results also demonstrate that these two factors have relatively independent effects on mobility behavior. Future research should focus on the relationship between life course change and mobility in later life so that researchers and planners may better understand the forces shaping the spatial distribution of older persons.

The authors analyze the Mexican household in the United States, emphasizing its complexity, which is expressed in the traits that distinguish it from both the original country and the new one, as well as in its changes over time. The authors seek to understand the composition of relatives within exiled families, their life-cycle evolution, and the interaction of civil and migratory calendars. They also consider household changes related to the migratory process, changes in marriage and birth civil calendars, changes produced by work or marital events, and the type of migratory decisions developed within families.

The share of elderly widows living alone rose from 18 percent in 1940 to 62 percent in 1990, wile the share living with adult children declined from 59 percent to 20 percent. This study analyzes the causes of this change and finds that income growth, in particular increased Social Security benefits, was the single most important factor causing the change in living arrangements, accounting for nearly two-thirds of the rise in the share of elderly widows living alone. Changes in benefits from the mean-tested OAA/SSI programs had a lesser impact on the decision to live alone but were a significant factor in explaining changes in the living arrangements of the poorest widows. Furthermore, contrary to recent work, we find no evidence that the effect of income on living arrangements became stronger over the period; income had a substantial positive effect on the propensity to live alone as early as the 1940s and 1950s. Finally, the substantial changes observed in the composition of the population with respect to age, race, immigrant status, schooling, and completed fertility explain a relatively small share of the changes in living arrangements.

The relationship between migration experience and family patterns among residents of the North and West [of the United States] is examined for three time periods--1940, 1970, and 1990. In general, an inverse association is observed between duration of residence in the North or West and family stability among African Americans. Although selective return migration to the South contributes to this association, it can account for only a minor part of the variation in family patterns by migration history. It is concluded that there is no evidence to support previous assumptions that southern migrants carried a dysfunctional family culture with them to the North and West, and thereby destabilized the nonsouthern African American family.

The relationship between migration experience and family patterns among residents of the North and West [of the United States] is examined for three time periods--1940, 1970, and 1990. In general, an inverse association is observed between duration of residence in the North or West and family stability among African Americans. Although selective return migration to the South contributes to this association, it can account for only a minor part of the variation in family patterns by migration history. It is concluded that there is no evidence to support previous assumptions that southern migrants carried a dysfunctional family culture with them to the North and West, and thereby destabilized the nonsouthern African American family.