We the People: One Nation, a Multicultural Society.

ABSTRACT

THIS ARTICLE EXAMINES SELECTED ECONOMIC and social indicators of our multiracial and multiethnic society at the end of the twentieth century to provide an information base for wise decisions about effective library services. The statistical profile describes the demography, economic well-being, and educational attainment of the principal racial and Hispanic origin groups. The data show that progress in our nation's well-being has occurred, but a great deal remains to be done to achieve the goals of equity and equality of opportunity.

INTRODUCTION

This article is a statistical profile of our nation at the end of the millennium. It examines selected economic and social indicators of our multiracial and multiethnic society that describe the journey our nation has taken and the distance that remains to achieve the goals of equity and equality of opportunity. An alternate title of this article could be "Correlates and Predictors of the `Digital Divide'" and yet another title: "What Must Be Done?"

Part one briefly discusses the origins of U.S. statistics on race and ethnicity, how these statistics represent membership in the polity, and caveats about the data. Part two describes the demographic characteristics of our nation--immigrants all, an extraordinary tapestry of races, ethnic groups, and cultures--and what our nation is projected to become by 2050. Parts three and four summarize a number of indicators of progress and disparities in the economic well-being and educational attainment of the major racial and Spanish-speaking populations; they focus our attention on distributional issues in the society.

Creating a statistical profile requires choices about which measures to include and exclude from a wealth of information collected by the U.S. government. This study was guided by a theoretical model that assumes the unequal distribution of life chances and creates a framework for understanding causes, effects, and outcomes. This model posits success (attainment) as a function of a complex array of related variables that include family background and choices made by society (through government), parents, and young adults (Haveman & Wolfe, 1994; Robbin, 2000c).

The statistical profile concentrates on the contribution that immigrants have made to the changing demography of our nation and on the family and its progress over the life course. The family is the single most important social agent for producing human capital, and its well-being significantly affects whether other social institutions have sufficient resources to carry out their missions. This conception of the determinants of attainment forms the basis for a large number of data and record keeping systems that are designed to carry out the constitutional, legislative, and administrative mandates of government as well as programmatic functions of non-governmental institutions, including the library.

How then, and in what ways, can the library, as an influential social institution, contribute to the well-being of people? The concluding remarks offer a cautionary note about the hype induced by the "Digital Divide" metaphor and whether social institutions can radically alter the choices that people make in their everyday lives. It seems indisputable, however, that, as a nation, social spending for educational resources must be significantly increased, in particular, to improve parental and child literacy. The data speak volumes about how education creates social capital, and it is here where the library community can make its most significant contribution.

ORIGINS OF RACIAL AND ETHNIC DATA IN THE NATIONAL STATISTICAL SYSTEM

Historical conceptions of race and which persons fall into majority and minority groups are fundamental to understanding the collection and reporting of racial and ethnic group statistics. Federal statistics on race and ethnicity are the product of more than two centuries of legislative initiative and public law that are policy responses to politics and social and economic conditions. The U.S. Constitution, hundreds of treaties, federal and state statutes, Supreme Court rulings, and case law have established the legal and political relationship of racial groups in the United States. Policies related to property rights, political representation, and citizenship in the polity have motivated administrative practices for collecting data on race and ethnicity (Anderson, 1988; Starr, 1987; Thernstrom, 1978).

Statistics on race and ethnicity have reflected the official record keeper's assessment of a social group's identity, status in society at a particular historical moment, and recognition of membership in the polity (Robbin, 1999). Government statistical systems that record race and ethnicity have created or erased social identity and excluded or included groups in the body politic (Robbin, 2000b). Until recent decades, immigration and naturalization laws contained explicit language that identified particular ethnic groups in racial terms and as ineligible for citizenship. At one time, the Irish, Syrians, and Spanish-speaking populations were considered "races" until they were officially designated an "ethnic" group (American Anthropological Association, 1997; Hayes-Bautista & Chapa, 1987; Ignatiev, 1995; U.S. Bureau of the Census, 1989). Federal statistics have created a similarity of identity where none existed, as with "Latino" identity based on shared language rather than culture and as with "Asian" or "Oriental" identity based on shared discrimination and ethnic stereotyping (Espiritu, 1992, pp. 13-14; Fugita & O'Brien, 1991; Oboler, 1992).

STATISTICS AND REPRESENTATION IN THE POLITY

Representation in the decennial census and administrative data and reporting systems has meant entry to the "political arena to have needs addressed and grievances resolved" (Feeney, 1994, p. 3). Being counted is the first step toward making claims for entitlements to resources and participating in the public policy process. Groups must be counted in order to make credible claims for political representation, demonstrate discriminatory practices against them, seek and obtain legal remedies, receive governmental assistance for a host of social programs, and evaluate current, as well as develop new, public policy.

National statistics on American Indians and Blacks, for whom data had been collected since the 1790 decennial census, constituted the most developed series on minority populations until the 1980s. Vital statistics records have historically identified a category related to the Black population. The decennial census began counting "Orientals" in the 1870 census, but restrictive immigration, exclusion, and right-to-work laws contributed to their numerically small size until after 1965. The concentration of Asian populations in only a few geographic areas of the country greatly limited their visibility and contributed to rationalizing the lack of statistics on these population groups until recent decades.

The status of the Spanish-speaking populations was established early in the history of the Republic by the Monroe Doctrine and Treaty of Guadalupe Hidalgo of 1848 and codified in a host of exclusionary federal and state immigration, naturalization, employment, taxation, education, and segregation laws. Statistical evidence of Spanish-speaking and Latin American identity became available in national statistics beginning in the mid-1800s and, specifically, in the 1930 decennial census on the Mexican. A key reason why government agencies did not collect statistical data on the Latino and Hispanic populations was the belief that they were geographically concentrated in only a few regions of the country (Robbin, 2000b).

The U.S. government first collected information on citizenship in the 1820 and 1830 censuses, again in 1870, and then from 1890 through 1990. During the mid-1800s, the U.S. government began collecting information on place of birth for both persons and their parents. During different periods in U.S. history, the census has collected information on ancestry and language, a response to changes in the demographic composition of the nation that were often perceived as threats to the body politic.

"Great Society" initiatives of the 1960s fostered the rise of identity politics and the entrance of minority population interest groups into the political arena. During the 1970s, African-Americans, Latinos, Asians, and Pacific Islanders mobilized to influence Congressional and administrative policy decisions about federal statistics on race and ethnicity (Robbin, 2000b). Minority statistics became an integral part of the federal statistical system because of their efforts.

The federal government issued "Statistical Policy Directive 15" in 1977 and "Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity" in 1997 (Edmonston et al., 1996; Robbin, 2000a). The 1977 version of the classification system established four racial categories (American Indian, Asian or Pacific Islander, Black, and other) and one ethnic group category (Hispanic/not of Hispanic origin). People were permitted to select only one category to identify their racial heritage. Racial and ethnic self-identification was the recommended way of collecting data, but observer identification was permitted.

Two major changes took place in the 1997 revision that were implemented in the 2000 decennial census. The "Asian or Pacific Islander" category was split into "Asian" and "Native Hawaiian or Other Pacific Islander" for a total of five categories. Recognizing the diversity and growing multiracial nature of American society, respondents were permitted to identify multiple racial identities. The official publications of the Census Bureau will report on sixty-three different combinations, but there will not be a category labeled "multiracial." It is important to emphasize that both the original 1977 and revised 1997 directive cautioned that the standard was not to be used to determine eligibility for participating in any federal program, nor were the categories to be construed as representing biological or genetic racial origins.

The directive became the foundation for collecting and reporting data on racial and ethnic groups in the U.S. population and ushered in a conceptual change in the official definition of race. Although the status differences of the White majority and various minority groups continued to be maintained, the "effect of the standard was to redefine the U.S. population beyond a White and non-White classification" (Lott quoted in Review of Federal Measurement, 1993, p. 38). Even though there were difficulties in how people interpreted the meaning of "race" and "ethnicity," the classification system provided evidence that the United States was no longer a "White" majority and "Black" minority society; the nation had become a rich tapestry of multiracial and multiethnic heritage.

In the decades that followed the issuing of Directive 15 in 1977, the classification system provided policy makers with a powerful tool for summarizing and tabulating a vast quantity of statistical data. Policy makers and administrators documented disparities and differentials in income, education, health, and access to information. A host of social welfare issues related to racial and ethnic identity found their way onto the public policy agenda. Constitutional, legislative, and administrative mandates of government, including the programmatic functions of non-governmental institutions, resulted in the development of extensive administrative record-keeping systems to register the health and welfare of social groups in the U.S. population. It is these record-keeping systems that illuminate our understanding of the demography and well-being of American society.

CAVEATS ABOUT THE PUBLISHED STATISTICAL DATA

There are at least four caveats about the statistics that must be stated. First, reflecting our history as an immigrant nation, "We The People" are a rich tapestry of nationalities, ancestries, and cultures. However, this statistical profile frames the discussion in terms of the official racial and Hispanic origin categories assigned by the government employing the rules issued under Statistical Policy Directive 15 issued in 1977 (i.e., four categories for race and one ethnic group category and self-identification for only one racial category).

The statistical profile thus reflects population aggregates for racial and Hispanic origin groups, although there are significant socioeconomic and demographic differences within all the groups that are related to their country of origin, time of arrival in the United States, incorporation experiences, generation or cohort, and membership in different ethnic groups. "Hispanic origin" includes Spanish speakers of any national origin, unless otherwise specified (the 2000 census will add "Latino" and "Chicano" as designators). There are, of course, distinct cultural, ethnic, and other differences that reduce the utility of the "Hispanic" label.

Similarly, the category label of"African-American" or "Black" masks a highly diverse population of recent immigrants from Caribbean and African nations, as well as the native born who experienced a history of slavery. The category "Asian" does not reveal the significant demographic and economic well-being differences in the very heterogeneous populations of Southeast Asia and the Pacific Islands (Hernandez, 1996). The category "American Indian" includes Eskimos and Aleuts, who have different histories from the American Indian population. Again, it is important to emphasize that this article is not designed as a comparative analysis of specific ethnic or nationality groups unless they are particularly significant. Nonetheless, these differences are critical for developing appropriate national and local library and other policies.

Second, much of our understanding of who we are as a nation derives from the decennial census, but the 2000 census was conducted after this article was written. Thus, the 1990 decennial census--the basis for government and non-governmental sampling frames--provides information that is somewhat degraded by its lack of recency. Nevertheless, what is important is the consistency in the trends and their direction and the differences in the indicators among the racial and ethnic-origin groups.

Third, the data reported here are from statistical series and record-keeping systems whose data collection and processing methodologies, data quality, procedures for reporting, time periods, and other factors result in numbers that are different, even when the published report describes the same subject matter. Projections of the racial and ethnic composition of the population may differ because, for example, one series derives from the complete enumeration of a decennial census and another series from sample data of the Current Population Survey. Even when estimates are drawn from the same series, such as the Current Population Survey, they may be issued in published tables whose numbers may have been computed with new information, assumptions, or methodology. In some cases, "large" differences may result, particularly when a new methodology is employed. It becomes essential to read the methodological reports in order to understand the differences. To the extent possible, the data presented in the following tables are consistent, and every table cites the original source of the statistics, so the reader can return to the original published reports and tabular data.

Fourth, although government surveys usually have sample sizes that are much larger than polls or surveys conducted by academic researchers or professional polling firms, sample surveys may have too few people of a particular racial or ethnic heritage to provide statistically significant results. Consequently, detail on a particular group will often not be provided in published reports, in particular, for the American Indian and Asian populations. The Aleut and Eskimo populations in the "American Indian" category and Pacific Islanders in the "Asian" category are also affected by extremely small sample sizes.

Typically, information is provided for the "White," "Black" ("African-American"), and "Hispanic" populations. Sometimes, publications provide more detail on the "Hispanic" population ("White" or "Black"), and the "White" population is further disaggregated as "White, non-Hispanic" and "Black non-Hispanic." The category "Asian" includes Pacific Islanders and is referred to as "API." "American Indian" includes the Eskimos and Aleuts.

Finally, a word about naming conventions. Many publications continue to refer to the "White" population as the "majority" race and all others as either the "minority" or "non-White" population. I use the naming conventions employed by the source I cite, although I place double quotes around the name, as in "non-White" when the reference is to "minority" populations. Official government documents and academic publications are inconsistent about whether "White" and "Black" are capitalized; however, this article capitalizes all racial group designations in order to make the discussion of racial and ethnic group differences easier to read. Percentaged data are subject to rounding when the text does not describe the data to a tenth of a percent.

OVERVIEW OF THE DEMOGRAPHIC COMPOSITION OF THE UNITED STATES: ON THE EVE OF THE MILLENNIUM THROUGH 2050

The United States grew from under 20 million people in the early part of the nineteenth century to almost 250 million people in 1990 (Gibson & Lennon, 1999). The U.S. population more than tripled in size during the twentieth century and, between 1970 and 1990, increased by nearly 50 million people. In March 1998, its population was estimated at 269,094,000 (U.S. Bureau of the Census, February 1996b).

The March 1998 Current Population Survey estimates that young people between 5 and 25 years old constitute about 30 percent of our total population, and the 35 to 44 and 45 to 54 year old cohorts about 38 percent. Of our nation's population, 70 percent is 21 years and over; 20 percent, 55 years and over; and almost 12 percent, 65 years and over. Married couples comprise somewhat more than 76 percent of the family household population, with about 18 percent headed by female single parents and 5.5 percent by male single parents. The "baby-boomer" generation and the elderly populations are two cohorts of the population that will constitute a much larger proportion of society during the twenty-first century (U.S. Bureau of the Census, February 1996a).

Educational attainment has changed markedly since the Census Bureau first collected information about schooling in 1940, when only 25 percent of the American people had completed high school and 4.6 percent had completed four or more years of college (U.S. Bureau of the Census, September 1998b, p. 25). Today, nearly 88 percent of the population has a high school diploma, 28.5 percent have had "some college" or received an associate degree, and about 27 percent have received a bachelor's or higher degree (U.S. Bureau of the Census, February 1999b).

On the eve of the millennium, our nation witnesses the longest running expansion of the economy in its history (Hershey, 1999; Uchitelle, 1999a, 1999b). More people are working than ever before, household income has climbed after remaining "flat" for most of the 1990s, the United States leads the world in the production of goods and services, consumer confidence is high, and poverty rates have declined. Selected indicators of social and economic well being for the principal racial and ethnic groups are examined below to determine whether all Americans have benefitted equally from these developments--i.e., does the "rising tide" carry everyone along?

CONTRIBUTION OF IMMIGRATION TO OUR NATION

Immigration has had major effects on the demographic composition of the nation, notably on the size of the foreign-born and "non-White" populations, number of interracial marriages, and the age distribution of the population (Gibson & Lennon, 1999). The immigrant share of the population has always been a function of immigration policy.

The Immigration and Naturalization Act of 1924 established permanent quotas on immigration and preferences for country of origin for immigrants from northern and western Europe, and the relative share of the population contributed by immigration decreased dramatically over the next several decades (Espenshade et al., 1996-97, p. 3). The 1965 Immigration and Nationality Act Amendments changed the rules for entry, including preferences, scope, magnitude, and country of origin, subsequently altering the size of the native and foreign-born stock, diversity and heterogeneity of the U.S. population, and geographic distribution of immigrants (Espenshade et al., 1996-97, p. 3). Among the most important changes made by these amendments were the family unification policy and elimination of national and ethnic group quotas. Education, skills, and national origin no longer determine whether people "get to pursue the American dream" (Nasar, 1999, p. BU6).

CHANGES BETWEEN 1970 AND 1990

At the time of the 1970 census, about 83 percent of the population was classified as non-Hispanic White; 11 percent Black/African-American; about 5 percent Hispanic; under 1 percent Asian; and .4 percent American Indian (Passel & Edmonston, 1994, p. 43). Twenty years later, the 1990 census enumerated a population classified as about 75 percent non-Hispanic White; 12 percent Black/African-American; 9 percent Hispanic; 2.9 percent Asian; and under 1 percent American Indian.

Between 1970 and 1990, the foreign-born stock increased from 4.7 percent to 7.9 percent of the population, about 9.6 million to almost 20 million people (Gibson & Lennon, 1999). By 1996, the foreign-born population was estimated to be 24.4 million people. Although the White foreign born remained about 5 percent of the population, immigration that originated in Latin America and Asia contributed to very large changes in the racial and ethnic composition of the foreign-born population.

Persons of Hispanic origin increased from about 20 to nearly 36 percent of the total foreign born population between 1970 and 1990. By 1996, about 11 million Hispanics constituted more than 44 percent of the foreign-born and about 38 percent of the total Hispanic population (U.S. Bureau of the Census, September 1998b, p. 45). Based on the 1990 census, Passel and Edmondston (1994, p. 52) calculated that almost two-thirds of the 1990 Hispanic population consisted of either immigrants who came to the United States since 1950 or were descendants of those immigrants. In 1990, Mexicans constituted the largest group, followed by persons of Puerto Rican and Cuban origin (U.S. Bureau of the Census, September 1999m).

Chinese, Japanese, and Korean constituted the largest Asian groups (46 percent), joined by Thai, Cambodians, Hmong, and Laotians (totaling about 8 percent) (U.S. Bureau of the Census, September 1993, p. 2). The Asian population increased from nearly 6 percent in 1970 to about 23 percent in 1990 of the foreign-born population. Blacks, many originating from the Caribbean and Africa, increased from 1 to 5 percent of the foreign-born population. "Races other than White" accounted for about 38 percent in the 1990 census, up from nearly 28 percent in 1970.

According to the 1990 census, 75 percent of the immigrant population is concentrated in only seven states with one-third of them living in California and residing "overwhelmingly" in urban areas (Espenshade et al., 1996-97, p. 3). About half of them live in only seven metropolitan areas: Houston, Los Angeles, New York, San Francisco, Miami, Chicago, and Washington, DC.

Immigration between the 1960s and 1980s contributed to modifying the composition of interracial marriages (U. S. Bureau of the Census, June 1998a, June 1998c). The 1970 census recorded about .7 percent of the total number of married couples and, in 1990, interracial marriages accounted for 2.7 percent of married couples, a nearly four-fold increase. By 1998, estimates indicate a more than 13 percent increase in interracial marriages with most of the interracial marriages between Whites and races other than African-Americans (U. S. Bureau of the Census, January 1999a). The number of children in interracial families also increased, and by 1990 represented about 4 percent of the more than 47 million children, a four-fold increase from 1970 (U.S. Bureau of the Census, June 1998f).

Immigration also contributed to changes in the age structure of the population. On the whole, the immigrant population is younger and with higher fertility rates than the native-born population. As immigration has rendered major changes in the demographic composition of our nation, it has also had significant effects on the services that institutions provide, "taxing some city resources like already overcrowded school systems" (Sachs, 1999, p. A24). Now, for example, the New York City public school system accommodates children of 196 different nationalities (Sachs, 1999, p. A24). Hollywood High School in Los Angeles records thirty-two different languages spoken by its student body (Swerdlow, 1999, p. 16). Immigration has also contributed to a growth in public library use, especially in major urban centers, making them the "busiest libraries" in the nation and leading to revisions in collection development policies (Toy, 1998, p. 30).

Although the U.S. statistical system does not collect data on religious affiliation, anecdotal evidence describes how immigration has also expanded the religious pluralism of the United States and transformed neighborhood religious life. For example, one block in Flushing, Queens in New York City--a microcosm of the world and probably the most heterogeneous small geographic area in the United States--is representative of a reality shared throughout the major urban centers of our country (Sengupta, 1999). Old Jewish synagogues share the neighborhood with Islamic mosques, Buddhist and Hindu temples, a Chinese evangelical church, a Sikh gurdwara, an African Methodist Episcopal parish church, and a Korean American Presbyterian church.

ECONOMIC EFFECTS OF IMMIGRATION

Immigration continues to provoke controversy and disagreements among thoughtful and well-informed people just as it did at the beginning of the twentieth century, about its contribution to our nation's economic well-being. Have immigrants contributed to the two decades of prosperity that the United States has experienced, or have they been responsible for the nearly stagnant wages that most Americans have experienced and also for the growing disparity between the have's and the have-nots? (Nasar, 1999). Have immigrants depressed opportunity and pay for the most disadvantaged, displacing African-Americans and other native-born at the lowest rungs of the economic ladder because they are willing to work hard for lower wages? (Borjas, 1999). Do immigrants create a burden that is shouldered by the native-born population because they have fewer years of schooling and fewer skills than the average native-born American? (Borjas, 1999). Or does the youthfulness of the immigrant population benefit the United States by reducing the effects of an aging work force?

These were some of the questions that the National Academy of Sciences Panel on the Demographic and Economic Impacts of Immigration considered in its deliberations during the mid-1990s (Smith & Edmonston, 1997). The panel concluded that there was a "wage gap between foreign- and native-born workers," but that "immigration has had a relatively small adverse impact on the wage and employment opportunities of competing native groups ... [and] is dispersed throughout the United States" (p. 7). Most affected by immigration were the immigrants themselves (p. 7). Yes, there is a wage gap, and immigration "has played some role in explaining the declining wages of high school dropouts, but it has played little part in the expanding wage inequality for any other group of native workers" (p. 7). Below is a very short discussion of the differences in the economic well-being of our foreign- and native-born populations.

PROJECTIONS OF THE DEMOGRAPHIC COMPOSITION OF THE UNITED STATES

Population projections are critical for developing sound public policy to meet the needs of our citizens. For example, we need to have a good estimate of the ratio of children and the elderly population to the working population because it is the latter that provides the financial support taxes for social, educational, library, medical insurance, and other programs.

Demographic tools are remarkable for their quality; nevertheless, it is still necessary to issue a caveat about the estimates and projections that are provided here. Projections are always subject to error. They are based on the best information that demographers have for the past and a set of assumptions about people's behavior. We know, however, that people do not always behave in the future as they have in the past. Thus, demographic projections are continually reassessed as new information becomes available, which will occur after the 2000 census.

The cumulative effects of births, deaths, net immigration, and interracial marriage are significant for the future racial and ethnic composition and age distribution of American society. The population is projected to be more than 310 million shortly after 2010 and nearly 400 million by 2050, and older than it is now (U.S. Bureau of the Census, February 1996a, p. 7). The forecast is that the United States will become a "nation of minorities," and no particular racial or ethnic group will dominate.

Table 1 shows the projected population change by racial and ethnic groups between 1995 and 2050. The Hispanic-origin population growth between 1995 and 2050 is projected to be close to 260 percent, while the Asian population growth during the same period is nearly 270 percent. In contrast, the White population is projected to experience only a 35 percent change, while the Black population, about an 83 percent change over the fifty-five years, and the American Indian population, a 95 percent change.

Table 1. Population Change by Race and Hispanic Origin: 1995 to 2050 (Middle series. As of July 1. Resident population).

                                    Race

Percent                                American
Change         Total   White   Black    Indian    Asian

1995 to 2050   49.9    35.1    82.8      95.0     267.1

                                 Not of Hispanic Origin

Percent        Hispanic                    American
Change         Origin(1)   White   Black    Indian    Asian

1995 to 2050     258.3      7.4    69.5      83.0     269.1

Note. Information includes all 50 states and District of Columbia but not residents of Puerto Rico. American Indian represents American Indian, Eskimo, and Aleut. Asian represents Asian and Pacific Islander. Source is U. S. Bureau of the Census, February 1996a.

(1) Persons of Hispanic origin may be of any race.

The White-Not-of-Hispanic origin will decline in size from an estimated 75.6 percent in 1990 to slightly more than 50 percent in 2050 but after 2030 would "contribute nothing in size to population growth" (U.S. Bureau of the Census, February 1996a, p. 14). Growth in the total White population after 2040 will be entirely due to White Hispanics and in the American Indian population due nearly all to natural increases (U.S. Bureau of the Census, February 1996a, p. 15). If current levels of interracial marriage continue, the proportion of the population of multiple ancestry will increase, "adding complexity and ambiguity to ethnic definitions ... and blurring ethnic and racial" identity (Smith & Edmonston, 1997, p. 4).

Birth trends also indicate significant age distributional changes in the total population. The White population will experience a decrease from about 78 to about 71 percent between 1995 and 2050. Slight increases from 17 to about 19 percent are projected for the African-American population. There will be a slight increase from 1 to 1.3 percent for the American Indian population. The most dramatic increase will be in the Spanish-speaking and Asian populations, more than tripling by 2050 (U.S. Bureau of the Census, February 1996a, p. 20). The Asian population is projected to grow from 4.1 to 7.6 percent. The Hispanic origin population will grow from 15.6 to nearly 33 percent.

AGE STRUCTURE

A growing proportion of the population will be elderly because of the maturing of the baby-boom cohort and "sharp declines in mortality at the adult and older ages in the recent past and the prospect of continuing low mortality" (U.S. Administration on Aging, February 1999, p. 4). Aging will be significant for the White population, now the oldest population group (median age 33.8 years) and projected to comprise the oldest group (median age nearly 40 years). The Hispanic-origin population, currently the youngest (median age 25.4), is projected to have a median age of 31 years by 2050. The median age of the African-American population will increase from 28 years in 1990 to nearly 33 years in 2050, and the American Indian population, from 26 years in 1990 to 31.6 in 2050.

According to Census Bureau projections, the elementary and high school population in the range 5 to 17 years will increase from an estimated 45.3 million in 1990, to 52.5 million in 2010, and to nearly 70 million by 2050 (U.S. Bureau of the Census, February 1996a, p. 9). This school age cohort will account for under 20 percent of the total population for each of the decades between 1990 and 2050. The college age 18 to 24 year old cohort will account for about 10 percent of the total population between 1990 and 2050. Asian and Hispanic families will contribute a larger percentage of children to the school systems than the other racial and ethnic groups.

It is, however, the 45 to 64 years and 65 years and older cohorts where large population increases are projected. By 2020, 45 to 64 year old cohorts will account for one-quarter of the total population, decreasing in size to about one-fifth by 2050. The population aged 65+ years will increase from all estimated 31.2 million in 1990 to a projected 47.8 million in 2020 and to a projected 85.5 million in 2050. The elderly population will more than double in size in absolute numbers between the 1990s and 2050, accounting for about one-fifth of the population beginning in 2030. The large projected growth of the very old, 85 years and older, is of particular concern to policy makers.

The "racial and ethnic composition of the elderly population is expected to change profoundly in the next 50 years" (U.S. Administration on Aging, 1999, p. 5). Projections indicate a growth from 4.5 percent in 1995 to 17.5 percent in 2050 for the Hispanic elderly population. The proportion of the elderly in the Hispanic, Black, and "Other Races" populations are "also expected to increase," but the "proportion of Whites in the elderly population will decrease from 90 to 82 percent" (p. 6). The "shift is even greater for the non-Hispanic White population, from 85 to 66 percent" (p. 6). What this means is "that in 2050 about one-third of the elderly population would be Black, Hispanic, or in the `Other Races' category" (p. 6).

CORRELATES AND PREDICTORS OF ECONOMIC SECURITY AND WELL-BEING

We live our lives in networks of social relations--families, friends, churches, schools, voluntary organizations, neighborhoods, and communities--that are responsible for, and contribute to, our well-being. Economic security of the household is essential for a host of quality of life issues, and family circumstances matter greatly, particularly parental income as a proxy for parental earnings capacity and educational attainment and as a predictor of exposure to the national information infrastructure. The well-being of families also affects the well-being of society.

Over the last fifty years, the United States has experienced growing income inequality due, in part, to changes in the composition of households and changing labor market conditions (Weinberg, 1996). There is concentrated advantage at the same time that there is concentrated disadvantage. Affluence is growing at the same time that many people experience poverty, and more people than ever do not have health insurance or adequate medical care.

The United States has seen large changes in living arrangements--no longer the "traditional" family of husband and wife but a shift to more single-parent households that "typically have lower incomes" (Weinberg, 1996, p. 4). There are also increased demands for highly skilled and well-educated college graduates but fewer opportunities for low-skilled and low-educated workers.

Economic well-being is differentially distributed among the racial and Spanish-speaking populations. The "Digital Age" has made more visible the fault lines of our society (Mueller & Schement, 1996; U.S. Department of Commerce, 1995, 1997, 1999).

HOUSEHOLD, FAMILY, AND PER CAPITA INCOME

Household income has risen significantly over the last decades principally because women joined the labor force in ever increasing numbers. In recent years, the nation has seen an increase in the median income of households (U.S. Bureau of the Census, September 1999m).(1) Even as income has risen, however, there remain large differences among the principal racial and Hispanic-origin groups (U.S. Bureau of the Census, March 1998).

In 1997, about 45 percent of the Asian population, in contrast to 40 percent of the White Non-Hispanic origin population, has a household income of $50,000 and higher (see Table 2). Data collected in March 1998 indicate that nearly 50 percent of the Hispanic and African-American households have incomes below $25,000 in contrast to about 30 percent of White households. About 20 percent of the Hispanic origin and about 24 percent of African-American populations are at or below the poverty line.

Table 2. Annual and Median Household Income by Race and Hispanic Origin in 1997 and 1998

                                  1997

                                               White,
                                               Not of
                                               Hispanic
                     All Races    Asian(1)     Origin
Household Income     % of Total   % of Total   % of Total

Number of
Households
(in thous.)           102,528       3,125       77,936
Under $5,000              3.4         4.7          2.5
$5,000 to $9,999          7.6         5.0          6.2
$10,000 to $14,999        8.1         6.4          7.5
$15,000 to 24,999        14.8        10.4         14.0
$25,000 to $34,999       13.3        10.8         13.0
$35,000 to $49,999       16.3        17.3         16.4
$50,000 to $74,999       18.1        19.2         19.5
$75,000 and over         18.4        26.1         20.8
Median Income         $37,004      45,248       40,576

                                  1998

                     All Races    Black        White
Household Income     % of Total   % of Total   % of Total

Number of
Households
(in thous.)           103.874      12,579       87,212
Under $5,000              3.2         7.3          2.6
$5,000 to $9,999          7.0        14.1          6.1
$10,000 to $14,999        7.8        10.6          9.0
$15,000 to 24,999        14.0        17.4         13.7
$25,000 to $34,999       13.2        13.5         13.2
$35,000 to $49,999       16.0        14.3         16.3
$50,000 to $74,999       18.6        13.2         19.3
$75,000 and over         20.1         9.0         21.3
Median Income          38,885      25,351       40,912

                       1998

                     Hispanic
                     Origin(1)
Household Income     % of Total

Number of
Households
(in thous.)             9,060
Under $5,000              4.7
$5,000 to $9,999         10.1
$10,000 to $14,999       10.9
$15,000 to 24,999        17.9
$25,000 to $34,999       16.3
$35,000 to $49,999       15.8
$50,000 to $74,999       14.0
$75,000 and over         10.2
Median Income          28,330

Note. Percentages calculated from raw numbers. Adapted from U.S. Bureau of the Census, July 1999i for 1997 income; U.S. Bureau of the Census, September 1999m for 1998 income.

(1) Hispanic origin may be of any race.

Household, family, and per capita median incomes in 1998 differ significantly among racial and Spanish-speaking groups (Table 3). African-Americans and Hispanic origin (all races) populations have much lower household median incomes than the Asian and Non-Hispanic White groups. It is important to recognize, however, that the higher API household income masks the fact that their household size is larger than the other racial groups (U.S. Bureau of the Census, July 1999i); thus, a better estimate of economic well-being for the Asian population is its per capita median income of $18,700, which is below the Non-Hispanic White population median per capita income. The Non-Hispanic White population is nearly double the per capita income of the Hispanic-origin (all races): $22,952 to $11,434. These large differences in family income across racial groups have not changed over the last forty years (U.S. Department of Education, 1996a, p. 45).

Table 3. Household, Family, and Per Capita Median Income by Race and Hispanic Origin: 1998 Households and people as of March of the following year. Income in current dollars)

                              Households            Families

Race and Hispanic
Origin(1)

                           Number     Median    Number     Median
                           (thous.)   Income    (thous.)   Income

All Races                  103,874    $38,885   71,551     46,737
African American/Black      12,579     25,351    8,452     29,404
Asian & Pacific Islander     3,308     46,637    2,381     51,860
White                       87,212     40,912   60,077     49,023
  Non-Hispanic White        78,577     42,439   52,871     49,640
Hispanic Origin(1)           9,060     28,330    7,273     29,608

Race and Hispanic
Origin(1)
                                 Persons

                           Number
                           (thous.)   Per Capita

All Races                  271,743    $20,120
African American/Black      35,070     12,957
Asian & Pacific Islander    10,897     18,709
White                      223,294     21,394
  Non-Hispanic White       193,074     22,952
Hispanic Origin(1)          31,689     11,434

Note. Data for American Indians, Eskimos, and Aleuts populations not tabulated from the CPS due to small size. Table adapted from U.S. Bureau of the Census, July 29, 1999i for "Income, All Races, African American/Black, White, and Hispanic;" (Table 9). U.S. Bureau of the Census, July 29, 1999j for "Income, Asian."

ROLE OF THE FAMILY

The social context of the family is essential for ensuring the economic well-being of children. Family resources contribute to the cognitive and social development of the child (Gamoran et al., 1999). The presence of two-parent family households is critical because children who live in two-parent families experience significantly lower levels of poverty than children in female-headed families with no husband present.

Economic security depends on having wage earners in the family in order to meet basic needs, pay bills (such as rent or a mortgage), and obtain medical care. Income provides choices about the neighborhoods where families will reside, and the affluence of neighborhoods and the surrounding community appear "instrumental in allowing many neighborhoods to achieve an efficacious environment" for meeting children's needs (Sampson et al., 1999, p. 656).(2) The key to ensuring economic security and well-being is parental educational attainment.

LIVING ARRANGEMENTS

Table 4 describes the difference that family structure and the age of the head of household make in terms of income in 1998 for all racial and Hispanic origin populations. Female-headed households make up about 45 percent of African-American and about 24 percent of Hispanic-origin families in contrast to about 11.6 percent for Asian families and 15 percent for White families.

Table 4. Median Income of Racial and Hispanic Origin Populations by Family Type and Age of Householder: 1998 Households as of March of the following year. Income in current dollars)

                           All Races             Black

                       Number    Median    Number    Median
Characteristic        (thous.)   Income   (thous.)   Income

Family Type

All families
 (all types)            71,551    46,737    8,452     29,404
Married
 couples                54,770    54,276    3,979     47,383
Male householder,
 no wife present         3,976    39,414      660     27,087
Female householder,
 no husband present     12,789    24,303    3,813     16,770

Age of
Householder
Under 65
 years                  60,053    50,259    7,489     30,946
65 years
 and over               11,498    31,568      962     22,102

                                              Hispanic
                            White             Origin(1)

                       Number    Median    Number    Median
Characteristic        (thous.)   Income   (thous.)   Income

Family Type

All families
 (all types)            60,077    49,093    7,273     29,608
Married
 couples                48,461    54,736    4,945     34,816
Male householder,
 no wife present         3,087    37,788      600     29,227
Female householder,
 no husband present      8,529    25,175    1,728     16,532

Age of
 Householder

Under 65
 years                  40,646    52,550    6,683     30,363
65 years
 and over               10,231    32,398      591     21,935

                            Asian(2)

                       Number    Median
Characteristic        (thous.)   Income

Family Type

All families
 (all types)            2,381     51,850
Married
 couples                1,946     55,864
Male householder,
 no wife present          156     44,010
Female householder,
 no husband present       278     30,303

Age of
Householder

Under 65
 years                   2147         NA
65 years
 and over                 234         NA

Note. Data for American Indians, Eskimos, and Aleuts populations are not tabulated from the CPS due to their small sample size. Totals for "All Races" apply only to data for African American, White and Hispanic origin. Data for Family Type for "Asian" are calculated on a different base, but are unavailable for "Age of Householder." Table adapted from the U.S. Bureau of the Census, September 1999 ("Family Type" and "Educational Attainment" for All Races, Black, White, and Hispanic); U.S. Bureau of the Census, July 29, 1999i, j, k ("Type" for Asian).

(1) Hispanic-origin may be of any race.

(2) Derived from percentaged data.

Among all races, married couple families are significantly better off than single parent families, and male-headed families fare significantly better than female-headed ones. The median income for married couple families is $54,276; for male-headed families, $39,414; and for female-headed families, $24,303. These income differences are even greater when disaggregated by racial groups. Although Asian families have a higher median income than all other groups, there remains a significant difference in family median income between married couples ($55,864) and female-headed households ($30,303). White married couples have a median family income of $54,736, but the median income of female-headed Black and Hispanic-origin families is under $17,000.

The presence of children under 18 influences economic well-being. Families with children have more difficulty meeting basic needs than adults 60 and older (Bauman, 1999, p. 3). All families experience less economic security when children are under 18. Of all families in 1997, 10 percent--but only 5 percent of all married couples--with or without children under 18 years of age, had incomes below the poverty line (U.S. Bureau of the Census, 1998a). Families with children under 18 are, however, poorer (16 percent for all races, 7 percent for married couples).

PRESENCE OF WAGE EARNERS

The key to economic security is the family wage earner. Here, too, racial group differences are visible (U.S. Bureau of the Census, September 1999b, pp. 14-16). In 1998, White families with no earners have a median income of $22,672; with one earner, $34,486; and with two or more earners, $62, 695. The median income for Black families with no earners is $9,422; one wage earner, $20,524; and two or more wage earners, $51,737. Hispanic-origin families have similar median income with no wage earners ($9,574) and one wage earner ($20,548), but have significantly less ($42,679) than Black families. Of Asian families, 20 percent have three or more wage earners in the family compared to the national average of 13 percent (U.S. Bureau of the Census, September 1993, p. 8; U.S. Bureau of the Census, July 1999l).

LIVING IN POVERTY(3)

The experience of poverty is a significant stressor in the life of all family members but especially its children. Children whose families live in poverty experience more low birth weight births, higher infant mortality in the first year of life, lower rates of vaccination, and debilitating chronic conditions that limit activities (U.S. Federal Interagency Forum, 1998a). Growing up in poverty means that children are less likely to graduate from high school, more likely to be a teenage mother, and even less likely to continue education beyond secondary school.

There are significant differences among the racial and ethnic groups who live in poverty, and the effects of family structure and number of children under 18 years of age persist. Again, an intact family significantly reduces the risk of falling into poverty. Somewhat more than 8 percent of all White families with or without children under 18 experienced poverty in 1997, whereas the percentage of White married couples was under 5 percent. Of Black married couple families, 8 percent--but 30 percent with children under 18--were poor. Nearly 27 percent of Hispanic families live in poverty, but even Hispanic married couples experience high rates of poverty (17.4 percent).

Female-headed family households with children under 18 years of age experience high rates of poverty, and the racial group differences are large. Nearly 38 percent of White, 47 percent of Black, and 54 percent of Hispanic-origin female-headed families are poor.

The poverty of Hispanic families and married couples increases if children are under 18:30.4 percent and 21 percent, respectively. The incidence of poverty in the Hispanic population varies considerably by place of origin, however: Cuban children are far less likely to experience extreme poverty than Puerto Rican and Mexican children (U.S. Administration for Children and Families, 1999, p. 63).

Educational attainment of the household head is key to whether families are at risk for falling into poverty. Years of schooling is highly related to future labor market and earnings success. Families whose household head is not a high school graduate are more than twice as likely to live in poverty than families whose household head is a high school graduate (U.S. Bureau of the Census, October 1998i). In 1997, more than 26 percent of family heads 15 years and older who are not high school graduates live below the poverty level. In contrast, 11.4 percent of high school graduates and 2.4 percent with a bachelor's degree or more, live below the poverty level.

There are large poverty status and educational attainment differences between the Asian Pacific Islander and White populations. Of Asian Pacific Islander family heads without a high school degree, 30 percent reside in poverty, but only 16 percent of White family heads live in poverty. These marked racial group differences remain, even for high school and college graduates: household heads with a high school diploma, 14.6 (Asian Pacific Islander) and 7.5 percent (White); and with at least a bachelor's degree, 6.0 (Asian Pacific Islander) and 1.8 percent (White).

HOUSEHOLD INCOME OF THE NATIVE- AND FOREIGN-BORN

U.S. society also sees differences in median household income between the native born ($39,677) and foreign born ($32,962) (U.S. Bureau of the Census, September 1999m). More striking, however, is the income differential between the naturalized and unnaturalized foreign-born populations ($41,028 and $28,278 respectively).

Recent arrivals come from poorer countries "where the average education and wage and skill levels are far below those in the United States" (Smith & Edmonston, 1997, p. 7). The recently arrived immigrants and those from Latin America earn the lowest wages. The wage gap between the newly arrived and those who have been here longer "closes significantly for entrants from Western Europe and Asia, somewhat for others, but not at all for those from Mexico" (p. 8).

These large household income differences also have fiscal impacts on the nation and on individual states with large numbers of immigrant households like New Jersey, California, Texas, and Florida. Taxes pay for government services and, at the beginning, immigrants contribute less revenue than the native born and also have larger families.

First- and second-generation children "experience somewhat higher poverty rates, overall, than third- and later-generation children, but the differences [for most children] are concentrated in the first generation" (U.S. Administration for Children and Families, 1999, p. 426). Twelve countries of origin account for nearly half the children of immigrant children who live in poverty, and two-thirds of these children are of Mexican origin.(4)

The traditional assimilationist model of immigrant incorporation has assumed that subsequent generations of immigrants are gradually absorbed into the dominant society, as achievement norms are inculcated by the second generation. This has not occurred with Hispanic and Latino immigrant groups. Not only is the recent Hispanic and Latino immigrant population on the whole poorer and less literate than other immigrant groups but, even after two or three generations, their educational attainment remains lower than other immigrant groups. The generational status of children and adults is also related to various educational outcomes, and its effect is most visibly seen in the educational and occupational differences of parents and children and between immigrants and the native born (U.S. Department of Education, 1998a).

EDUCATIONAL INDICATORS OF WELL-BEING

The most significant predictors of economic well-being are related to education--literacy or a home environment where parents read to children; educational attainment of both parents and children; years of schooling; and English fluency. Comparative educational advantage is a function of family background. Future labor market and earnings success depends on years of schooling. Investing in our children also means that the community provides adequate educational resources to ensure successful outcomes. Racial and Hispanic origin differences persist.

Contribution of the Family to Literacy Activities, Schooling, and Educational Attainment

Parental involvement in the educational experiences of their children is known to have positive effects on student achievement and success in school (U.S. Department of Education, 1998i). Their involvement can mitigate the negative influences of the environment (U.S. Department of Education, 1999d).

Participation in literacy activities, achievement in elementary and secondary school, high school completion, and attendance at a four-year college are related to parental educational attainment, income, and race/ethnic background (Hauser et al., 1997; U.S. Department of Education, 1998c). Computer ownership, for example, is highly correlated with family income and educational attainment (Department of Commerce, 1995, 1997, 1999). Coming from a family whose parents did not complete high school--families at risk for low income and lower employment opportunities--greatly reduces the probability that parents engage in literacy activities with their children and that their children will complete high school, do well enough in high school to qualify for admission to college, or attend and complete college.

Early Literacy Experiences In and Outside the Home

Parental literacy is essential for children's success in school. Children who improve their literacy do better in school and increase their chances for graduating from high school. Data collected from the 1996 National Household Education Survey indicate that Hispanic and Black children ages 3 to 5 are less likely than White, non-Hispanic children to be read to by a family member "every day" (U.S. Federal Interagency Forum, 1998c). Even when the data are calculated for "three or more times in the past week" for all races, the percentage of children who are read to increases but the differences among the racial and ethnic groups remain: 65 percent Hispanic, 76 percent Black, and 89 percent White, respectively.

As part of the assessment of early childhood literacy activities, the National Center for Education Statistics also collects information on whether or not children "visited a library in the past month." Overall, only 39 percent of children of all races (includes the Asian population) and ethnic groups frequented libraries in 1996; however, there are differences among the Hispanic, Black, and White populations (U.S. Department of Education, 1999a). Only 26 percent of Hispanic children visited a library in contrast to 34 percent of Black children and nearly 43 percent of White children.

Black young children are much more likely (63 percent) than White non-Hispanic (54 percent), and significantly more than Hispanic (37 percent) children to be enrolled in center-based programs (U.S. Federal Interagency Forum, 1998c). However, Hispanic children under the age of 6 (not yet entered kindergarten) participate much more (54 percent) than White non-Hispanic (38 percent) and Black non-Hispanic (38 percent) children in child care and early education programs on a regular basis (U.S. Federal Interagency Forum, 1998d).

Reading to young children and visiting a library are highly correlated with parental or mother's education (U.S. Department of Education, 1999d; U.S. Federal Interagency Forum, 1998c). For example, in 1995, 59 percent of parents with less than a high school diploma and 90 percent with a bachelor's degree read to their children ages 3 to 5 three or more times a week. Only 19 percent of parents with less than a high school diploma visited a library during the previous month, and 30 percent of those with a high school diploma. The percentage of library visits rises for parents with a bachelor's degree (52 percent) and a graduate or professional degree (60 percent). Similar differences in mother's educational attainment are also observed in the number who read every day to children 6 to 12 years: 37 percent with less than a high school education; 49 percent have a high school or GED; 62 percent have a vocational/technical schooling or some college; and 77 percent are college graduates. Hispanic (all races) children who are "told a story at least once in the past week" increases to 79 percent for mothers with a college education (U. S. Department of Education, 1999c).

PARENTAL EDUCATIONAL ATTAINMENT, EMPLOYMENT, AND INCOME

Overall, there are significant improvements in two of the three family characteristics that strongly predict children's school completion rates (U.S. Department of Education, 1998e). The educational attainment of parents with children ages 6 to 12 years old increased substantially between 1972 and 1997 from 66 to 84 percent of mothers having completed at least high school (U.S. Department of Education, 1999c). Fathers also experienced similar gains in educational attainment.

Between 1972 and 1997, attainment of a high school education for White mothers with children age 15 to 18 years significantly declined from 32 to 8 percent. Significant increases (from 39 to 50 percent) have been observed for White mothers earning a high school diploma. White mothers with "some years of college" more than doubled (11 to 29 percent) and nearly tripled (8 to 22 percent) for those with a bachelor's degree or higher (U.S. Department of Education, 1998e).

The decline in Black mothers with less than a high school diploma has been even greater, approximately 64 to 22 percent. The percentage of Black mothers with a high school diploma has increased approximately 26 to 40 percent, and there has been more than a 22 percent increase for Black mothers with "some college." With the exception of a 22 percent decline in mothers with less than a high school diploma, Hispanic mothers have not experienced the same percentage gains in educational attainment between 1972 and 1997: from 16 to 25 percent for a high school diploma, about 6 to 14 percent for some college, and 2 to 6.5 percent for a bachelor's degree or higher. These increases have generally been matched by fathers, although the gains for Black and Hispanic fathers graduating from college are slightly greater than for Black and Hispanic mothers.

Between 1972 and 1997, the percentage of employed mothers for children between 6 and 18 years of age increased for all races by more than 20 percent (49 to 73 percent), but father's employment status for all races declined somewhat from 92 to 89 percent (U.S. Department of Education, 1998e). Employment rates for mothers of 6 to 12 year olds rose between 1972 and 1997 from 39 to 91 percent while fathers experienced a slight decline from 93 to 91 percent.

More mothers and fathers of young people ages 15 to 18 years are employed than ever before (U.S. Department of Education, 1999c). White mothers increased their employment participation from nearly 50 to 77 percent; Black mothers approximately 51 to 70 percent; and Hispanic mothers from slightly more than 31 to 56 percent. More than 90 percent of White fathers and around 85 percent of Black fathers were employed in 1972 and 1997. However, the percentage of employed Hispanic fathers declined from 90 to 85 percent.

Family income is also related to whether high school graduates are qualified for admission to a four year institution. Between 1992 and 1994, data collected from the National Educational Longitudinal Study of 1988 (NELS88) and Third Follow-up (1994) indicate that nearly half the marginally or unqualified students came from families whose income was less than $25,000. In contrast, 32 percent came from middle income ($25,000-74,999) and only 14 percent from high income ($75,000 or more) families (U.S. Department of Education, 1998g).

FLUENCY IN THE ENGLISH LANGUAGE

Fluency in the dominant language is viewed as critical for incorporation into the cultural, social, and economic life of the majority group. Three questions asked in the 1980 and 1990 censuses provide information about what languages are spoken in the home and how well English is spoken: Does this person speak a language other than English at home? What is this language? and, for those who spoke another language, How well does this person speak English--i.e., very well, well, not well, not at all?

The 1990 census data for language use indicate that nearly 14 percent of the total U.S. population 5 years and older speaks a language other than English at home (see Table 5). The 1990 census records more than twenty-two language groups (e.g., Uralic, Romance, Scandinavian, Germanic, Slavic, Indic) represented in the foreign-born population, with more people speaking Indo-European, Spanish, and Chinese languages than other languages (U.S. Bureau of the Census, March 1999c).

Table 5. Language Use and Percentage Distribution of English Language Ability by Age of Persons: 1990 Census

(Persons in households. Based on sample data)

                                                Speaks Non-English
                                                 Language at Home
                                    English
Age of Person          Total         Only         Total       %

5 years and older   230,445,777   198,600,798   31,844,979   13.8
5 to 17 years        45,342,448    39,019,514    6,322,934   13.9
18 years and over   185,103,329   159,581,284   25,522,045   13.8
Total               460,891,554   397,201,596   63,689,958

                    Speaks Non-English Language at Home

Age of Person       Ability to Speak English(1)

                     %             %       %
                    Very    %     Not     Not
                    well   Well   well   at all

5 years and older    56     23     15      6
5 to 17 years        62     23     12      2
18 years and over    55     23     16      7
Total

                    Speaks Non-English Language at Home

Age of Person               In Households

                                      All Speak
                    Linguistically   Non-English
                     Isolated(2)      Language

5 years and older      7,741,259      22,347,838
5 to 17 years          1,763,173       4,834,637
18 years and over      5,978,086      17,513,201
Total                 15,482,520      44,695,676

Note. Percentaged data subject to rounding. Table sources are U.S. Bureau of the Census, April 1999d, 1999e, 1999f.

(1) Percentages calculated on Total of "Speaks Non-English Language at Home."

(2) A linguistically isolated household is one in which no person 14 or older speaks English at least very well.

According to Current Population Surveys, the number of children who spoke another language at home other than English increased from about 40 to 66 million between 1979 and 1995 (U.S. Federal Interagency Forum, 1998b). Nearly 75 percent of Spanish-speaking and somewhat more than 45 percent of Asian children accounted for the vast majority of these children, with the West region of the United States experiencing the greatest growth.

How well do people assess their ability to speak English? According to the 1990 census, 56 percent of people 5 years and older say they speak English "very well" and another 23 percent "well." However, the number of children who had "difficulty speaking English" increased from 1.25 million in 1979 to 2.4 million in 1995--5 percent of the school age children in the United States (U.S. Federal Interagency Forum, 1998f).

Data tabulated from the 1995 Current Population Survey indicate that 31 percent of Hispanic (all races) children and 14 percent of Asian and other origin children between the ages of 5 and 17 years are identified as "having difficulty speaking English" compared with 1 percent of White non-Hispanic or Black non-Hispanic children (U.S. Federal Interagency Forum, 1998b, 1998f). Data tabulated directly from the 1990 census indicate that more than 40 percent of Spanish-speakers 65 years and older speak English "not well" or "not at all," in contrast to about 28 percent for the 8 to 64 years cohort and about 15 percent for persons 5 to 17 years of age (Robbin, 2000c).

The Bureau of the Census constructs a measure of "linguistic isolation" based on how well people assess their English language fluency. A linguistically isolated household is one in which no person 14 or older speaks English at least very well (U.S. Bureau of the Census, April 1999d). Nearly 25 percent of people 5 years and older, 28 percent between 5 and 17 years, and 23.4 percent 18 years and over are deemed "linguistically isolated."

These statistics mask significant differences in English fluency by national origin and racial groups (U.S. Bureau of the Census, April 1999h). More than 70 percent of people of German and French origin, followed by Italian (67 percent), Polish (63 percent), Filipino (60 percent), and Hispanic/Latino (52 percent) heritage speak English "very well." The Spanish-speaking and Asian populations rank highest in the total number of people who are not fluent English language speakers.

SCHOOL ACHIEVEMENT, HIGH SCHOOL COMPLETION RATES, AND EDUCATIONAL ATTAINMENT

How well our elementary and secondary students are learning has been assessed by the National Assessment of Educational Progress (NAEP) surveys conducted since the 1970s. Overall school performance in science and mathematics has improved, but no gains have been made in reading (U.S. Department of Education, 1999b). Nevertheless, between 1971 and 1988, average reading proficiency scores rose for Black students by about 20 or more on the scale, and the gap between Black and White students narrowed. The narrowing of the gap has been less pronounced between White and Hispanic students.

Since the late 1980s, the gap between Black and White students has remained about the same. White non-Hispanic students "consistently [had] higher reading and math scores than either Black, non-Hispanic, or Hispanic students ages 9, 13, and 17" (U.S. Federal Interagency Forum, 1998e). The gap between Hispanic students and White non-Hispanic students in the 1996 mathematics achievement scores was about the same across the three age groups--between 21 and 25 points lower for Hispanic children (U.S. Federal Interagency Forum, 1998g). Hispanic 17 year olds (all races) did somewhat better in mathematics achievement than Black non-Hispanic students. Reading achievement scores of Hispanic children for ages 9, 13, and 17 demonstrate a similar differential between White non-Hispanic and Hispanic children that is consistent across all cohorts with scale scores that average between 26 and 29 points.

HIGH SCHOOL COMPLETION

Completing high school is critical for success in the workplace and in becoming a fully functioning member of society. According to data collected in 1996 by the U.S. Department of Education, 5 percent of students who were in grades 10 through 12 had dropped out of school the previous year (U.S. Department of Education, 1998d). Black and Hispanic young people are disproportionately at risk of not graduating from high school because, proportionately, more Black and Hispanic young people come from families with lower parental education, lower income, and headed by single parents (U.S. Department of Education, 1999b).

Black and Hispanic young people are more likely to drop out of school than White teenagers. The drop out rate between 1975 and 1994 among persons aged 16 to 24 declined for White non-Hispanic children from 11.4 to 7.7 percent. The Black non-Hispanic rate declined from 22.9 to 12.6 percent. However, drop out rates for the Hispanic population show no decline over the two decades, remaining at about 30 percent (U.S. Department of Education, 1996b).

The Black population has, however, made very large gains between 1940 and 1996, closing the White/Black educational attainment gap at the high school level (U.S. Department of Education, 1998f; 1999b). High school completion rates have improved markedly for Black children since the early 1970s, with most of the improvement taking place during the 1980s. The proportion of Black high school graduates increased from about 8 percent in 1940 to more than 74 percent in 1996, compared to the White high school graduate population from 26 to more than 82 percent (U.S. Department of Education, 1996c). African-Americans increased their high school graduation rates to 86 percent in 1997 (U.S. Bureau of the Census, June 1998g).

This progress has not been reflected in Hispanic population school completion rates, which have remained at about 57 percent between 1980 (first year of data collection) and 1995. The gap between White and Hispanic rates has not narrowed (U.S. Department of Education, 1996c, 1999b).

Asian American students continue to demonstrate very high completion rates. According to data collected in 1997, 90 percent of the API population has graduated from high school (U.S. Bureau of the Census, October 1998j).

HIGHER EDUCATION ENROLLMENT AND COMPLETION

More than 85 percent of parents expect that their children will "at least acquire some education beyond high school [and] ... nearly 60 percent expect that their children will at least finish college" (Carbonaro, 1999, p. 653; italics in the original). Are high achievement norms and expectations of the home environment consistent with the reality of college completion and educational attainment?

The Condition of Education 1998 reports that "about half of the Black and Hispanic high school graduates were qualified for college admission, and Asian and Pacific Islanders, 73 percent, while 68 percent of White high school graduates were qualified" (U.S. Department of Education, 1998g; see also Chronicle of Higher Education, 1998a, 1998b). Nevertheless, students of all races and ethnic groups who are academically prepared enroll in college at the same rates (U.S. Department of Education, 1999g). Once African-American and Hispanic high school students graduate from high school, differences between the two population groups disappear.

Minority students now make up about 25 percent of all students enrolled in colleges and universities. This reflects an increase of 10 percent between 1976 and 1996 (U.S. Department of Education, 1998i). Most of this increase is due to enrollment growth in the Hispanic and Asian populations. American Indian students increased their enrollments by 81 percent; Asian students, 319 percent; Black students, 46 percent; Hispanic students, 204 percent; and White students, 13 percent (Chronicle of Higher Education, 1999a).

For the ten year period between 1987 and 1997, Black high school graduates increased their enrollment in college by nearly 10 percent, Hispanic students by 7.5 percent, and White students by 8.7 percent (Chronicle of Higher Education, 1999b). Black students now make up the majority of minority students enrolled in colleges and universities, accounting for 10 percent of the total enrollment, and Hispanic and Asian populations for 8 and 6 percent, respectively. At the same time that minority student enrollments have climbed, White student enrollments have climbed even faster, widening even further the gap in educational attainment between White and minority populations (U.S. Department of Education, 1999e).

About 68 percent of White students and 54 percent of Black and Hispanic students complete "some college" (U.S. Department of Education, 1999g; 1999i). Bachelor degree or higher completion rates for the 25 to 29 year olds in 1997 are similar for African-Americans (16 percent) and Hispanic (18 percent) students; and 35 percent of White students graduate with a bachelor's degree (U.S. Department of Education, 1998h). Of the Asian population 25 years and older, 42 percent has a bachelor's degree or higher (U.S. Bureau of the Census, October 1998j).

Another measure of persistence toward degree completion is reflected in the number of degrees conferred by U.S. colleges and universities. White students account for nearly 75 percent of all degrees awarded; Hispanic and API students, 5 percent; American Indian students, .6 percent; and African-American students, 7.5 percent in 1993-94 and 1995-96 (Chronicle of Higher Education, 1999c, 1999d).

Asian American and Hispanic students experienced the greatest growth in the number of degrees conferred between 1993-94 and 1995-96, about 14 percent. Asian Americans and Hispanic students experienced very large increases in Associate degrees (21.4 percent and 15.4 percent, respectively), bachelor's degrees (about 13 percent), and master's degrees (12.5 percent and 14.7 percent, respectively). Asian and Hispanic students received around 14 percent more professional degrees over the period. The number of doctorates received by Asian American students increased 23.1 percent.

Overall, the number of degrees awarded to African-American and American Indian groups increased 8.5 and 9.0 percent, respectively. American Indian students received 14 percent more doctoral and 22 percent more professional degrees over the period. Black students saw increases of about 12 percent for master's and doctoral degrees. White students saw an overall decline of 4.2 percent and a 9.4 percent decrease in the number of bachelor's degrees awarded.

Although the number of professional degrees awarded to minorities has increased, very few translate into library school degrees (Association for Library and Information Science Education, 1998). Minority populations constitute a very small percentage of public, school, and academic librarians, well below their share of the total population; the White population accounts for nearly 87 percent of academic and public librarians according to the 1998 ALA Survey of Librarian Salaries (American Library Association, 1998a).

OUR NATIONAL INVESTMENT IN EDUCATION AND LIBRARIES

Education is the principal vehicle for equalizing opportunity. Our commitment is measured by various indexes that reflect how much our nation spends to educate our children and to provide access to educational resources through the public school system and public libraries.

The United States spent 3.5 percent of its gross national product (GNP) on primary and secondary education and 1.1 percent on higher education in 1995, a smaller percentage than what the nation spent during the early- to mid-1970s (U.S. Department of Education, 1999e). The Digest of Education Statistics 1998 reports Fiscal Year 1998 estimates for funding for elementary and secondary education at $37 billion; postsecondary education at $16 billion; research at universities and related institutions at $17 billion; and other programs at $5.2 billion (U.S. Department of Education, 1991). Libraries, museums, and federal institutions received 14 percent of the total federal support for education in FY 1998 (U.S. Department of Education, 1999e).(6)

K-12 AND HIGHER EDUCATION

Per pupil expenditures for elementary and secondary education increased slightly between the 1989-90 and 1995-96 school years from $6,700 to nearly $6,900, but the amount spent on each student varies considerably by the wealth of the school district (U.S. Department of Education, 1999h). Wealthier school districts, whose median household income was $35,000 or more, spent about $7,500 for each pupil. In contrast, school districts whose median household income was less than $20,000 expended $6,000 per pupil.

Public school spending also varies by the size of the minority population in the district (U.S. Department of Education, 1998j).(7) In the 1993-94 school year, districts with low minority enrollments spent on average $500 less per pupil than districts with high minority enrollment (Table 6). As a percentage of total per pupil expenditures, those districts with 50 percent or more minority populations spend considerably less on capital outlays than school districts where the minority enrollment is between 5 and 49 percent; about the same for "other services" than districts of less than 5 percent but considerably more than districts between 5 and 49 percent minority enrollment.

Table 6. Public School Expenditures per Pupil by Function and Percentage of Minority School-age Children: School Year 1993-94

(in 1997 constant dollars)

                                              Support
Selected District             Instruction     Services
Characteristics       Total   Total   %      Total   %

Median household
income

Less than $20,000     5,634   3,052   54.0   1,745   31.0
20,000 - 24,999       5,899   3,190   54.7   1,772   27.9
25,000 - 29,999       6,361   3,480   54.7   1,849   29.0
30,000 - 34,999       6,124   3,292   53.8   1,851   30.2
35,000 or more        7,027   3,806   54.2   2,174   30.9

Percentage of
minority school-age
children

Less than 5           6,347   3,492   55.0   1,838   28.9
5 - 19                6,362   3,426   53.9   1,923   30.2
20 - 49               6,018   3,204   53.2   1,878   31.2
50 or more            6,847   3,779   55.2   2,054   30.9

Selected District     Capital Outlay     Other
Characteristics       Total    %       Total    %

Median household
income

Less than $20,000       407   7.2       407    7.6
20,000 - 24,999         470   7.4       470    7.3
25,000 - 29,999         551   8.7       551    7.6
30,000 - 34,999       1,851   8.9       545    7.2
35,000 or more        2,174   8.6       606    6.3

Percentage of
minority school-age
children

Less than 5             509   8.0       508    8.0
5 - 19                  620   9.7       393    6.2
20 - 49                 536   8.9       400    6.6
50 or more              448   6.5       565    8.3

Note. Percentages by function calculated on the basis of total dollars expended. Percentages subject to rounding. Table source is U.S. Department of Education, 1998j.

The difference in capital outlay expenditures between low and high minority population school districts suggests lower spending for information technology by districts with high minority populations. This is confirmed by the U.S. Department of Education and market research firms. Data collected by the U.S. Department of Education (1995a, 1996a, 1998d) since the early 1990s show that Internet access is correlated with income and minority population enrollment, and the lower the income and higher the minority population, the lower the access to information technology.

The November 1999 survey conducted by Market Data Retrieval, a unit of Dun & Bradstreet Corporation that specializes in the education market, found that "schools located in areas with high poverty rates and a high percentage of minority students are less likely to have up-to-date technology than others" (Mendels, 1999, p. 2). More than 90 percent of schools in the wealthiest communities, but 84 percent in the poorest areas, have Internet access.

Nevertheless, considerable progress has been made between 1995 and 1998. In 1995, 31 percent of high poverty school districts had access to the Internet in contrast to 82 percent in districts with very low poverty (U.S. Department of Education, 1999k).(8) By 1998, most of these differences disappeared, and connections to the Internet increased to 89 percent (U.S. Department of Education, 1998k; 1999j). Still, "schools with high poverty were slightly less likely to have Internet access than schools in lower poverty districts" (U.S. Department of Education, 1999j).

Other indicators, however, show the distance that we must go to achieve equal educational opportunity through our school system. Access to the Internet is unequally distributed inside schools where education takes place. High minority enrollment districts have a lower percentage of instructional room access, a lower ratio of students to instructional computers, and fewer information technology resources available in a school media center or library (U.S. Department of Education, 1997). The National Center for Education Statistics (U.S. Department of Education, 1998k) reports that, in 1998, public schools with 50 percent or more minority enrollment had Internet access in 37 percent of instructional rooms in contrast to schools with lower minority enrollment (between 52 and 59 percent with Internet access).(9)

There are also racial differences in information technology use in higher education. Hoffman and Novak (1999, p. 3) report that more than 90 percent of private college freshmen used the Internet for research, but only 78 percent of students entering public Black colleges report doing so. About 80 percent of private college freshmen report using e-mail regularly, while 42 percent of Black public college students do. Novak and Hoffman also found that "access translates into usage" and that "White students were more likely than Black students to use the Web and were more likely to have used the Web at locations other than home, work or school" (p. 2).

PUBLIC LIBRARIES

Can public libraries mitigate the effects of existing inequity and inequality? A 1998 survey of connectivity in public libraries found that about 20 percent of all libraries are located in rural and urban poverty areas, but that public access to the Internet "is about equal in poverty and non-poverty areas" (about 73 percent) (American Library Association, 1998b, p. 3; Bertot & McClure, 1998). More libraries located in "extreme poverty" areas do, however, offer Internet access (79 percent).(10)

But are these libraries accessible to people in poverty? Work by Jue (1999) and his colleagues is instructive in this regard, although they caution that their sample is based on the 1990 decennial census and their conclusions are subject to a variety of caveats. They find that public library outlets are "disproportionately located in low poverty/middle class census tracts and may be inaccessible to potential library users in poverty" (p. 313). More people in poverty are not being served in the western than in the eastern part of the United States (13 percent versus 6.7 percent) (p. 316).

The study by Jue and his colleagues examines user activities within public libraries that serve lower income and "majority-minority" markets (Koontz et al., 1999a). Reading/writing and browsing constitute the major activities for minorities with 19 percent of African-American, 27 percent of Asian American, and 23 percent of Hispanic users engaged in reading/writing, and 27 percent of Native American users engaged in browsing (Koontz et al., 1999b).(11) There are also differences in question-asking: more questions about computer usage by African-American users; education and homework by Asian American and Hispanic users; and leisure and entertainment by Native American users (Koontz et al., 1999c). But there are virtually no differences in the types of questions asked by minority adults, the elderly, and children of all ages (Koontz et al., 1999c).

CONCLUSION

We arrive at a new century having made enormous progress in improving the life chances of many of our citizens. Our immigrant nation is the most dynamic in the world. Yet, the "fault lines" of society remain, as they have historically, between the "haves and have nots" and among the diverse races and ethnic groups that make up the American "melting pot."

Some people argue that the "Digital Divide" can be overcome with a change in priorities and by substantially increasing our investment in the national information infrastructure. This conception of problem solving, however, assumes that the disparities among the races and ethnic groups can be eliminated by substantial infusions in government spending in technology. Yes, there is a "Digital Divide" in our society, but the metaphor is overused and, some would argue, misused (see Powell, 1999). Yes, federal priorities may make a difference; however, a cautionary note must be issued: We cannot be sure that "a given outcome is attributable to a social program or to a multiplicity of other possible causes ... [because the] causal pathways between policies and desirable outcomes are seldom direct" (Fellegi & Wolfson, 1999, pp. 375, 376).

The "Digital Divide" is an outcome of long-standing inequalities in society and choices that people make. Our attention is thus more usefully directed toward understanding the antecedents, correlates, and predictors of this inequality, and toward determining the statistical data we need to help us develop wise and effective library policies that provide equal access to information for everyone and eliminate barriers to library and information services. We need to focus on a new millennium that brings a beautiful but very complex mosaic of ethnicities and cultures, and a large increase in the number of lowd-come, non-English speaking, and elderly people whose needs will require significant planning to provide appropriate services. The well-being of families is central and, in particular, the educational attainment of parents.

The library community will make its most important contribution to enhancing our social capital with its current commitment to programs that raise literacy levels, improve our children's interest in reading, and coordinate more closely the activities of school and public libraries. The research conducted on library markets needs to be taken very seriously: libraries need to be close to their users to be used. Finally, there is incontrovertible evidence that high minority and poverty school districts are comparatively disadvantaged in terms of instructional resources and capital funds. These are items on an agenda for action, where political activism of the library profession will make a difference. It is here where commitment is translated into national priorities.

NOTES

(1) The U.S. Bureau of the Census (September 1999m, p. A-1) defines a "family" as a group of two or more people related by birth, marriage, or adoption who reside together. A "household" consists of all people who occupy a housing unit. It includes the related family members and all unrelated people, if any, such as lodgers, foster children, wards, or employees who share the housing unit. Group quarters are excluded from the count of households. A household may contain more than one family. The distinction between "family" and "household" is important for calculations of all wealth and income statistics. These definitions are foundational for interpreting statistics.

(2) "Neighborhoods have effects, too, but these are not discussed in this article. For relevant discussions, see Sampson et al. (1999) and Brooks-Gunn et al. (1997).

(3) In 1997, the poverty threshold for a family of four was $16,400.

(4) These countries are the former Soviet Union, Cambodia, Laos, Thailand, Vietnam, El Salvador Guatemala, Nicaragua, Mexico, Honduras, Haiti, and the Dominican Republic.

(5) Earlier censuses asked questions about "mother tongue" (the language spoken when the person was a child) or only for a select group (e.g., the foreign-born) (U.S. Bureau of the Census, April 1999d).

(6) This index measures the amount of revenue raised for educating our students relative to the income of taxpayers adjusted for the number of students and the total population. The U.S. Department of Education (1999f) explains that "the numerator measures average financial resources available for the education of each student. The denominator measures the taxpayer's average ability to pay. The index is the number of dollars of revenue raised for each student from each $100 of income received by each member of the population."

(7) Another indicator of social investment in education is the "national index of public effort" (U.S. Department of Education, 1999f). Our national investment in higher education has fallen precipitously since 1970 (9.9 in 1996, 31.1 in 1970). The index for national investment in elementary and secondary education increased from 19.5 in 1970 to 23.5 in 1996. The differential in spending is reduced when adjustments are made for cost of living (U.S. Department of Education, 1995b).

(8) The "poverty" indicator used by the U.S. Department of Education is the number of students eligible for a free or reduced price lunch (U.S. Department of Education, 1999k).

(9) Yet, even were these existing inequalities alleviated by an infusion of funding to build an information infrastructure, significant barriers would remain. Extensive work by Schofield and colleagues offers a cautionary note regarding implementation of computer technology in the classroom (Eurich-Fulcer & Schofield, 1995; Schofield, 1994, 1995, 1997; Schofield & Davidson, 1998; Schofield et al., 1997). Computer technology use depends on social and political processes inside schools and the relationship between schools and the outside world, and these, Schofield and colleagues find, are significant barriers to technology implementation.

(10) "Extreme poverty" is defined as a condition where more than 40 percent of the population in a particular geographic area (e.g., census tract) lives below the poverty level. A particular geographic area is "poor" if 20 percent or more of its inhabitants live at the poverty level. A "low poverty area" is one where 20 percent or less of the population live at the poverty level. See Jue et al. (1999) for more information about the effects of defining poverty and low income tracts on siting public libraries.

(11) The user activities are: reading/writing, browsing, using computer, checking out/library card, library program/tours, sitting alone/socializing, schoolwork, and non-library program.

REFERENCES

American Anthropological Association. (1997). Response to OMB Directive 15. Retrieved November 15, 1997 from the World Wide Web: http://www.ameranthassn.org/ comsumm.htm.

American Library Association. (1998a). Distribution of librarians by racial/ethnic/sexual group in academic and public libraries (unpublished data). Chicago: ALA.

American Library Association. (1998b). New report shows more libraries connect to the Internet, access still limited. Washington Office Newsline. Retrieved December 1, 1999 from the World Wide Web: http://ala.org/washoff/alawon/alwn7149.html.

Anderson, M. J. (1988). The American census: A social history. New Haven: Yale University Press.

Association for Library and Information Science Education. (1998). Library and Information Science Education Statistical Report 1998. Washington, DC: ALISE.

Bauman, K. J. (1999). Extended measures of well-being: Meeting basic needs (Current Population Reports P70-67). Washington, DC: U.S. Bureau of the Census.

Bean, F.; Chapa, J.; Berg, R. R.; & Sowards, K. A. (1994). Educational and sociodemographic incorporation among Hispanic immigrants to the United States. In B. Edmonston & J. S. Passell (Eds.), Immigration and ethnicity: The integration of America's newest arrivals (pp. 73-100). Washington, DC: Urban Institute Press.

Bertot, J. C., & McClure, C. R. (1998). Moving toward more effective public Internet access: The 1998 national survey of U.S. public library outlet Internet connectivity. Summary results. Retrieved December 1, 1999 from the World Wide Web: http://www.ala.org/olitp/ 1998plo.pdf; http://www.ala.org/olitp/survey98.pdf.

Borjas, G. J, (1999). Heaven's door: Immigration policy and the American economy. Princeton, NJ: Princeton University Press.

Brooks-Gunn, J.; Duncan, G. J.; & Abet, L. J. (Eds.). (1997). Neighborhood poverty: Context and consequences for children. New York: Russell Sage Foundation.

Carbonaro, W. J. (1999). Opening the debate on closure and schooling outcomes. American Sociological Review, 64(5), 682-686.

Chan, S. (1991). Asian Americans: An interpretive history. Boston: Twayne Publishers.

Chronicle of Higher Education. (1998a). Average scores on the scholastic assessment test by sex and racial and ethnic group, 1996. In The chronicle almanac 97-98: Facts & Figures. Retrieved November 30, 1999 from the World Wide Web: http://chronicle.com/ facts.

Chronicle of Higher Education. (1998b). Average scores on the ACT Assessment by sex and racial and ethnic group, 1997. In The chronicle almanac 97-98: Facts & Figures. Retrieved November 30, 1999 from the World Wide Web: http://chronicle.com/facts.

Chronicle of Higher Education. (1999a). College enrollment by racial and ethnic group, selected years. In The chronicle: 1999-2000 almanac. Retrieved November 30, 1999 from the World Wide Web: http://chronicle.com.

Chronicle of Higher Education. (1999b). Proportion of 18-24-year-olds enrolled in college by high-school-graduation status and race, 1987 to 1997. In The chronicle: 1999-2000 almanac. Retrieved November 30, 1999 from the World Wide Web: http:// chronicle.com.

Chronicle of Higher Education. (1999c). Degrees conferred by racial and ethnic group, 1993-94. In The chronicle: Facts and figures. Retrieved November 30, 1999 from the World Wide Web: http://chronicle.com.

Chronicle of Higher Education. (1999d). Degrees conferred by racial and ethnic group, 1995-96. In The chronicle: 1999-2000 almanac. Retrieved November 30, 1999 from the World Wide Web: http://chronicle.com.

Edmonston, B.; Goldstein, J.; & Lott, J. T. (Eds.). (1996). Spotlight on heterogeneity: The federal standards for racial and ethnic classification summary of a workshop. Washington, DC: National Academy of Sciences.

Espiritu, Y. L. (1992). Asian American panethnicity: Bridging institutions and identities. Philadelphia: Temple University Press.

Espenshade, T. J.; Fix, M.; Zimmerman, W.; & Corbett, T. (1996-1997). Immigration and social policy: New interest in an old issue. Focus, 18(2), 1-10.

Eurich-Fulcer, R., & Schofield, J. W. (1995). Wide-area networking in K-12 education: Issues shaping implementation and use. Computers & Education, 24(3), 211-230.

Feeney, P. (1994). The 200 Years' War and counting: Power politics and census controversies: 1790-1990. Unpublished doctoral dissertation, Mount St. Mary's College, Emmitsburg, Maryland.

Felligi, I., & Wolfson, M. (1999). Towards systems of social statistics--Some principles and their application in Statistics Canada. Journal of Official Statistics, 15(3), 373-393.

Fugita, S. S., & O'Brien, D. J. (1991). Japanese American ethnicity: The persistence of community. Seattle: University of Washington Press.

Gamoran, A.; Mare, R. D.; & Bethke, L. (1999). Effects of nonmaternal child care on inequality in cognitive skills (Institute for Research on Poverty Discussion Paper no. 1186-99). Madison: University of Wisconsin.

Gibson, C. J., & Lennon, E. (1999). Historical census statistics on the foreign-born population of the United States: 1850-1990 (Population Division Working Paper No. 29). Washington, DC: U.S. Bureau of the Census.

Hauser, R. M.; Brown, B. V.; & Prosser, W. R. (Eds.). (1997). Indicators of children's well-being. New York: Russell Sage Foundation.

Haveman, R., & Wolfe, B. (1994). Succeeding generations: On the effects of investments in children. New York: Russell Sage Foundation.

Hayes-Bautista, D. E., & Chapa, J. (1987). Latino terminology: Conceptual bases for standardized terminology. American Journal of Public Health, 7(1), 61-68.

Hernandez, E. W. (1996). Comparisons of selected social and economic characteristics between Asians, Hawaiians, Pacific Islanders, and American Indians (including Alaskan Natives) (Population Division Working Paper No. 15). Washington, DC: U.S. Bureau of the Census.

Hershey, R. D. (1999). Economy proving it's more robust than anticipated. The New York Times. Retrieved on November 25, 1999 from the World Wide Web: http:// www.nytimes.com/yr/mo/day/news/financial/econ-report.html.

Hoffman, D. L., & Novak, T. P. (1999). The evolution of the digital divide: Examining the relationship of race to Internet access and usage over time. Retrieved December 1, 1999 from the World Wide Web: http://ecommerce.vanderbilt.edu/papers/.

Ignatiev, N. (1995). How the Irish became white. New York: Routledge.

Jue, D. K.; Koontz, C. M.; Magpantay, J. A.; Lance, K. C.; & Seidl, A. M. (1999). Using public libraries to provide technology access for individuals in poverty: A nationwide analysis of library market areas using a geographic information system. Library & Information Science Research, 21(3), 299-325.

Koontz, C.; Jue, D. K.; & Lance, K. C. (1999a). Market-based adult lifelong learning performance measures for public libraries servicing lower income and majority-minority markets, September 1, 1996-August 31, 1999 (Final performance report to the U.S. Department of Education). Tallahassee: Florida State University, GeoLib Program.

Koontz, C.; Jue, D. K.; & Lance, K. C. (1999b). Comparison of user activities within libraries: Minority markets. Retrieved December 1, 1999 from the World Wide Web: http:// www.geolib.edu/Activities_minorities.html.

Koontz, C.; Jue, D. K.; & Lance, K. C. (1999c). Comparison of user questions within libraries: Minority markets. Retrieved December 1, 1999 from the World Wide Web: http:// www.geolib.edu/Questions_minorities.html.

Koontz, C.; Jue, D. K.; & Lance, K. C. (1999d). Top two user activities within libraries per each age group: Minority markets. Retrieved December 1, 1999 from the World Wide Web: http://www.geolib.edu/Top2Activities_minorities.html.

Mendels, P. (1999). Internet access spreads to more classrooms, survey finds. The New York Times. Retrieved December 1, 1999 from the World Wide Web: http:// www.nytimes.com/library/tech/99/12/cyber/education/01education.html.

Mueller, M. L., & Schement, J. R. (1996). Universal service from the bottom up: A study of telephone penetration in Camden, New Jersey. Information Society, 12, 273-292.

Nasar, S. (1999). A gloomy view of immigration. The New York Times, October 10, p. BU6.

Oboler, S. (1992). The politics of labeling: Latino/a cultural identities of self and others. Latin American Perspectives, 19(4), 18-36.

Passel, J. S., & Edmonston, B. (1994). Immigration and race: Recent trends in immigration in the United States. In B. Edmonston & J. S. Passel (Eds.), Immigration and ethnicity: The integration of America 5 newest arrivals (pp. 31-72). Washington, DC: Urban Institute Press.

Powell, A. C., III (1999). Falling for the gap: Whatever happened to the "digital divide"? Reason, (November), 43-44.

Review of Federal Measurement of Race and Ethnicity: Hearings before the Subcommittee on Census, Statistics and Postal Personnel of the Committee on Post Office and Civil Service of the House of Representatives, 103rd Cong., 1st Sess. (1993). (testimony of Juanita Lott).

Robbin, A. (1999). The problematic status of statistics on race and ethnicity: An "imperfect representation of reality." Journal of Government Information, 26(5), 467-483.

Robbin, A. (2000a). Classifying racial and ethnic group data in the United States: The politics of negotiation and accommodation. Journal of Government Information, 27(2), 129-156.

Robbin, A. (2000b). The politics of representation in the U.S. national statistical system: Origins of minority population interest group participation. Journal of Government Information, 27(4), 1-22.

Robbin, A. (2000c). Hispanic demographics and implications for library services for Hispanic youth. In B. Immroth & K. d. l. P. McCook (Eds.), Library services to youth of Hispanic heritage (pp. 137-154). Jefferson, NC: McFarland & Company, Inc..

Sachs, S. (1999). As New York immigration thrives, diversity widens. The New York Times, November 8, p. A24 (National edition).

Sampson, R. J.; Morenoff, J. D.; & Earls, E (1999). Beyond social capital: Spatial dynamics of collective efficacy for children. American Sociological Review, 64(5), 633-660.

Schofield, J. W. (1994). Barriers to computer usage in secondary school. In C. Hull & T. Finholt (Eds.), Social issues in computing: Putting computing in its place (pp. 547-578). New York: McGraw-Hill.

Schofield, J. W. (1995). Computers and classroom culture. New York: Cambridge University Press.

Schofield, J. W. (1997). Computers and classroom social process--a review of the literature. Social Science Computer Review, 15(1), 27-39.

Schofield,J. W., & Davidson, A. L. (1998). The Internet and equality of educational opportunity. Unpublished paper presented at the World Conference on ED-Media/ED-TELECOM, June 1998, Freiburg, Germany.

Schofield, J. W.; Davidson, A.; Stocks,J. E.; & Futoran, G. (1997). The Internet in school: A case study of educator demand and its precursors. In S. Kiesler (Ed.), Culture of the Internet (pp. 361-381). Mahwah, NJ: Lawrence Erlbaum Associates.

Schofield, J. W.; Stocks, J.; & Davidson, A. L. (1998). Connecting schools to the world via the Internet: Do the walls come tumbling down or just get lower? In Proceedings of the Thirtieth Annual Hawaii International Conference on System Sciences (pp. 665-672). Los Alamitos, CA: IEEE Computer Society Press.

Sengupta, S. (1999). Snapshot of world faiths on one block in Queens. The New York Times, November 7, p. 31 (National ed.).

Smith, J. P., & Edmonston, B. (Eds.). (1997). The new Americans: Economic, demographic, and fiscal effects of immigration. Washington, DC: National Academy Press.

Starr, P. (1987). Introduction. In W. Alonso & P. Starr (Eds.), The politics of numbers (pp. 7-57). New York: The Twentieth Century Fund.

Swerdlow, J. (1999). Global culture. National Geographic, 196(2), 22-33.

Thernstrom, A. (1987). Whose vote counts? Affirmative action and minority voting rights. New York: The Twentieth Century Fund.

Toy, V. S. (1998). Bustling Queens library speaks in many tongues. The New York Times, May 31, pp. 30, 33.

Uchitelle, L. (1999a). Big increases in productivity by workers. The New York Times, November 13, pp. B1, B14 (National ed.).

Uchitelle, L. (1999b). Stocks surge on positive labor data. The New York Times, December 4. Retrieved December 4, 1999 from the World Wide Web: http://www.nytimes.com/ yr/mo/day/news/financial/econ-report.html.

U.S. Administration on Aging. (1999). Aging into the 21st century. Washington, DC: U.S. Department of Health and Human Resources. Retrieved November 5, 1999 from the World Wide Web: http://www.aoa.dhhs.gov/aoa/stats/aging21/.

U.S. Administration for Children and Families. (1999). Trends in the well-being of America's children and youth 1998. Washington, DC: U.S. Department of Health and Human Resources. Retrieved November 4, 1999 from the World Wide Web: http:// aspe.os.dhhs.gov/hsp/98trends/trend98.htm.

U.S. Bureau of the Census. (1989). 200 Years of U.S. Census taking: Population and housing questions, 1790-1990. Washington, DC: U. S. Bureau of the Census.

U.S. Bureau of the Census. (1993). We the Americans: Asians (We the People series. WE-10). Washington, DC: U. S. Bureau of the Census. Retrieved November 20, 1999 from the World Wide Web: http://www.ceasus.gov/apsd/www/wepeople/asian/we-3.pdf.

U.S. Bureau of the Census. (February, 1996a). Population projections of the United States by age, sex, race, and Hispanic origin: 1995 to 2050 (Current Population Reports P25-1130, February). Washington, DC: U. S. Bureau of the Census.

U.S. Bureau of the Census. (1996b). Resident population of the United States: Middle series projections. Retrieved November 5, 1999 from the World Wide Web: http:// www.census.gov/population/projections/nation/nsrh/nprh9600.txt; nprh0105.txt; nprh0610.txt; nprh1530.txt; nprh3550.txt.

U.S. Bureau of the Census. (March, 1998a). Table 4. Poverty status of families, by type of family, presence of related children, race, and hispanic origin: 1959 to 1997. Historical income tables. Retrieved June 5,1999 from the World Wide Web: http://www.census.gov/ income/histpov4.txt.

U.S. Bureau of the Census. (September, 1998b). 1997 population profile of the United States (Current Population Reports Special Studies P23-194: September). Washington, DC: U.S. Bureau of the Census.

U.S. Bureau of the Census. (March,1998c). Table H-5. Race and Hispanic origin of householder--households by median and mean income: 1967 to 1997. Historical Income Tables. Retrieved June 15, 1999 from the World Wide Web: http://www.census.gov/h05.html.

U.S. Bureau of the Census. (June 10, 1998d). Table 1. Race of wife by race of husband: 1960, 1970, 1980, 1991, and 1992. Interracial tables. Retrieved November 4, 1999 from the World Wide Web: http://www.census.gov/intertab1.txt.

U.S. Bureau of the Census. (June 10, 1998e). Table 2. Race of couples: 1990: June 10. Interracial tables. Retrieved November 4, 1999 from the World Wide Web: http:// www.census.gov/intertab2.txt.

U.S. Bureau of the Census. (June 10, 1998f). Table 4. Race of child by race of householder and of spouse or partner: 1990 (June 10). Interracial tables. Retrieved November 4, 1999 from the World Wide Web: http://www.census.gov/intertab4.txt.

U.S. Bureau of the Census. (June 29, 1998g). Young African Americans boost high-school completion rate, Census Bureau reports. News. Retrieved November 24, 1999 from the World Wide Web: http://www.census.gov/Press-Release/cb98-106.html.

U.S. Bureau of the Census. (July 27, 1998h). Interracial married couples: 1960 to present (Table MS-3). Retrieved November 6, 1999 from the World Wide Web: http:// www.census.gov/population/socdemo/ms-la/tabms-3.txt.

U.S. Bureau of the Census. (October 1, 1998i). Table 5. Selected characteristics of families, by type, region, and race of householder: March 1997 (October 1). Retrieved November 23, 1999 from the World Wide Web: http://www.census.gov/population/ socdemo/race/api97/table05.txt.

U.S. Bureau of the Census. (October 15, 1998j). Table 1. Selected social characteristics of the population, by region and race: May 1997 (October 1). Retrieved November 23, 1999 from the World Wide Web: http://www.census.gov/population/socdemo/race/ api97/table01.txt.

U.S. Bureau of the Census. (January 7, 1999a). Table MS-3. Interracial married couples: 1960 to present. Interracial tables. Retrieved November 4, 1999 from the World Wide Web: http://www.census.gov/couples-tabms3.txt.

U.S. Bureau of the Census. (February 2, 1999b). Table 1. Selected social characteristics of the population, by sex, region, and race: March 1998. Retrieved October 29, 1999 from the World Wide Web: http://www.census.gov/tab01.txt.

U.S. Bureau of the Census. (March 9, 1999c). Table 8. Language spoken at home for the foreign-born population 5 years and over: 1980 and 1990 (March 9). Retrieved November 25, 1999 from the World Wide Web: http://www.census.gov/population/www/ documentan/twps0029/tab05.html.

U.S. Bureau of the Census. (April, 1999d). Language use. Retrieved November 25, 1999 from the World Wide Web: http://www.census.gov/population/www/socdemo/ lang_use.html.

U.S. Bureau of the Census. (April, 1999e). Table 1. Language use and English ability: Persons 5 years and over, by state: 1990 census. Language use. Retrieved November 25, 1999 from the World Wide Web: http://www.census.gov/population/www/socdemo/ language/table1.txt.

U.S. Bureau of the Census. (April, 1999f). Table 2. Language use and English ability: Persons 5 to 17 years, by state: 1990 census. Language use. Retrieved November 25, 1999 from the World Wide Web: http://www.census.gov/population/www/socdemo/language/table2.txt.

U.S. Bureau of the Census. (April, 1999g). Table 3. Language use and English ability: Persons 5 to 17 years, by state: 1990 census. Language use (April). Retrieved November 25, 1999 from the World Wide Web: http://www.census.gov/population/www/socdemo/ language/table3.txt.

U.S. Bureau of the Census. (April, 1999h). Table 5. Detailed language spoken at home and ability to speak English for persons 5 years and over--50 languages with greatest number of speakers: United States 1990. Language use. Retrieved November 25, 1999 from the World Wide Web: http://www.census.gov/population/www/socdemo/language/ table5.txt

U.S. Bureau of the Census. (July 29, 1999i). Table 1 Selected social characteristics of the population by region and race: March 1998. Retrieved from the World Wide Web on November 11, 1999: http://www.census.gov/population/socdemo/race/api98/ table01.txt.

U.S. Bureau of the Census. (July 29, 1999j). Table 4. Selected characteristics of households, by type, region, and race of householder: March 1998. Retrieved from the World Wide Web on November 11, 1999: http:www.census.gov//population/socdemo/race/ api98/table04.txt.

U.S. Bureau of the Census. (July 29, 1999k). Table 9. Total money income in 1997 of families, by type, region, and race. Retrieved from the World Wide Web on November 11, 1999: http://www.census.gov/population/socdemo/race/api98/table09.txt.

U.S. Bureau of the Census. (July 29, 1999l). Table 5. Selected characteristics of families, by type, region, and race of householder: March 1998. Retrieved from the World Wide Web on November 21, 1999: http://www.census.gov/socdemo/race/api98/table05.txt.

U.S. Bureau of the Census. (1999m). Money income in the United States (Current Population Reports Consumer Income P60-206 (September). Washington, DC: U.S. Bureau of the Census. Retrieved November 10, 1999 from the World Wide Web: http://hhes/ income/income98/inc8sum.html (Table A. Comparison of summary measures of income by selected characteristics).

U.S. Department of Commerce. National Telecommunications and Information Administration. (1995). Falling through the net: A survey of the "have nots" in rural and urban America. Retrieved October 29, 1999 from the World Wide Web: http://www.ntia.gov/.

U.S. Department of Commerce. National Telecommunications and Information Administration (1997). Falling through the net II: New data on the digital divide. Retrieved October 29, 1999 from the World Wide Web: http://www.ntia.gov/.

U.S. Department of Commerce. Nation al Telecommunications and Information Administration. (1999, July). Falling thru the net: Defining the digital divide. A report on the telecommunications and information gap in America. Retrieved October 29, 1999 from the World Wide Web: http://www.ntia.gov/.

U.S. Department of Education. National Center for Education Statistics. (1995a). Advanced telecommunications in U.S. public schools, K-12 (NCES 95-731). Retrieved December 1, 1999 from the World Wide Web: http://nces.ed.gov/pubs95/95731.pdf.

U.S. Department of Education. National Center for Education Statistics. (1995b). Public school district funding differences (Indicator 53). The condition of education 1995. Retrieved December 3, 1999 from the World Wide Web: http://nces.ed.gov/pubs95/ condition95/c9553a01.html.

U.S. Department of Education. National Center for Education Statistics. (1996a). Youth indicators, 1996 (NCES 96-027). Washington, DC: U. S. Department of Education.

U.S. Department of Education. National Center for Education Statistics. (1996b). Indicator 29. Dropouts. Percent of high school dropouts among persons 16 to 24 years old, by sex and race/ethnicity: October 1967 to October 1994. Youth indicators, 1996 (NCES 96-027). Retrieved November 24, 1999 from the World Wide Web: http://nces.ed.gov/ pubs/yr/y9629a.html.

U.S. Department of Education. National Center for Education Statistics. (1996c). Indicator 28. School completion. Percent of 25- to 29-year olds completing high school and college, by age and race/ethnicity: 1940 to 1995. Youth indicators, 1996 (NCES 96-027). Retrieved November 24, 1999 from the World Wide Web: http://nces.ed.gov/ pubs/yr/y9628a.html.

U.S. Department of Education. National Center for Education Statistics. (1996d). Advanced telecommunications in U.S. public elementary and secondary schools, 1995 (NCES 96-854, February). Retrieved December 1, 1999 from the World Wide Web: http:// nces.ed.gov/pubs/96854.pdf.

U.S. Department of Education. National Center for Education Statistics. (1997). School library media resources (Indicator 6). The condition of education 1997. Retrieved December 3, 1999 from the World Wide Web: http://nces.ed.gov/pubs97/condition97/ c9706a01.html.

U.S. Department of Education. National Center for Education Statistics. (1998a). Generational status and educational outcomes among Asian and Hispanic 1988 eighth graders (NCES 1999-020). Washington, DC: U.S. Department of Education.

U.S. Department of Education. National Center for Education Statistics. (1998b). Trends in the reading proficiency of 9-, 13-, and 17-year-olds. The condition of education 1998. Retrieved on November 24, 1999 from the World Wide Web at: http://nces.ed.gov/ pubs98/condition98/c9816a01.html.

U.S. Department of Education. National Center for Education Statistics. (1998c). Access, participation, and progress. The condition of education 1998. Retrieved on November 24, 1999 from the World Wide Web at: http://nces.ed.gov/pubs98/condition98/ c98seca.html.

U.S. Department of Education. National Center for Education Statistics. (1998d). Recent high school dropouts. The condition of education 1998. Retrieved on November 24, 1999 from the World Wide Web at: http://nces.ed.gov/pubs98/condition98/ c9806a01.html.

U.S. Department of Education. National Center for Education Statistics. (1998e). Table 44-1. Family characteristics of 15- to 18-year-olds: 1972-97. The condition of education 1998. Retrieved on November 24, 1999 from the World Wide Web at: http:// nces.ed.gov/pubs98/condition98/c980644d01.html.

U.S. Department of Education. National Center for Education Statistics. (1998f). Conclusions. The condition of education 1998. Retrieved on November 24, 1999 from the World Wide Web at: http://nces.ed.gov/pubs98/condition98/summary-conclusionx.html.

U.S. Department of Education. National Center for Education Statistics. (1998g). Access to higher education. (Table 8.1. Percentage distribution of 1992 high school graduates qualified for admission to a 4-year institution, by race/ethnicity and family income: 1992-94). The condition of education 1998. Retrieved on November 24, 1999 from the World Wide Web at: http://nces.ed.gov/pubs98/condition98/c9808a01.html; http://nces.ed/gov/pubs98/condition98/c9808d01.html.

U.S. Department of Education. National Center for Education Statistics. (1998h). Table 22-2 and 22-3. Percentage of 25-to 29-year-old high school completers with 1 or more years of college, by race/ethnicity and sex: March 1971-97; Table 22-3 Percentage of 25-to 29-year-old high school completers with 4 or more years of college, by race/ ethnicity and sex: March 1971-97. The condition of education 1998. Retrieved on November 24, 1999 from the World Wide Web at: http://nces.ed.gov/pubs98/condition98/c9808d02.html; http://nces.ed.gov/pubs98/condition98/c9808d03.html

U.S. Department of Education. National Center for Education Statistics. (1998i). Climate and diversity of educational institutions. The condition of education 1998. Retrieved on November 24, 1999 from the World Wide Web at: http://nces.ed.gov/pubs98/ condition98/c98sece.html

U.S. Department of Education. National Center for Education Statistics. (1998j). Public school expenditures per pupil (in 1998 constant dollars), by function and selected district characteristics: School year 1993-94. The condition of education 1998. Retrieved on November 24, 1999 from the World Wide Web at: http://nces.ed.gov/pubs98/ condition98/c9853a01.html.

U.S. Department of Education. National Center for Education Statistics. (1998k). Internet access in public schools (NCES 98-031, March). Retrieved December 1, 1999 from the World Wide Web: http://nces.ed.gov/pubs98/98031.pdf.

U.S. Department of Education. National Center for Education Statistics. (1999a). Early literacy experiences in the home (NCES 1999-003). Retrieved October 25, 1999 from the World Wide Web: http://nces.ed.gov/literacy-98.pdf.

U.S. Department of Education. National Center for Education Statistics. (1999b). Learner outcomes. The condition of education 1999. Retrieved November 24, 1999 from the World Wide Web: http://nces.ed.gov/pubs00/condition99/summary1x.html.

U.S. Department of Education. National Center for Education Statistics. (1999c). Family characteristics of 6- to 12-year-olds. The condition of education 1999. Retrieved November 24, 1999 from the World Wide Web: http://nces.ed.gov/pubs00/condition99/indicator-36.html.

U.S. Department of Education. National Center for Education Statistics. (1999d). Percentage of children ages 3-5 who participated in various reading activities with a parent or family member, by selected characteristics: 1991, 1994, 1995, and 1996. The condition of education 1999. Retrieved November 24, 1999 from the World Wide Web: http:// nces.ed.gov/pubs99/condition99/indicator-34.html.

U.S. Department of Education. National Center for Education Statistics. (1999e). Conclusions. The condition of education 1999. Retrieved November 24, 1999 from the World Wide Web: http://nces.ed.gov/pubs99/condition99/summary-conclusionx.html.

U.S. Department of Education. National Center for Education Statistics. (1999f). National indicators of public investment in education. The condition of education 1999. Retrieved November 24, 1999 from the World Wide Web: http://nces.ed.gov/pubs99/condition99/ summary5x.html.

U.S. Department of Education. National Center for Education Statistics. (1999g). Educational participation and progress. The condition of education 1999. Retrieved November 24, 1999 from the World Wide Web: http://nces.ed.gov/pubs99/condition99/ summary5x.html.

U.S. Department of Education. National Center for Education Statistics. (1999h). Social support for learning. The condition of education 1999. Retrieved November 24, 1999 from the World Wide Web: http://nces.ed.gov/pubs99/condition99/summary4x.html.

U.S. Department of Education. National Center for Education Statistics. (1999i). Higher education expenditures per student. The condition of education 1999. Retrieved November 24, 1999 from the World Wide Web: http://nces.ed.gov/pubs99/condition99/ indicator-40.html.

U.S. Department of Education. National Center for Education Statistics. (1999j). Internet access in public schools and classrooms: 1994-1998. Retrieved December 1, 1999 from the World Wide Web http://nces.ed.gov/pubs99/1999017.html.

U.S. Department. of Education. Nation:mi Center for Education Statistics. (1999k). Table 419--Percent of public schools and school classrooms having access to the Internet, by school characteristic: 1994 to 1998. Retrieved December 3, 1999 from the World Wide Web: http://nces.ed.gov/pabs99/table414.pdf.

U.S. Department of Education. National (:enter for Education Statistics. (1999l). Federal programs for education and related activities. The digest of education statistics 1999. Retrieved November 29, 1999 from the World Wide Web: http://nces.ed.gov/pubs99/ digest98/chapter4.html.

U.S. Federal Interagency Forum on Child and Family Statistics. (1998a). America's children 1998. Retrieved January 17,1999 from the World Wide Web: http://www.childstats.gov/ac 1998.

U.S. Federal Interagency Forum on Child and Family Statistics. (1998b). Table POP4. Difficulty speaking English: Children ages 5 to 17 who speak a language other than English at home, and who are reported to have difficulty speaking English by race and Hispanic origin, region, selected years 1979-95. America's children 1998. Retrieved January 17, 1999 from the World Wide Web: http://www.childstats.gov/ac1998.

U.S. Federal Interagency Forum on Child and Family Statistics. (1998c). Table ED1. Family reading: Percentage of children ages 3 to 5 who were read to every day in the last week by a family member by child and family characteristics, selected years, 1993-96. America's children 1998. Retrieved January 17, 1999 from the World Wide Web: http:// www.cldstate.gov/ac1998/ed1.htm.

U.S. Federal Interagency Forum on Child and Family Statistics. (1998d). Table ED2B. Early childhood education: Percentage of children Ages 3 to 4 enrolled in center-based programs by child and family characteristics, selected years, 1991-96. America's children 1998. Retrieved January 17, 1999 from the World Wide Web: http://www.childstats.gov/ ac1998/ed2b.htm.

U.S. Federal Interagency Forum on Child and Family Statistics. (1998e). Table SPECIAL2.. Number and percentage of children under age 6 participating in child care and early education programs on a regular basis by type of arrangement and child and family characteristics, 1995. America's children 1998. Retrieved January 17,1999 from the World Wide Web: http://www.childstats.gov/ac1998/special2.htm.

U.S. Federal Interagency Forum on Child and Family Statistics. (1998f). Population and family characteristics. America's children 1998. Retrieved January 17, 1999 from the World Wide Web: http://www.childstats.gov/ac1998/poptxt.htm.

U.S. Federal Interagency Forum on Child and Family Statistics. (1998g). Table ED3.A. Mathematics achievement: Average scale scores of students ages 9, 13, and 17 by age and child and family characteristics, selected years 1982-96. Retrieved June 15, 2000 from the World Wide Web: http://www.childstats.gov/ac1998/ed3a.htm.

Weinberg, D. H. (1996). A brief look at postwar U.S. income inequality. Current population reports (Household Economic Studies [P60-191]). Washington, DC: U.S. Bureau of the Census.

Alice Robbin, School of Library and Information Science, Indiana University, Bloomington, IN 47401

ALICE ROBBIN is currently associate professor at the School of Library and Information Science at Indiana University at Bloomington. Her research focuses on the political controversy surrounding the reclassification of racial and ethnic data, the symbolic aspects of classification systems, the role of the decennial census in creating identity, the history of minority statistics in the federal statistical system, and administrative decision making. Recent articles have appeared in the Journal of the American Society for Information Science, Administration & Society, and Journal of Government Information.