CHAPTER 2
Socioeconomic and Demographic Risks

Very little information is available about the effects of either risk or, to an even greater extent, protective factors for children in immigrant families, or indeed about other factors that seem likely to be influential specifically for children in immigrant families (see Chapters 1 and 3 for a discussion of these). In order to expand available information about the incidence of socioeconomic and demographic risk factors among first-, second-, and third- and later-generation children, new analyses of decennial census data for 1910, 1960, and 1990 were conducted for the committee (Hernandez and Darke, 1998).

These analyses focused especially on information from the 1990 decennial census, because it is the most recent source of information on risk factors for children with origins in a large number of countries. These results were supplemented insofar as possible with comparative data using the 1910, 1960, and 1990 decennial censuses to examine historical changes in key risk factors following the decade of peak immigration to the United States (1901-1910), the subsequent era of very low immigration (1931-1960), and the most recent decades of increasing immigration for which census data are available (1970-1990).

This assessment is the first to use decennial census data with children as the unit of analysis to study long-run historical changes in foreign-born and U.S.-born families. It draws on ana-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families CHAPTER 2 Socioeconomic and Demographic Risks Very little information is available about the effects of either risk or, to an even greater extent, protective factors for children in immigrant families, or indeed about other factors that seem likely to be influential specifically for children in immigrant families (see Chapters 1 and 3 for a discussion of these). In order to expand available information about the incidence of socioeconomic and demographic risk factors among first-, second-, and third- and later-generation children, new analyses of decennial census data for 1910, 1960, and 1990 were conducted for the committee (Hernandez and Darke, 1998). These analyses focused especially on information from the 1990 decennial census, because it is the most recent source of information on risk factors for children with origins in a large number of countries. These results were supplemented insofar as possible with comparative data using the 1910, 1960, and 1990 decennial censuses to examine historical changes in key risk factors following the decade of peak immigration to the United States (1901-1910), the subsequent era of very low immigration (1931-1960), and the most recent decades of increasing immigration for which census data are available (1970-1990). This assessment is the first to use decennial census data with children as the unit of analysis to study long-run historical changes in foreign-born and U.S.-born families. It draws on ana-

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families lytical approaches to identifying first-, second-, and third- and later-generation children developed during the last few years (Hernandez, 1993; Jensen and Chitose, 1997; Landale et al., 1997; Oropesa and Landale, 1995, 1997a, 1997b). These data are the best available for assessing the process of assimilation by comparing the socioeconomic and demographic risk factors of first-, second-, and third- and later-generation children from various countries (see Appendix Tables B-1A through B-1E and B-2A through B-2E). Nevertheless, conclusions must be treated as preliminary here, as throughout the report. Differences from generation to generation may to some extent result from changes in the characteristics of immigrants from decade to decade or even year to year (Borjas, 1991). In addition, in census data, second-generation children can be identified by the foreign-born status of their parents only if they live in the home; hence some second-generation children may be misclassified as belonging to later generations.1 This chapter presents information on the exposure of first-, second-, and third- and later-generation children to risk factors known to affect children generally: poverty, limited parental education and employment, living in a one-parent family or with many siblings, and in overcrowded housing. It then discusses potential risk factors specific to children in immigrant families: English language fluency, living in a linguistically isolated household, and not being a U.S. citizen or having parents who are not U.S. citizens. GENERAL CHILDHOOD RISK FACTORS For children generally, negative outcomes have been demonstrated to result not only from poverty, but also independently from low parental educational attainments and from living in families with only one parent or with a large number of siblings 1   In addition, among children not living with a parent, second generation children cannot be distinguished from third- and later-generation children. To ensure that estimates for various risk factors are maximally comparable across generations, children with no parent in the home are excluded from these estimates.

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families (Alwin, 1984; Blake, 1985; Blau and Duncan, 1967; Featherman and Hauser, 1978; Kohn, 1969; Kohn and Schooler, 1983; Kominski, 1987; McLanahan and Sandefur, 1994; Sewell and Hauser, 1975; Sewell et al., 1980). In addition, overcrowded housing conditions can facilitate the transmission of communicable diseases such as tuberculosis, hepatitis A, and intestinal and respiratory infections (Coggon et al., 1993; Fall et al., 1997; Guberan, 1980; Paul et al., 1993; Rosenberg et al., 1997). We discuss each of these risk factors in turn for children in immigrant families. Poverty In 1990, children in immigrant families were somewhat more likely than U.S.-born children in U.S.-born families to live in poverty (22 versus 17 percent).2 Most of the difference was accounted for by the high poverty rate of first-generation immigrant children (33 percent), whereas the second generation was only slightly more likely (19 percent) to be poor than third- and later-generation U.S.-born children (17 percent). In 1960, in contrast, first- and second-generation children were less likely to be poor than third- and later-generation children (23 and 19 percent versus 26 percent), although, as in 1990, the proportion in poverty was greater for the first than for the second generation. Variations by Country of Origin Poverty rates differed enormously in both 1960 and 1990 for first- and second-generation children from various countries of origin, and for third- and later-generation children by race and ethnicity. For example, in 1990 the poverty rate for third- and later-generation white children was only 11 percent, and it was 2.5 to 4 times greater for third- and later-generation black, Hispanic, and American Indian children, at 40, 28, and 35 percent, respectively. 2   For limitations of the current official poverty measure for current and historical comparisons see National Research Council (1995) and Hernandez (1993).

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families Similarly, in 1960 and 1990 poverty rates for first- and second-generation children ranged widely, from a low of about 5 percent to a high of about 50 percent. For children in immigrant families from about two dozen countries in Latin America and the Caribbean, Asia, Europe, the Middle East, and Africa, poverty rates were about equal to, or even substantially less than, the rate of 11 percent for third- and later-generation white children in 1990 (see Appendix Table B-1A). Children with origins in these countries accounted for 28 percent (2.3 million) of all children in immigrant families in 1990 (8.4 million). At the other end of the spectrum, for children in immigrant families from 12 countries, poverty rates exceeded 25 percent in 1990 (the range for this group was 26 to 51 percent, depending on the country of origin). In view of the negative risks associated with poverty generally, the situation of children from these 12 countries may be particularly serious. Children with origins in these 12 countries accounted for 46 percent (3.9 million) of all children in immigrant families in 1990. Mexico alone accounted for 31 percent of children in immigrant families (2.6 million). Moreover, children from these 12 countries whose family income was below the official poverty threshold in 1990 accounted for about 80 percent of all children in immigrant families who lived in poverty; these estimates may be low, given evidence that the decennial census underestimates the number of Mexican-origin children living in poverty.3 Of the 12 countries, 5 are the source of many officially recognized refugees (the former Soviet Union, Cambodia, Laos, Thailand, Vietnam); 3 are war-torn countries in Central America (El 3   Analyses carried out for the committee using the National Agricultural Workers Survey indicate that more than 67 percent of U.S.-based children in migrant farmworker families lived in poverty in each year from 1993 to 1995, that is, more than 590,000 of the 880,000 total (Mines, 1998). Insofar as a substantial portion of migrant farmworker families and their children are of Mexican origin (69 percent of the U.S.-based children in the survey were from Mexico) and are not counted in the decennial census, the total number (and percentage) of children in immigrant families, especially of Mexican origin, who were living in poverty is higher, perhaps by several hundred thousand (and several percentage points), than indicated by the decennial census data.

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families Salvador, Guatemala, Nicaragua); and 3 are small, impoverished Central American or Caribbean countries (Honduras, Haiti, Dominican Republic) that are sources of unskilled labor. The 12th country is Mexico, which currently sends the largest number of both legal and illegal unskilled immigrants, and which has been a ready source of unskilled labor for the U.S. economy throughout the 20th century (Romo, 1996; Rumbaut, 1996). Within the racial and ethnic stratification system of the United States, most children from these 12 countries, except the former Soviet Union, are classified as minority—Hispanic, Asian, or black. Variations by Generation In 1990, poverty rates for children in immigrant families were lower, sometimes much lower, for second-generation children than for first-generation children for nearly all countries of origin, including most of the 12 countries with the highest poverty rates (see Appendix Table B-2A). But for children from Mexico, who account for about two-thirds of the children in immigrant families from these 12 countries, the poverty rates for the second and the later generations were quite similar, at 32 and 28 percent, respectively, which is 2.5 to 3 times greater than for third- and later-generation white children. For children from the 4 Central American countries in the cluster of 12 high-poverty countries (El Salvador, Guatemala, Nicaragua, Honduras), the decline in poverty from the second to the third and later generations is somewhat larger than for Mexican-origin children. It is significant that the poverty rates for the later generations drop to the range of 14 to 17 percent, only somewhat greater than for third- and later-generation white children. The lower poverty rates among third- and later-generation children from these four countries, compared with those from Mexico, may reflect differences in the socioeconomic status of their parents at the time they entered the United States rather than intergenerational socioeconomic assimilation. In 1990, children in immigrant families with origins in the Dominican Republic and Haiti had extremely high poverty rates, and they were nearly the same for the first and second generations (41 and 42 percent, respectively, for the Dominican Repub-

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families lic, 30 and 26 percent for Haiti). Available data for the Dominican Republic indicates no change for the third and later generations (40 percent). The high poverty rates of second- and third- and later-generation children from these Caribbean countries and from Mexico suggest the possibility that racial and ethnic stratification has been restricting socioeconomic opportunities across the generations. In the case of immigrants from Mexico, the pattern has remained the same for many generations—a conclusion that is strikingly similar to that for U.S.-born black children. Since at least 1960, the intergenerational pattern for Mexican-origin children has been quite similar, with poverty rates for the first, second, and third and later generations of 58, 48, and 53 percent, respectively, about 2.5 to 3 times greater than the rate of 19 percent in 1960 for third-and later-generation white children. 4 This persistent pattern of very high poverty rates across generations suggests that ethnic stratification may have continuing power in determining the life chances of children of Mexican origin. Potential additional or alternative explanations that merit attention in future research, particularly for the Mexican-origin population, include their continuing high levels of immigration, the extent of back-and-forth movement between the United States and Mexico, and the large size and residential concentration of the Mexican-origin population, as well as the declining need for very low-skilled entry-level workers in the U.S. Economy. 4   Of course, it is possible that a substantial portion of third- and later-generation children with a Mexican parent or grandparent also have a non-Mexican parent or grandparents, and that such children tend not to be reported as being of Mexican origin and tend to have lower poverty rates than children with two Mexican parents who are reported as being of Mexican origin. As of 1990, because only 9 percent of third- and later-generation children who were identified as Mexican or as having at least one Mexican parent had a Mexican parent but were not themselves identified as Mexican, the exclusion of these children from the poverty estimates above could not affect the poverty rates of third- and later-generation Mexican children by more than a percentage point or two. Additional research is required to assess the effect of marriage between Mexican origin and non-Mexican-origin grandparents.

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families Parents' Education and Employment The percentage of children whose parents have graduated from college was very similar for those in immigrant and U.S.-born families in 1990. At the same time, however, children in immigrant families were much more likely than children in U.S.-born families to have parents with very low educational attainments. In other words, there is comparability at the top of the spectrum but, below this threshold, children in immigrant families are more likely to be concentrated at the lower end. Specifically, first- and second-generation children in families with fathers in the home were about as likely in 1990 to have fathers who were college graduates (23 percent) as were third-and later-generation children (26 percent). First-, second-, and third- and later-generation children in families with mothers in the home were also about equally likely to have mothers who were college graduates (14 to 18 percent). In addition, in 1990, among children in immigrant families from about two dozen countries, 35 percent or more had a father in the home who was a college graduate—a rate that is higher than the 28 percent recorded for third- and later-generation white children (see Appendix Tables B-1D and B-2D). At the other end of the spectrum, in 1990, first- and second-generation children living with fathers were 2 to 3 times more likely than third and later generations to have fathers who had not graduated from high school, at 49, 36, and 15 percent, respectively. Those with fathers in the home who had completed no more than eight years of schooling for the three generations were 34, 23, and 3 percent, respectively. Patterns in mothers' educational attainment were quite similar. Mexican-origin children constitute a large portion of the children in immigrant families with very low parental educational attainments (see Appendix Tables B-1A, B-1D, B-2A, and B-2D). In 1960, parental educational attainment followed a similar pattern of improvement from the first to the third generation. It is striking, however, that the second and third generations showed substantially higher rates of very low educational attainment in 1960 than in 1990. Among first-, second-, and third- and later-generation children with fathers in the home in 1960, the

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families proportions with fathers in the home who had completed fewer than eight years of schooling were 41, 40, and 31 percent, respectively. The only measure of educational attainment in the 1910 census is the literacy rate. Second and later generations of children were similar in their chances of having a parent in the home who was illiterate, at 9 to 14 percent, but the first generation was substantially more likely to have an illiterate father (22 percent) or mother (34 percent). Variations by Country of Origin and Generation Parental educational attainment varied enormously by country of origin for children in immigrant families, and by race and ethnicity among U.S.-born children in U.S.-born families (see Appendix Tables B-1A, B-1D, B-2A, and B-2D), both historically and today. In 1990, with the exception of the former Soviet Union, children in immigrant families from the 12 countries with the highest poverty rates were somewhat to much more likely than third- and later-generation white children to have parents in the home who had not graduated from either high school or elementary school. For children from 11 of these countries (all but Mexico), parents' educational attainment generally increased substantially from the first to the second to the third generation. Although most third- and later-generation children from these countries have parents who have completed eight years of schooling or more, it is also true that most have parents who have not completed college. The proportion of third- and later-generation Mexican-origin children with parents not graduating from high school remains in the range of 30 to 34 percent for all generations, similar to the level for third- and later-generation black children (26 to 29 percent), and substantially higher than for third- and later-generation white children (12 percent). Earlier in the century, too, there was a substantial deficit in parental educational attainment (1960) and literacy (1910) among children of Mexican origin of all generations, and among third- and later-generation black children, compared with third- and later-generation white children. Throughout the century, then, these racial and ethnic minorities

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families have been more likely to experience low parental educational attainment than third- and later-generation white children. Relation Between Parental Education, Employment, and Poverty Throughout the century, the vast majority of children of all generations had fathers who were in the labor force. The proportions of children with fathers in the home who were in the labor force were essentially identical in 1910 (95 to 96 percent) and 1960 (97 percent), but slightly lower for the first generation in 1990 than for the second and third and later generations, at 88, 94, and 95 percent, respectively. Children of all generations have also been similar, historically, in their likelihood of having a mother in the labor force, although enormous increases occurred for all generations during the past century. Among first-, second-, and third- and later-generation children with mothers in the home, the proportions with their mother in the labor force were 7, 6, and 12 percent, respectively, in 1910, increasing to 33, 25, and 27 percent, respectively, in 1960. By 1990, 55, 58, and 66 percent, respectively, had mothers in the labor force. It is not surprising that very low parental educational attainment characterizes children from the 12 countries of origin with very high poverty rates in 1990. Yet among these children, poverty and low parental educational attainment were not usually associated with especially lower rates of labor force participation by fathers and mothers. For example, only for children in immigrant families from Cambodia, Laos, and Thailand was low parental educational attainment accompanied by low parental labor force participation rates. The high proportions of children in immigrant families with parents in the labor force, including those with origins in these 12 countries with especially high poverty and lower parental educational attainments, suggests that immigrants may be more likely to have high levels of ambition and motivation to work than those who do not immigrate. It is the case, however, that low parental educational attainment and poverty in these 12 countries are associated with especially high pro-

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families portions of children in immigrant families whose fathers do not have full-time, year-round jobs (ranging from 31 to 68 percent) (see Appendix Tables B-1C and B-2C). Children in immigrant families from an additional and sizeable number of countries (especially Israel, Nigeria, Pakistan, Jamaica, Syria, and Japan) also have very high proportions of fathers who do not work full-time year-round, but they are not characterized by extremely high poverty rates. What appears to distinguish these children with lower poverty rates is that most live with at least one person who speaks English exclusively or very well, and most have parents with more than eight years of education. Thus, despite generally high proportions with fathers and mothers in the home who are in the labor force, very high poverty rates for children in immigrant families tend to occur among children from countries with very low parental educational attainment (no more than eight years of schooling), fathers who cannot find full-time, year-round work, and parents who do not speak English well. One-Parent Families First-generation children from most countries of origin, and, to an even greater extent, second-generation children, were less likely to live in a one-parent family in 1990 than were third- and later-generation white children. Important exceptions are children in immigrant families with origins in Cambodia and most Central American and Caribbean countries. First- and second-generation children of Mexican origin were about as likely in 1990 to live in a one-parent family as were third- and later-generation white children (23, 18, and 18 percent, respectively). The proportion of children living in one-parent families increased, sometimes dramatically, from the second to the third generation. For example, the share of third- and later-generation children with origins in most countries of Central and South America and the Caribbean living in one-parent families was at least twice as great as for third- and later-generation white children.

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families Relation Between Poverty and One-Parent Families Among children in immigrant families in 1990, poverty was not necessarily associated with high proportions living in a one-parent family. For example, looking at the 12 countries of origin with very high poverty rates, children in immigrant families from Laos, the former Soviet Union, Mexico, Thailand, and Vietnam showed rates of living in one-parent families that were lower than or approximately the same as the rate for third- and later-generation white children. In contrast, those from the Central American and Caribbean countries and from Cambodia were substantially more likely to live in one-parent families than third- and later-generation white children. Historical Trends in One-Parent Families First-, second-, and third- and later-generation children were about equally likely, overall, to live in one-parent families in 1910 and in 1960, at 9 to 10 percent for children who lived with at least one parent. First-generation children in families from Mexico were, however, much more likely in both 1910 and 1960 (but not in 1990) than third- and later-generation white children to live in one-parent families, although the differences disappeared by the second generation. The same was true for first- and second-generation children from Central America and the Caribbean in 1960. One-fourth of third- and later-generation black children lived in one-parent families in 1960, more than for nearly every country of origin for first-generation children, except perhaps the Dominican Republic. In 1910, third- and later-generation black children were substantially more likely (19 percent) than others to live in a one-parent family, with the exception of first-generation children from Mexico (24 percent). Families with Many Siblings The proportion of children living in families with five or more siblings in 1990 declined from 17 percent for the first generation to 9 percent for the second and to 5 percent for the third and later

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families generations. For most specific countries of origin, not only did smaller proportions of second-generation children in 1990 live in large families than first-generation children from the same countries, but also the proportions for the second generation by country of origin were usually similar to third- and later-generation white children, at 5 percent or less. These differences in the number of siblings in the homes of first-, second-, and third- and later-generation children are consistent with changes in fertility measured in the 1990 census (Chiswick and Sullivan, 1995); immigrant women have higher fertility than native-born women, but there is a convergence in fertility to U.S. norms across generations. In 1910, first-, second-, and third- and later-generation children as a whole were about equally likely to live in families with many siblings; the range was 38 to 43 percent. The proportions in such large families were much smaller in 1960, but they remained similar across the generations, within the range of 17 to 21 percent. But the proportions living in large families in 1960 were much higher for first- and second-generation children of Mexican origin and for third- and later-generation black and Hispanic children, at 40 to 51 percent. These proportions were nearly as high as they had been in 1910, when the range was 47 to 61 percent. By 1990, among first- and second-generation children of Mexican origin and among third- and later-generation black and Hispanic children, the proportions living in large families had fallen to the range of 8 to 19 percent. Overcrowded Housing In 1990, only 12 percent of third- and later-generation children lived in overcrowded housing with more than one person per room, compared with 38 percent for the second generation and 62 percent for the first generation. Children in immigrant families from most specific countries of origin in 1990 also had high proportions living in overcrowded housing, and children in immigrant families from the 12 high-poverty countries were much more likely than most to live in such conditions (Appendix Tables B-1B and B-2B). For children from most of these 12 countries, declines in overcrowding are substantial across the first, second, and third and later generations, but the third generation from

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families countries for which data are available continued to experience high levels of overcrowding, especially Mexican-origin children at 31 percent. Overcrowding cannot be measured in the 1910 census, but overall levels of overcrowding were much higher among children in 1960 than in 1990. In 1960, the proportions living in overcrowded housing were about equal for first-, second-, and third- and later-generation children, at 31 to 36 percent. However, 75 percent of first- and second-generation Mexican-origin children and 69 percent of third- and later-generation black and Hispanic children lived in crowded conditions in 1960. Summary Children in immigrant families in 1990 were less likely than U.S.-born children in U.S.-born families to have only one parent in the home, but they were substantially more likely to live in poverty, with many siblings, with parents who had not finished school beyond the eighth grade, and in overcrowded housing. Children in immigrant families were similar to those in U.S.-born families with fathers in the home in having a father who was in the labor force, but substantially more likely to have a father who did not work full-time, year-round. But the socioeconomic and demographic risk factors experienced by children in immigrant families from various countries are extremely diverse. Children in immigrant families from about two dozen countries experience socioeconomic and demographic circumstances similar to or better than third- and later-generation children. At the other extreme are the children in immigrant families from 12 other countries with very high poverty rates, who experience socioeconomic and demographic circumstances in the range experienced by third- and later-generation black, Hispanic, and American Indian children. Children from these 12 countries, or their parents, entered the United States as officially recognized refugees from Southeast Asia or the former Soviet Union in order to leave the dangerous conditions behind, fled warring countries in Central America, or

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families came as unskilled migrant laborers from the Caribbean or Mexico in search of improved economic opportunities. With the exception of the former Soviet Union, children from these countries experience not only very high levels of poverty, but also very low parental educational attainment and a high likelihood of living in overcrowded housing. Low parental educational attainment appears to contribute to poverty not by leading to low rates of labor force participation (Laos, Cambodia, and Thailand are the main exceptions), but instead because fathers with very limited education who do not speak English very well do not have full-time, year-round work. The proportions of children exposed to important socioeconomic and demographic risk factors declined for most of these factors between the first and second generation for children from most of the 12 countries of origin with high child poverty rates. But data available for selected countries suggest that, for third- and later-generation children from Mexico and the Dominican Republic, and perhaps the Central American countries, the proportions living in poverty with parents who have not graduated from high school, in overcrowded housing conditions, and with only one parent remain quite high. Children in immigrant families from Mexico account for nearly two-thirds of the children with origins in these 12 countries, and for nearly one-third of all children in immigrant families in 1990. Thus, Mexican-origin children account for a large proportion of children in immigrant families who experience high rates of poverty, low parental educational attainments, and fathers who do not work full-time, year-round. The history of immigration from Mexico is unique, in that Mexico has for many decades been an important source of low-skilled labor for the U.S. economy, and it has also been the most important source of undocumented immigrants. Mexican-origin children of all generations have also, throughout the century, been among those with high proportions exposed to elevated socioeconomic and demographic risks.

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families RISK FACTORS SPECIFIC TO CHILDREN IN IMMIGRANT FAMILIES English Language Fluency Children in immigrant families from countries in which English is not the native language or is not widely taught may be at special risk, compared with U.S.-born children in U.S.-born families, because they may not speak English well or they may live with parents who do not speak English well. A lack of English fluency can limit effective communication and functioning in health facilities, schools, and other settings that provide essential resources to children and their families. In 1990, at least 60 percent of children in immigrant families spoke a language other than English at home, regardless of their own proficiency with English. The exceptions were English-speaking countries of origin, as well as Austria, Germany, the Netherlands, Nigeria, and South Africa. In contrast, using as an outcome the percentage of children who do not speak English ''exclusively or very well,"5 only in 13 countries of origin did the proportion of children in immigrant families reach the substantial proportion of 30 percent or more; 11 of these countries are among the 12 high-poverty countries (all but Haiti); the remaining two were China and Hong Kong (see Appendix Tables B-1E and B-2E). Generational differences are large, however. The proportion of children who speak English "exclusively or very well" is only 54 percent for the first generation but 81 percent for the second generation. For children in immigrant families from the 12 countries with very high poverty rates, the range is only 35 to 53 percent for the first generation, but this rises for 10 of the 12 countries (excepting only Laos and Cambodia) to 68 percent or more for the second generation. 5   Among children age 5 years and older, those who speak only English in the home are categorized as speaking English exclusively, and those who speak a language other than English in the home are distinguished according to 4 categories, whether they speak English very well, well, not well, or not at all.

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families Looking to the future, it will be important to assess whether and to what extent factors that may result from the expanding Spanish-speaking immigrant population, such as growing up in homogeneously Spanish-speaking neighborhoods or watching Spanish-language TV, will affect the acquisition of fluent English among children in immigrant families from Spanish-speaking countries. Linguistic Isolation Lack of English fluency may not pose enormous difficulties for immigrants in communities that have a large number of people with the same national origin—but it can isolate them from mainstream society. The Census Bureau defines a linguistically isolated household as one in which no person age 14 or older speaks English either "exclusively" or "very well." In 1990, 76 to 78 percent of children in immigrant families lived with a mother or a father who did not speak English at home. In households with both mother and father at home, the proportion was 70 percent. No language information was collected in the 1960 census, but historical changes are best measured by comparing data on "mother tongue" for 1910 with data on "language spoken'' in 1990. In 1910, 84 to 85 percent of children in immigrant families lived with either a father or a mother whose mother tongue was not English. For 79 percent of children in immigrant families in 1910 with two parents in the home, neither parent spoke English as a mother tongue. Although these measures of language are not identical, they are similar, and the similarity of the results for 1910 and 1990 suggests that differences in the proportion of children in immigrant families with parents speaking or not speaking English were about the same at the beginning and the end of the century. Among children in immigrant families from each of the 12 high-poverty countries, 30 percent or more lived in linguistically isolated households. The proportion was over 40 percent for 9 of these countries, and at 60 percent for two of them (Laos and Cambodia). Four additional countries of origin had 30 percent or more children living in linguistically isolated households (China, Hong Kong, Taiwan, Colombia).

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families Citizenship Of the 8.4 million children in immigrant families in 1990, 75 percent were U.S. citizens by birth, 4 percent were naturalized citizens, and 21 percent (1.7 million) were not citizens. Of the citizen children, 54 percent (3.6 million) had at least one parent in the home who was not a citizen; thus, approximately two-thirds of children in immigrant families in 1990 were either themselves not a citizen or lived with a noncitizen parent. Children who are illegal immigrants are ineligible for most public benefits and services and, under welfare reform, those who are legal immigrants but not citizens may also be ineligible for important medical and social services (see Chapter 4). Equally important, U.S.-born children in immigrant families who are eligible for such services may not receive them, because immigrant parents who are not themselves eligible may not be aware that their children are eligible, or they may fear or resist contact with government agencies administering the services. Because legal immigrants and citizens experienced essentially the same eligibility prior to welfare reform, for legal immigrants the fact of not being a U.S. citizen has only recently become a potential risk factor. Welfare eligibility exclusions are most significant to children living in poverty. in 1990, the official poverty rate was 34 percent among children who were not citizens and 23 percent among citizen children with at least one noncitizen parent. For all children in immigrant families, the poverty rate in 1990 was 27 percent. Children in immigrant families from 8 of the 12 high-poverty countries of origin were especially likely to be noncitizens, with rates of 30 percent or more. For children from the four remaining high-poverty countries (the Dominican Republic, Mexico, Honduras, Haiti), the proportion who were not citizens was 20 to 29 percent. For three additional countries of origin (Venezuela, Romania, Guyana) with child poverty rates at least as high as the rate for third- and later-generation white children (11 percent), the proportion of children who were not citizens was 30 percent or more (see Appendix Table B-1E). In 1990, among children from 10 of the 12 high-poverty countries or origin, the proportion of children who were not citizens or

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families had at least one parent who was not a citizen was 73 percent or more. Among the two remaining high-poverty countries, the former Soviet Union and Vietnam, the proportions of children who were not citizens or who had at least one parent who was not a citizen were 62 to 63 percent. The figure was 50 percent or more for 18 of the other 26 countries of origin with child poverty rates at least as high as the rate for third- and later-generation white children. Thus eligibility rules that exclude persons who are not citizens from public benefits and services may have important consequences for children from many different countries of origin. Traumatic Circumstances Children who have witnessed the horrors of war firsthand, including the killing of parents or siblings, may have special needs, particularly for mental health services, that are especially serious. Children with these experiences and others who enter the United States unaccompanied by an adult and without documentation face severe difficulties. However, very little information is available about them. The Immigration and Naturalization Service (INS) arrested 2,028 minors in 1995, the last year for which data were available (McDonnell, 1997). In 1990, the INS arrested 8,500 undocumented minors (under age 18), 70 percent of whom were unaccompanied by an adult (Human Rights Watch, 1997). Most detained children are placed in foster care or a relative's care or are deported in a matter of days. At any one time, several hundred to more than 1,000 undocumented children are in longer-term detention centers. Minors are held in about 100 detention centers that range from nonsecure foster care facilities to adult correctional facilities. About one-third of children in the legal guardianship of the INS in places of detention are younger than 15 years old, and some are under age 10. In addition to lack of legal counsel, one brief survey of undocumented children in INS detention revealed that they were exposed to physical and sexual assaults, verbal abuse, denial of medical services, and nutritional deprivation (one meal per day) in some facilities. Lack of health care, especially mental health

OCR for page 40
From Generation to Generation: The Health and Well-Being of Children in Immigrant Families services, is a particular concern, because these minors may have been seriously traumatized prior to entering the United States or while held in detention (Nadeau et al., 1997). The number of minors imprisoned by the INS does not appear to be large, but the lack of publicly available information about unaccompanied children in detention is an important gap in the knowledge about immigrant children at risk. Summary Children in immigrant families may be subject to risk factors that grow out of their immigrant circumstances. Lack of English fluency can limit effective communication and functioning in health facilities, schools, and other settings that provide resources essential to children and their families. However, although most children in immigrant families speak a language other than English at home, the vast majority of them (73 percent) speak English exclusively or very well, and language assimilation occurs rapidly across generations (National Research Council, 1997). With the passage of welfare reform legislation, lack of U.S. citizenship became a potentially important risk factor, by limiting eligibility and access of noncitizens to public benefits and services. A large majority of children in immigrant families may be ineligible for important benefits, or have parents who are ineligible and who are therefore hesitant to secure benefits on behalf of their children. Moreover, reductions in benefits available to such families will reduce their overall resources. One other very high risk group merits attention: children who have emigrated under traumatic circumstances, especially from Southeast Asia and Central America, including those with firsthand experience of the horrors of war.