Evidence for residential segregation for Hispanics and Asians, taking immigrants and native-born persons together, has come from several recent censuses. Asians, as a group, display low to moderate levels of segregation from whites (Massey and Denton, 1992; Fong, 1994). Asians tend to be highly suburbanized and, because they have low levels of segregation within suburbs, their overall levels are reduced. Asian-white segregation is reduced as the socioeconomic status of Asians increases, with very low segregation levels for higher education and income levels. Taken as a whole, "Asians appear to experience few barriers to residential mobility and display remarkably low levels of segregation" (Massey and Denton, 1992:170).3

One difficulty with using the census is that the geographic areas identified are at a very high level of aggregation. Borjas (1995) used data from the National Longitudinal Survey of Youth to look at the probability that other survey respondents living in the same zip code area had the same ethnic background as the respondent.4 These segregation indices are shown in Table 8.1. Some ethnic groups display high levels of segregation. The average black respondent lived in a neighborhood that was 63 percent black, and the average Mexican respondent lived in a neighborhood that was 50 percent Mexican. The typical respondent lived in a neighborhood with a much lower index: only 30 percent of the respondents had a similar ethnicity. Many other immigrant groups, including Chinese, Filipinos, other Hispanics, Polish, and particularly the older immigrant groups (the Irish and the Italians) had much lower segregation indices. These indices imply that, except for Mexicans, geographic segregation for immigrant groups is not particularly great.

Moreover, segregation weakens as the generations succeed one another. Because the 1970 census included a question about the nativity of the parents of the respondent, it permits researchers to classify people as foreign-born, native-born of foreign-born parents, and native-born of native-born parents—providing information on the first, second, and third and later immigrant generations.5 In


Although we highlight general segregation patterns for Hispanics and Asians above, there are widespread differences within nationality groups of the population. For example, Cubans are highly segregated from both blacks and whites, and Mexicans are highly segregated from blacks but only moderately segregated from whites. Asians as an overall group are highly segregated from blacks. Within the Asian population, the Japanese have the lowest segregation from whites. The Vietnamese evidence the highest segregation levels. Other Asian groups fall between the two extremes. Because the Japanese have resided in the United States for several generations and the Vietnamese are among the most recent immigrant groups, these results are consistent with a story of declining segregation with generational assimilation.


This survey was not designed to be a survey of immigrants. Rather, we cite these results as evidence that segregation varies among ethnic groups, some of whom are heavily affected by recent immigration.


Questions about parental nativity were omitted from the 1980 and 1990 censuses. Immigration researchers have argued that it would be valuable to include a question on parental nativity in the future (Edmonston, 1996).

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