Census 2000 results showed that Hispanics or Latinos (hereafter referred to as Hispanics) already are or are about to become the largest minority group in the United States. About 14.6 million people identified as Hispanic in 1980 (6.4 percent of the total population); this number grew by over 50 percent to 22.4 million people in 1990 (9.0 percent of the total population), and to 35.3 million people (12.5 percent of the total) in 2000.(1)
Table 6-1 shows the residential segregation indexes for Hispanics for 1980, 1990, and 2000.(2) The dissimilarity, isolation and spatial proximity indexes showed an increase in segregation between 1980 and 2000, and the delta and absolute centralization indexes showed a decrease in segregation.
This lack of a consistent pattern is illustrated by Figure 6-1(a-e), except that the isolation index showed an increase throughout its distribution (rightward shift), as did, to a lesser extent, the spatial proximity index. An examination of Figure 6-2b shows that only two of the 123 selected metropolitan areas had a decrease in the isolation of Hispanics between 1980 and 2000.
For regions, this mixed pattern persisted for the Northeast and Midwest (see Table 6-2). However, four of the five indexes indicated a decline in the residential segregation of Hispanics in Southern metropolitan areas between 1980 and 2000, while four of the five indexes indicated an increase in residential segregation in Western metropolitan areas over the same period.
While the picture was also mixed for metropolitan areas of 1 million or more and areas of under 500,000 people, for medium-sized metropolitan areas (500,000-999,999) three of the five indexes increased, and the other two showed no change. Despite these increases, the medium-sized areas tended to have lower levels of segregation than areas of larger or smaller size.
The highest level of residential segregation among Hispanics was in areas with the highest percentage of Hispanics, in some cases substantially higher segregation (note the isolation index particularly). In 2000, the dissimilarity index was 10 percent higher in areas where the population was 17.5 percent Hispanic or more (highest quartile) than in areas that were under 3.9 percent Hispanic (lowest quartile). When we compared the same group of metropolitan areas for the other indexes, the same pattern emerged -- the isolation index was 147 percent higher, the delta index was 3 percent higher, the absolute centralization index was 1 percent higher, and the spatial proximity index was 11 percent higher. With a few minor exceptions, the increase in segregation was monotonic from the under 3.9 percent category to the 3.9 to 7.3 percent category, to the 7.3 to 17.5 percent category, and to the highest quartile.
Metropolitan areas with the largest increases (over 213.9 percent) in Hispanic or Latino population between 1980 and 2000 generally experienced larger increases in segregation than metropolitan areas with relatively small increases in the Hispanic or Latino population. The highest quartile was also the only one which experienced increases in all five dimensions of segregation over the 20-year period. In contrast, metropolitan areas with the smallest increases in the Hispanic or Latino population experienced decreases in three of the five indexes and increases in the other two.
Table 6-3 gives the distribution of percent change in each index by decade. This table also confirms the findings described earlier. The isolation and spatial proximity indexes increased over the 1980-2000 period, the absolute centralization index showed a possible decrease, and the other two indexes did not change much, on average.
Table 6-4 displays 36 large (1 million or more) metropolitan areas with 3 percent, or 20,000 or more, Hispanics in 1980. In terms of the most commonly used residential segregation index -- dissimilarity -- the five most segregated metropolitan areas for Hispanics were, in order: Providence-Fall River-Warwick, New York, Newark, Hartford, and Los Angeles-Long Beach. When the other four indexes are taken into account, and the ranks averaged across the five indexes, the five most segregated metropolitan areas for Hispanics in 2000 were, in order, New York, Providence-Fall River-Warwick, Phoenix-Mesa, Los Angeles-Long Beach, and Chicago and Newark (tied). The top ten were rounded out by Denver, Riverside-San Bernardino and Houston (tied), and five others that were roughly tied for tenth. Figure 6-3 (2.3M) presents the settlement patterns for Hispanics in New York in 2000.
While New York has been the most segregated large metropolitan area for Hispanics since 1980, Providence has risen from 27th of 36 in 1980 to 10th in 1990, and to 2nd most segregated in 2000. Miami moved in the other direction, from 3rd most segregated in 1980, to 4th in 1990, and to 17th in 2000.
The five least segregated metropolitan areas for Hispanics, based on the dissimilarity index, were, in order: St. Louis, Seattle-Bellevue-Everett, Fort Lauderdale, Portland-Vancouver, and Baltimore. Using all five indexes averaged, the five least segregated metropolitan areas for Hispanics were, in order: Baltimore; St. Louis; Fort Lauderdale; Nassau-Suffolk; and Tampa-St. Petersburg-Clearwater (Detroit is roughly tied with Nassau-Suffolk and Tampa-St. Petersburg). Figure 6-4 (1.6M) presents the settlement pattern for Hispanics in Baltimore in 2000.
Table 6-5 presents the changes by decade in the 1980 and 2000 period for these large metropolitan areas. Those showing the largest percentage declines (or smallest increases) in residential segregation of Hispanics over the 1980-2000 period (averaging ranks across the five indexes) were, in order: San Antonio, Tampa-St. Petersburg-Clearwater, Miami, Hartford, and St. Louis. The five large metropolitan areas showing the smallest percentage declines (largest increases) were, in order: Providence-Fall River-Warwick, Atlanta, Fort Lauderdale and Portland-Vancouver (tied), and Washington.
Not only was Providence-Fall River-Warwick one of the most segregated large metropolitan areas for Hispanics in 2000, it was also the metropolitan area with the largest percentage increase over the 1980-2000 period when all 123 selected areas (areas with 20,000 or 3 percent or more Hispanics in 1980) were considered. The other four with the largest increases were Fort Lauderdale, Las Vegas, Santa Rosa, and Richland-Kennewick-Pasco. In contrast to Providence-Fall River-Warwick, Fort Lauderdale was one of the least segregated large areas in 2000.
Of the five metropolitan areas showing the largest percentage decrease, based on the ranks of all selected metropolitan areas in residential segregation over the 1980-2000 period, four were in Texas: El Paso; Odessa-Midland; San Antonio; Jersey City, NJ; and Laredo (tied). Of the next five, another was in North Carolina (Fayettville), another two were also in the South (Miami FL and Galveston-Texas City TX), and two were in the Midwest (Gary IN and Saginaw-Bay City-Midland MI).
Overall, the residential segregation picture for Hispanics in the United States is mixed, with increases slightly outnumbering declines when all measures are considered. There was some slight evidence of declines in segregation in the South, but increases for medium-sized metropolitan areas, and increases in metropolitan areas with large percentages of Hispanics. While New York continued to be the most segregated large metropolitan area for Hispanics, as it had been for two decades, several areas showed significant changes -- Providence-Fall River-Warwick, for example, became much more segregated and Miami much less so.
1. The ethnicity question was moved ahead of the race question on the 2000 Census because research showed it reduced nonresponse to this item.
2. Using the approach described in Chapter 2 to determine substantive changes as one percent of the index range over three years, the following critical values are used: dissimilarity, 0.006; isolation, 0.008; delta, 0.005; absolute centralization, 0.010; spatial proximity, 0.009.