WASHINGTON - Urban Indian segregation has decreased between six percent and 11 percent over the last two decades, according to the broadest measure in a new study by the U.S. Census Bureau. But in a couple of narrower indices the Bureau tracks, the separation actually rose. And, if measured by those who designated themselves only as Natives, and not as mixed race including Native, segregation actually increased even by the broadest measure.
The Bureau measured residential segregation between 1980 and 2000 in the country's metropolitan statistical areas (MSAs). It found an 11 percent drop in American Indian and Alaska Native "dissimilarity" in all MSAs during that time, and a smaller, six percent drop in 13 MSAs with substantial Native populations. It did not track non-metro areas, which would include most reservations.
Dissimilarity, according to the bureau, is a measure of evenness, which compares "the spatial distributions of different groups among units in a metropolitan area. Segregation is smallest when majority and minority populations are evenly distributed."
The 13 MSAs used by the census study of Natives are Albuquerque, N.M.; Tulsa, Okla.; Anchorage, Alaska; Rapid City, S.D.; Fort Smith, Ark.-Okla.; Lawton, Okla.; Great Falls, Mont.; Yakima, Wash.; Bellingham, Wash.; Yuma, Ariz.; Oklahoma City, Okla.; Phoenix-Mesa, Ariz. and Los Angeles-Long Beach, Calif.
Twelve of these showed declines in dissimilarity over the two decades, with only one showing an increase towards more segregation.
The four Oklahoma MSAs showed less segregation than the other nine in both decades studied. The Bureau noted that the 13 snapshot MSAs only have 12.7 percent of the total Native population and only 19.4 percent of urban Indians.
The bureau's "spatial proximity" index, which measures clustering, also showed less segregation for urban Indians, by 10 percent for all MSAs and 15 percent for the 13 MSAs. Clustering is defined as "the extent to which units inhabited by minority members adjoin one another, or cluster, in space."
Urban Indians showed more segregation, however, for the index measuring isolation, and mixed results for ones called delta (concentration) and absolute centralization.
For isolation, meaning the probability a minority member shares an area with another minority person rather than a majority one, the Census Bureau index for urban Indians increased from 0.177 in 1980 to 0.188 in 1990 and 0.205 in 2000 for the 13 selected MSAs. The increases mean an increase in segregation (with the white population as a reference group). For all MSAs, the values increased from 0.082 to 0.102 to 0.103.
Results were mixed on the other two measures. Delta, or concentration, decreased for urban Natives in all MSAs but increased in the 13 selected ones. Delta "computes the proportion of (minority) members residing in units with above average density of minority members," according to the study.
The last category, absolute centralization, measures "the distribution of the minority group around the center" or most densely populated area of the city. Here again, urban Indians in all MSAs showed an increase in segregation, while those in the 13 MSAs with higher concentrations of Native people showed a decrease.
The authors of the report, John Iceland, Daniel H. Weinberg, and Erika Steinmetz, did some cross-minority group analysis and concluded that urban Natives generally are less segregated than urban African-Americans, Hispanics and Asians.
Only one MSA, Los Angeles-Long Beach, had enough members of each minority group to allow them to be broken out separately. In Los Angeles, the same descending order of segregation applied: blacks, Hispanics, Asians and Natives.
As of the 2000 Census, Native Hawaiians and Pacific Islanders are now their own group for Census purposes. In previous Census counts, they were lumped in with Asians. Since their levels of segregation have been measured only once, there is no earlier national Pacific Islander study against which to measure them, as there will be ten years from now.
However, they are included in the individual MSAs, and can be measured against other ethnic groups. So in the Los Angeles area, where 49,514 Native Hawaiians and Pacific Islanders lived in 2000, the Bureau measured a dissimilarity of 0.506. That would make them slightly more segregated than the 136,696 Indians or Alaska Natives in the MSA, who measured 0.474, and Asians (0.479), but less segregated than blacks (0.664) and Hispanics (0.631).
Also for the first time in 2000, the Census Bureau added a mixed race category. The segregation study included anyone who indicated any American Indian/Native Alaskan/Native Hawaiian ancestry. The figures for just those who termed themselves fully Native found "modestly larger" segregation levels.
The Census Bureau studied only the fact of residential segregation and not the reasons for it, so it presented no statistics and advanced no conclusions about housing discrimination.