end of header

Housing Patterns

Chapter 5

Residential Segregation of Blacks or African Americans: 1980 to 2000

The number of African Americans in the United States grew over the last few decades from 26.5 million in 1980, to 30.0 million in 1990, and to 36.4 million in 2000.(1) Blacks comprised 11.7 percent of the total U.S. population in 1980, 12.1 percent in 1990, and 12.9 percent in 2000. About 86.5 percent of Blacks lived in metropolitan areas in 2000.

Table 5-1 shows the extent of residential segregation of Blacks in 1980, 1990, and 2000. There were 220 metropolitan areas (of the 330 total) with 3 percent or 20,000 or more Blacks in 1980. All five measures of segregation indicate a reduction in the residential segregation of Blacks between 1980 and 1990, and a further reduction between 1990 and 2000. The two-decade reduction ranges from 4 percent (absolute centralization and spatial proximity) to 12 percent (dissimilarity), regardless of whether all metropolitan areas or just "selected" metropolitan areas are examined. All indexes declined by one percent or more in each decade (the threshold for a change that we consider substantive as described in Chapter 2). This decline demonstrates a clear trend toward lower residential segregation for Blacks.(2)

The overall reduction in residential segregation is illustrated in Figures 5-1a through 5-1e. These figures demonstrate the clear shift in most of the index distributions toward less residential segregation for the dissimilarity, isolation, and delta indexes, although this was less pronounced for the absolute centralization index and unclear for spatial proximity index.

This reduction seemed to take place throughout the distribution of segregation, but with different indexes showing different patterns. Dissimilarity, isolation, and delta indexes showed a greater decline in areas of higher segregation, while changes in absolute centralization and spatial proximity were perhaps more uniform. The percentage declines in each decade were similar for each index: sometimes the 1980-1990 change was larger than the 1990-2000 change, and sometimes the reverse was true.

The largest metropolitan areas (1 million or more population) had higher residential segregation than the middle-sized ones (500,000 to 999,999 population), which, in turn, had higher residential segregation than the smallest metropolitan areas (see Table 5-2). This was true for all indexes for all 3 years, but for several indexes, the difference between small and medium metropolitan areas was small. The 1980-1990 and 1990-2000 reductions in the residential segregation of Blacks took place in all regions for all five indexes (with the exception of the spatial proximity index for the Northeast), and for metropolitan areas of different sizes for four of the five indexes.(3) In 2000, the West region had the lowest level of residential segregation for three of the five indexes, and the South was lowest for the remaining two. The Midwest had the highest level of residential segregation for four of the five indexes; the Northeast had the highest level for the remaining one.

Residential segregation varied by the percentage (expressed in quartiles) of the population that is Black. While all four metropolitan area quartiles showed a pattern of decreasing residential segregation over time, three of the five indexes showed a pattern of higher segregation in places with a higher percentage of Blacks in 2000, while two showed the reverse. As the percentage of the population that is Black increased, Blacks were

The relationship between segregation and quartiles of percent change in the African American population does not show a clear pattern. For example, metropolitan areas with both the largest and smallest percent increases in the African American population experienced significant decreases in dissimilarity, isolation, delta, and spatial proximity.

Figures 5-2a through 5-2e show that a majority of all of the selected metropolitan areas declined in residential segregation between 1980 and 2000, though most only had a small change (are clustered near the 45-degree line).(4) Table 5-3 shows the percentage of metropolitan areas experiencing change in segregation scores (in five ranges). The proportion of metropolitan areas with increases of 1 percent or more between 1980 and 2000 ranged from only 3 percent (dissimilarity) to 34 percent (spatial proximity). However, the proportion with decreases of 1 percent or more between 1980 and 2000 ranged from 43 percent (spatial proximity) to 92 percent (dissimilarity). Thus, the most widely used index, dissimilarity, showed that only 8 of 220 metropolitan areas had an increase in residential segregation between 1980 and 2000, while 203 metropolitan areas had a decrease. The others indicated a much less uniform pattern but still tended to show a decline in segregation.

Table 5-4 presents the level of residential segregation for the 43 large metropolitan areas with 1 million or more population in 1980 and at least 3 percent or 20,000 or more Blacks. In terms of the most commonly used residential segregation index, dissimilarity, the five most segregated metropolitan areas for Blacks were, in order, Detroit, Milwaukee-Waukesha, New York, Newark, and Chicago (Newark at 0.801, is not substantively higher than Chicago, 0.797, but both are higher than number six -- Cleveland-Lorain-Elyria, at 0.768).

When the other four indexes are taken into account and the ranks averaged across the five indexes, the five most segregated metropolitan areas for Blacks in 2000 were, in order, Milwaukee-Waukesha, Detroit, Cleveland-Lorain-Elyria, St. Louis, and Newark (Milwaukee- Waukesha and Detroit are less than one average rank apart). Cincinnati, Buffalo-Niagara Falls, and New York, are roughly tied for number six, but each is more than one average rank behind Newark. The top ten are rounded out by Chicago and Philadelphia (the latter roughly tied with Kansas City, New Orleans, and Indianapolis). Figure 5-3 (961k) PDF shows the settlement pattern of Blacks in 2000 in Milwaukee-Waukesha.

Averaging the ranks across the five indexes, the most segregated areas in 2000 were also the most segregated in 1990, and among the six most segregated in 1980 (Kansas City comes in at number 5 in 1980). In 1990, Milwaukee-Waukesha was the most segregated, followed by Detroit, and in 1980, Detroit was followed by St. Louis.

The five least segregated metropolitan areas for Blacks among the large ones analyzed here were, in order using just the dissimilarity index: Orange County, San Jose, Phoenix-Mesa, Riverside-San Bernardino, and Norfolk-Virginia Beach-Newport (which is substantively similar to Portland-Vancouver). When using all five indexes averaged, the five least segregated metropolitan areas for Blacks were, in order: Orange County, San Jose, Norfolk-Virginia Beach- Newport News, Tampa-St. Petersburg-Clearwater, and San Diego (the latter two are roughly tied, and tied with Providence-Fall River-Warwick). Figure 5-4 (1.1M) PDF shows the settlement pattern of Blacks in Orange County in 2000.

The top ten most segregated large metropolitan areas were in the older Northeast-Midwest "Rust Belt," which has tended to lose population in recent decades. All but one of the least segregated large metropolitan areas were in the West and South, where metropolitan areas have tended to gain population. The exception was Providence-Fall River-Warwick in the Northeast, which tied for the fifth-least-segregated and does not fit the broader pattern.

Also of interest is how segregation has been changing for these large metropolitan areas. Table 5-5 presents these results for the 1980 to 2000 period. Of the 43 large metropolitan areas, 40 showed a decline in residential segregation using the dissimilarity index between 1980 and 1990, while the other three showed virtually no change. This was also true for the 1990 to 2000 period. Combined, all large metropolitan areas showed a decline in the residential segregation of Blacks and African Americans between 1980 and 2000, but some of the changes are not substantively significant.

The metropolitan areas showing the largest percentage declines (averaging ranks across the five indexes) were, in order: Los Angeles-Long Beach, Oakland, Seattle-Bellevue-Everett, Tampa-St. Petersburg-Clearwater, and Orange County (the last two were among the five least segregated large metropolitan areas in 2000). The five large metropolitan areas showing the least percentage declines were, in order: Rochester, New York, Riverside-San Bernardino, Pittsburgh, and New Orleans. No large metropolitan area showed an increase in dissimilarity between 1980 and 2000, 10 showed an increase in isolation, 1 in delta, 3 in absolute centralization, and 14 in spatial proximity.

When we examined all selected metropolitan areas (not shown), the five most residentially segregated for Blacks in 2000 were the five large areas already noted (using the averaging over five ranks method) and the five least segregated were Orange County MSA CA, Hickory-Morganton-Lenoir MSA NC, Fort Walton Beach MSA FL, Charlottesville MSA VA, and Auburn-Opelika MSA AL, all in the South or West; the 10 least segregated were all in the South except for Orange County and Hamilton-Middletown.

Based on the ranks of all selected metropolitan areas, the five metropolitan areas showing the greatest increase in residential segregation over the two-decade 1980-2000 period were Dover MSA DE, Columbus MSA GA-AL, Goldsboro MSA NC, Athens MSA GA, and Danville MSA VA. The seven metropolitan areas showing the greatest increases were all in the South region (the 8th was in Michigan and the 9th in New Jersey). The five metropolitan areas showing the greatest decrease in residential segregation over the 1980-2000 period were all in Florida: Melbourne-Titusville-Palm Bay, Daytona Beach, Fort Myers-Cape Coral, Fort Pierce-Port St Lucie, and Sarasota-Bradenton. Of the next five largest declines, four were in Texas, and another in Florida.

In conclusion, it is clear that the decline in the residential segregation of African Americans in the 1980-1990 period continued apace over the 1990-2000 period. Most strides seemed to have been made in the West and South, particularly in California, Florida, and Texas, although increases in segregation were apparent for some small metropolitan areas in the South. Less progress was made in the Northeast and Midwest, and the large metropolitan areas that had been the most segregated at the beginning of the period remained at or near the top of the list.

Yet, only 8 of the 220 metropolitan areas examined in this chapter showed an increase in the dissimilarity index of residential segregation for Blacks of 1 percent or more, and 203 showed a decline of 1 percent or more -- indicating widespread reductions in residential segregation between 1980 and 2000. The reduction of African American residential segregation remained slow, but steady.

[PDF] or PDF denotes a file in Adobe’s Portable Document Format. To view the file, you will need the Adobe® Reader® Off Site available free from Adobe.


1. The 2000 figure includes all people who identified as Black or African American alone or in combination with another race. The number of people who identified as Black or African American alone was 34.7 million.

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.

3. The absolute centralization index increased slightly for metropolitan areas under 1 million between 1980 and 1990. The index decreased back to its 1980 value in 2000 for areas with 500,000-999,999 and decreased from its 1990 value for small metropolitan areas but not fully back to the 1980 level.

4. These figures are presented for 1980 versus 1990 and 1990 versus 2000 in Appendix D.

Chapter 5:
Go to Chapter 6: Residential Segregation of Hispanics or Latinos
Go to "Racial and Ethnic Residential Segregation in the United States: 1980-2000" in HTML format
Contact the Census Call Center Staff at 1-800-923-8282 (toll free) or visit ask.census.gov for further information on Housing Patterns Data.

Source: U.S. Census Bureau, Housing and Household Economic Statistics Division
Last Revised: October 31, 2011

This symbol Off Site indicates a link to a non-government web site. Our linking to these sites does not constitute an endorsement of any products, services or the information found on them. Once you link to another site you are subject to the policies of the new site.
  Is this page helpful?
Thumbs Up Image Yes    Thumbs Down Image No
No, thanks
255 characters remaining
Thank you for your feedback.
Comments or suggestions?