Explanation of segregation measures

There are many ways to think about and measure segregation. Here we focus on two standard types of measures. One is the Index of Dissimilarity (D). It measures how differently two groups are spread across neighborhoods. The other is a measure of group isolation or intergroup contact, reporting what the racial composition of the average group member’s neighborhood is. If it measures the share of neighbors who are of the same group (e.g., the share of a white person’s neighborhoods who are white), it is called the Isolation Index. If it measures the share of neighbors who are of some other specific group (e.g., the share of a white person’s neighbors who are Hispanic), it is called an Exposure Index.

Index of Dissimilarity

The standard measure of segregation is the Index of Dissimilarity (D), which captures the degree to which two groups are evenly spread among census tracts in a given city. Evenness is defined with respect to the racial composition of the city as a whole. With values ranging from 0 to 100, D gives the percentage of one group who would have to move to achieve an even residential pattern - one where every tract replicates the group composition of the city. A value of 60 or above is considered very high. For example, a D score of 60 for black-white segregation means that 60% of either group must move to a different tract for the two groups to become equally distributed. Values of 30 to 60 are usually considered moderate levels of segregation, while values of 30 or less are considered low.

Demographers typically interpret change either up or down in the following way:

• Change of 10 points and above in one decade - Very significant change
• Change of 5-10 points in one decade - Moderate change
• Below 5 points in one decade - Small change or no real change at all

Change can be cumulative, and small changes in a single decade – if they are repeated over several decades – can constitute a significant trend. Therefore we pay attention not only to what has happened since 2000 but also to the longer term trajectory for each group.

For smaller metropolitan regions, or for groups with small populations in a given metro, readers should interpret results with caution. In this report, the average values are weighted by the size of the group in a given year, and consequently less reliable measures for small areas introduce little error in these averages.

Exposure and Isolation Indices
Another widely used measure of segregation is a class of Exposure Indices (P*) that refers to the racial/ethnic composition of a tract where the average member of a given group lives. Exposure of a group to itself is called the Index of Isolation, while exposure of one group to other groups is called the Index of Exposure. Both range from 0 to 100. For example, an Isolation score of 80.2 for whites means that the average white lives in a neighborhood that is 80.2% white. An Exposure score of 6.7 for white-black exposure indicates that the average white lives in a neighborhood that is 6.7% black.

Even if segregation (measured by the Index of Dissimilarity) remains the same over time, growth in a minority population will tend to leave it more isolated - that is, leaving group members in neighborhoods where they are a larger share of the population. But at the same time the minority group’s growth also tends to increase the exposure of non-Hispanic whites to that minority population. These are common phenomena in recent years when the white share of the typical metropolis is declining. Even if there were no change in the distribution of whites and minorities across census tracts (which is what we measure with D), there could be change in each one’s exposure to the other (measured by P*).