Index of Dissimilarity
The index of dissimilarity allows us to measure the degree to which two population characteristics are distributed differently within a metropolitan area. We might use the measure to see the degree to which the African-American population is segregated from the rest of the population, or the degree to which foreign-born people are distributed differently from native-born people. Different metropolitan areas might be compared with one another, or we could see how the index of dissimilarity has changed in a single area through time.
In the hypothetical example shown above there are three spatial units. In a real metropolitan area we would be working with many hundreds of block groups or other spatial units. Variable x is the percentage of the total population of x that is found in each area. Variable y is the percentage of the total population of y that is located in each area. The third column is the difference between the two percentages in each area (all positive numbers). To calculate the index we take the sum of the differences in the third column (70) and divide by two. The index in our hypothothetical example therefore has a value of 35.
For a real metropolitan area we could use census data found on the LandView III compact disk distributed by the Bureau of the Census (CD-TGR95-LV3-1). We could export data for block groups in the metropolitan area to an Excel spreadsheet. Then we might save the columns that deal with the subject in which we are interested, and use the Excel spreadsheet to calculate an index of dissimilarity.
Go Back to List of Topics
Go to SRU Main Page