Diversity and segregation of the child population

The Boston region has historically had only modest minority populations.  Table 1 shows that as recently as 1990 nearly 90% of the population was non-Hispanic white.  Blacks and Hispanics each accounted for close to 5% of the total, and Asians were 2.7%.  During the 1990s, as the region grew slowly, the white population actually declined by about 60,000, while minorities grew at a rapid rate.  Hence whites now are only 81.2% of the total.  Hispanics are the largest minority with 6.6%, followed by blacks (5.9%) and Asians (4.7%).

The decline in the white population stems partly from the fact that this is an older group, and Table 1 shows that the under-18 population is substantially less white than the total.  By 2000 whites were only 75% of the children in the region, while Hispanics were almost 10%, blacks 8.1% and Asians 5.2%.  So there is now considerable racial and ethnic diversity among children in the region.

This diversity in the region as a whole is not very well reflected at the level of neighborhoods.  Instead, there is a strong tendency for children to live separately from those of other backgrounds.  We measure this tendency in Table 2 in two ways, based on data for census tracts (geographic areas that typically have 3000-4000 residents). Researchers often use both measures together, because they each tell us about a distinct aspect of people’s neighborhoods: 

• The traditional and most commonly used measure of segregation is the Index of Dissimilarity (D), which measures the “unevenness of the distribution?of children of different races across neighborhoods.  If all neighborhoods had the same racial composition, D would have a value of 0.  In a situation of complete apartheid (for example, where neighborhoods were either all white or all black) D would have a value of 100.  Based on many studies of different racial and ethnic groups, researchers generally consider values below 30 to represent a low level of segregation (typical, for example, of separation between different white ethnic groups in most cities).  Values between 30 and 50 or 55 are interpreted as moderate segregation.   Higher values are considered extreme levels of segregation.  This index isn’t affected by the relative size of the white or black or other populations.  Whatever the diversity of the population, it measures whether they are similarly distributed across neighborhoods.