March 17, 2016: Exploring Educational Segregation in New Jersey's Counties

In this post, we conclude our exploration into diversity of educational attainment to understand if our counties are segregated in terms of the educational attainment of their adult population. Segregation is a product of the relationship between a county’s overall diversity and its neighborhood diversity. Therefore, to understand the level of segregation a county exhibits, we have examined the difference between every county’s two Diversity Index Scores. To conduct this analysis, we calculated an Integration/Segregation Index Score for each county. This final metric in our analysis is determined by subtracting a county’s Countywide Diversity Index Score from its Average Neighborhood Diversity Index Score.

This calculation assumes that a county is perfectly integrated if its neighborhood diversity matches its overall diversity, meaning a perfectly integrated county would receive an Integration/Segregation Index Score of zero (0.00). This scenario of absolute integration is not seen in any New Jersey counties. However, while no New Jersey county is perfectly integrated by this measure, we can still compare the index scores of counties to determine which ones are especially segregated compared to their peers. The closer to zero a county’s Integration/Segregation Index Score is (i.e. the smaller the difference between its neighborhood and countywide diversity), the less segregated is the county, while scores farther from zero indicate more segregation.

The scatterplot below helps us visualize and interpret the results of our integration/segregation analysis. In the scatterplot, we can see an important trend emerging in the relationship between a county’s overall diversity and its neighborhood diversity. Rarely do counties that score highly in their overall diversity exhibit the same level of diversity in their average neighborhood. Often in this analysis, the higher a county’s Countywide Diversity Index Score, the larger the distance is between the county’s location on the scatterplot and the 45-degree line of absolute integration. This trend leads us to suggest that counties that are more diverse in the educational attainment of their adult population are usually the most segregated as well. This can be seen on the scatterplot by observing how far the counties that score highest on the Average Neighborhood Diversity Index (i.e. are closest to the top of the scatterplot) fall from the line of absolute integration. 

While Atlantic County exhibits high scores in both countywide and neighborhood diversity, this fact does not make it the most integrated county. Consistent with the trend discussed above, the difference between Atlantic County’s two diversity index scores is greater than counties that are less diverse overall. This trend is seen again when comparing the Integration/Segregation Index Scores of Mercer and Essex counties, which are the two most segregated counties in the state and which have relatively high countywide diversity, with the scores of Hunterdon and Sussex counties, which are the two most integrated counties but which also have lower countywide diversity.

What are some factors that contribute to such a great difference in countywide and neighborhood diversity in Mercer and Essex counties? The universities and colleges in these counties actually contribute more to the educational diversity of neighborhoods in which they are located. Neighborhoods surrounding Princeton University and The College of New Jersey in Mercer County and Rutgers-Newark and NJIT in Essex County are home to a mixture of current students, recent graduates, and even faculty, making for more diverse neighborhoods that are not dominated by one group. The reason these counties are so segregated has more to do with the affluent and the distressed neighborhoods that are found throughout and the homogenous nature of the populations found within both types of neighborhoods. It is the presence of numerous neighborhoods of both affluence and distress within the same county that indicate its high segregation, and vice versa. Therefore, it is both correct to say that these counties have lots of pockets of affluence and distress because they are so segregated, and to say that these counties are likely to be highly segregated, given the mere presence of numerous pockets of both affluence and distress found within.   

The presence of affluent or distressed neighborhoods, and the homogenous and segregated nature of their populations, is not a naturally occurring phenomenon. Rather, the nature of the educational attainment of residents (and many other demographic and socioeconomic characteristics) in these neighborhoods has much to do with county and local land use and development practices. The types of jobs and housing that are located within a neighborhood have a great impact on the type of residents who live within. Neighborhoods that are dominated by particular occupations are also likely to be dominated by particular housing types reflective of the income earned by residents employed in the dominant occupation. In order for neighborhoods to be more educationally integrated, they need to be home to jobs that are available to students, highly educated professionals, and working-class residents, while also being home to housing options reflective of the income earned by individuals employed in all these types of occupations.

Use the interactive map and chart feature below to uncover more information on the level of segregation seen in each county. Happy exploring!

Author: John Manieri, AICP

Research, Analysis, and Technical Assistance: Steve Scott


Data Sources: 

U.S. Census Bureau, 2010-2014 5-year American Community Survey. Table S1501