Does School Segregation Facilitate the Formation of Criminal Networks?

People are hard-wired to form social networks, and an individual’s social network can play a role in shaping his or her behavior. Unfortunately, this social dynamic is as true for criminal networks as it is for more benign social networks. Among young people, if the behavior of an individual’s group affects the probability that he or she will commit a crime, it is possible that segregation in schools could lead to the formation of tighter and more active criminal networks by grouping disadvantaged youth together. In fact, neighborhood and school segregation policies tend to precede increases in local criminal activity. Thus, analyzing the role of neighborhoods and schools in the formation of criminal networks is pivotal to understanding how such networks develop and prosper.

In their 2016 paper, Stephen Billings, David Deming, and Stephen Ross examine the concentration of disadvantaged youth in schools and neighborhoods in Mecklenburg County, North Carolina, and its effect on total crimes committed in that county.

Mecklenburg County has a complicated history with segregation policies, making it a particularly interesting setting for this research study. Following a 1971 Supreme Court ruling, which held that the county’s schools were in effect segregated, Mecklenburg County prioritized school desegregation and became a successful example of racial integration. The county established a variety of programs with the goal of maintaining the county’s average race ratio – 60 percent white and 40 percent black – in each school. However, a legal battle initiated by parents in the late 1990s led to a court-ordered mandate that put a stop to the desegregation process. As of 2002, each student’s school assignment is based simply on the neighborhood in which he or she lives.

The authors identified 129 census tracts that were affected by the 2002 change in school boundaries and compared the arrest outcomes of students who live on either side of the boundaries. Individuals who commit a crime are included in the registry of adult arrests beginning at age 16 in North Carolina, which allowed the researchers to identify arrests during high school and when multiple individuals were arrested for the same crime. In addition to criminal records, the researchers also examined administrative school records to obtain information about each student’s date of birth, race, gender, standardized test scores, days absent, and days suspended.

The researchers followed 14-year-old students who attended public schools and lived within the 129 census tract areas between the 2002–2003 and 2008–2009 school years. On the basis of this sample, they found that 28 percent of students were involved in a criminal partnership, with an average of 4.26 arrests and 1.23 unique “partners in crime” per student in this group. Furthermore, these partners in crime tended to be black males of the same age.  They also were more likely than not to have relatively lower test scores and to live in low-income neighborhoods near their school peers.

This study indicates that increasing the number of school peers of the same grade, race, and gender by 8.3 students increases the probability of being arrested by 3.9 percentage points. This represents a 23 percent increase in the probability of arrest compared to the average student. Controlling for distance, the authors found that these results do not hold for distances larger than one kilometer. This suggests that the probability of being arrested increases when students who are similar across race, gender, income and grade level attend the same school and live in close physical proximity to each other.

The authors also analyzed the observed criminal partnerships to examine whether any patterns could be detected. They found that the probability of partnership is small between students who do not attend the same school. Two individuals who attend the same school, are in the same grade, and live near each other are six times more likely to form a criminal partnership than individuals who do not share these characteristics.

This research demonstrates that, for already disadvantaged students, the probability of a student committing a crime increases as the concentration of similar students living near one another and attending the same school increases. The authors also find that criminal partnerships might begin to develop at school, not only in neighborhoods in which students live. Thus, the court’s decision to separate students in Mecklenburg County based on their neighborhoods, which has indirectly increased the racial and socioeconomic segregation in a county that was once an example of successful racial integration, is problematic because the segregation appears to be increasing the crime rate among disadvantaged students. These findings indicate that school systems nationwide should consider the ramifications of districting that fails to diversify students along socioeconomic lines, even if based on residency rather than explicit segregation.

Article source: Billings, Stephen B. Deming, David J. Ross, Stephen L. “Partners in Crime: Schools, Neighborhoods and the Formation of Criminal Networks.” National Bureau of Economic Research Working Paper No. 21962 (2016).

Featured photo: cc/(maroke, photo ID: 514855272, from iStock by Getty Images)