When we began our journey in 2003 to dismantle our zero tolerance disciplinary system in our schools, an interesting thing happened along the way.
The crime rates went down, but not until we changed the way we disciplined kids in school.
I am not sure how many folks similarly situated in juvenile justice understand that how we treat kids in our schools is one of the most essential factors in reducing crime among juveniles, and later in reducing crime among the adult population.
As go kids, so go adults.
Our graduation rates were at an abysmal 58 percent when we embarked on our journey. Since creating our school-justice partnership program, our rates have increased every single year. This significant hike in graduation rates parallels a significant decrease in school arrests and suspensions.
Our school-justice partnership approach has resulted in a 91 percent decrease in school referrals to the court while simultaneously the overall graduation rates increased 24 percent from 2003-15. Since changing in 2012 how we report these rates from overall to on-time rates, our rates have continued to climb another 16 percent.
The correlation between school arrests as an independent variable and graduation rates as a dependent variable should not be underestimated. A regression analysis done on these two variables shows a strong statistical relationship between school referrals and graduation rates.
This is also true for suspension rates.
When undertaking reform in your system, you should guard against unintended consequences, and this requires an understanding of how differing systems interact with shared populations. When a change occurs in one place, there is a risk that a shift may occur in another place that will create problems.
This brings me to another important requirement: that data must be collected along the decision-making continuum.
For example, by removing school arrests as an option in the schematics of school discipline, how does it impact the other remaining disciplinary options? We did not plan for this, but we did collect data on out-of-school suspensions after we implemented the protocol to reduce school arrests. We discovered that these kids were shifted to the suspension option despite providing an educational alternative for these kids.
We were still battling the punitive disciplinary culture.
This brings me to another best practice in school-justice partnerships, sustainability using written interagency agreements. We would not have identified this problem shifting had we not placed our protocol in writing to include a continuing partnership. That group meets to review data to determine if the protocol was being implemented effectively, to watch out for unintended and harmful consequences and to provide oversight.
This conversation about our population being shifted from one poor response to another did two things: 1) it identified that the good news of decreasing school referrals by 54 percent within six months of implementing our partnership was not so good if we replaced it with another harsh option, and 2) it provided the opportunity to discuss how to solve the problem.
[Related: Strong Education Programs, Supports Can Be Potent for Justice-Involved Youth]
This led to the school system changing its code of conduct so that the minor school offenses that were no longer subject to referral to court were also removed from the list of suspensions. This was to prevent administrators from using out-of-school suspensions to replace arrests. We also introduced Positive Behavioral Interventions and Supports (PBIS) as a tool for educators in the classroom. These changes have reduced the suspension rate to the levels they were in 1994 before zero tolerance policies were introduced and wreaked havoc on our graduation rates along with school arrests.
When including suspensions along with school referrals as independent variables in a multiple regression analysis, we found what we hypothesized in the beginning: Keeping kids in school and out of courts will significantly improve graduation rates.
But this brings me to the one thing we didn’t consider — the influence of graduation rates on crime as the new dependent variable.
Despite those who deny a “school to prison pipeline” exists, the research shows that school arrests double the likelihood a kid will not graduate and quadruple the likelihood if the kid appears in court. I am surprised that these “deniers” do not connect this statistic to the extremely high number of inmates in adult prisons who dropped out of school (70 percent in Georgia).
If that doesn’t grab you, consider that only 30.73 percent of Georgia kids with 15 or more days absent from school (including suspensions) graduate from school. The more we suspend and arrest kids on campus, the more likely they will drop out of school and enter our adult prisons.
A regression analysis of multiple variables including the reduction in school arrests and suspensions and the increase in graduation rates revealed a statistically significant relationship between these variables and a reduction in juvenile crime.
We hypothesized early on that as go kids, so go adults: Reducing juvenile delinquency will eventually influence the adult crime rate. Thus we looked at the crime index rates before and after enacting the protocol. What we found suggests there is a relationship between this model and growing kids into a healthier adult population with reduced risks for criminality.
Our crime rate index for the baseline year (2004) was 57 per every 1,000. The year we enacted the protocol, our crime index rate was 57 crimes for every 1,000 persons. By 2015, the rate fell to 48 for every 1,000. We also found that in the years of zero tolerance preceding our protocol, the crime index rate increased into the 60s per 1,000 suggesting a statistical relevance between zero tolerance policies and the increase in crime rates.
If such a correlation exists between crime rates and graduation rates, it would follow that the more kids we graduate, the fewer crimes. If true, this would suggest that over time, as more kids graduate each year, as in my county, and become economically mobile, the crime rate increase will slow and eventually take a downward turn.
I opine that a community that pursues a school-justice partnership with fidelity will not only improve the academic success of kids, but will improve the quality of life for the entire community in a matter of years.
It is my hope that one day our understanding of what really works to reduce crime will be so ingrained in our culture of justice that it will become intuitive for us, and no longer a challenge to do what is right for kids, which we consider counterintuitive today.
Steven Teske is chief judge of the Juvenile Court of Clayton County, Georgia, and vice chairman of the Governor's Office For Children and Families. He is a past president of the Council of Juvenile Court Judges and has been appointed by the governor to the Children & Youth Coordinating Council, Department of Juvenile Justice Judicial Advisory Council, Commission on Family Violence and the Governor's Office for Children and Families.
More related articles:
Racial Disparities Persist Even as School Suspensions Decrease, Federal Data Shows
GAO Calls for Reducing Barriers to College for Foster and Homeless Youth
Resilience: Our Nation’s Great Social Justice Issue for Kids