All studies

Education policy / Ohio · 8 districts · AP Research · April 2026

Balancing education and discipline in school drug policy

I scored the drug policies of eight Ohio school districts against 20 evidence-based tenets, then compared them with each district’s drug-related incident rate. Stronger policies tracked with fewer incidents, and districts whose discipline far outweighed their education did worse.

Policy vs. incidents, district by district

Switch the measure, then select a district point or use the menu to open its scorecard.

District scorecard: 20 policy tenets

Each tenet was scored 1 (absent) to 5 (explicit, with a detailed implementation plan). Every district scored higher on discipline than on education.

Education tenets

Discipline tenets

What if a district rebalanced?

Move the sliders to set total education and discipline scores (10–50 each). The fitted Poisson model estimates the incident rate a district with that profile would expect.

Illustrative only: the model was fitted on eight districts and is correlational, so it cannot show what would happen if a district actually changed its policy.

Inside the study

Question · Data · Methods · Findings · Limits
Question

To what extent does the balance between comprehensive education-based prevention and thorough disciplinary measures affect drug-abuse-related disciplinary incident rates in Ohio high schools?

Why it matters

Research on school drug prevention has mostly studied education and discipline separately, and it relies on student surveys rather than on the policies themselves. This study scores both kinds of policy side by side and links them to recorded incidents.

Data

From December 2025 through March 2026 I requested policy documents and aggregate, de-identified incident counts from Ohio districts, starting with contacts from my deputy superintendent and then working outward by distance from my district’s central office. Eight districts took part, with enrollments from 1,800 to 8,500. An incident is any formally documented drug, alcohol or nicotine offense in the category districts already report to the state under the Ohio Revised Code.

Methods

I scored each district on 10 education tenets (Faggiano 2014; Griffin & Botvin 2010; Durlak & DuPre 2008; Pettigrew 2013) and 10 discipline tenets (Evans-Whipp 2007; Patte 2017; Hemphill 2006), using anchored 1–5 Likert scales. Every policy was scored twice, at least two weeks apart, and any differences were reconciled a week later. Strength is EPS + DPS, and imbalance is |EPS − DPS|. A Poisson regression with enrollment as the exposure models incidents as log(μ) = β₀ + β₁·Strength + β₂·Imbalance.

Findings

Total policy strength was strongly negatively correlated with incident rate (r = −0.937), and imbalance was moderately positively correlated with it (r = 0.593). In the regression, each additional strength point was associated with about 2.7% fewer incidents (β₁ = −0.0267, IRR 0.974, p = 0.012), and each additional imbalance point with about 5% more (β₂ = 0.0483, IRR 1.050, p = 0.008). District #7 had the lowest imbalance (7) and the lowest rate (3.0 per 1,000). District #2 had the highest strength but also the greatest disciplinary tilt, and a 37% higher rate.

Limitations

The sample is eight districts, chosen by convenience and proximity, so the estimates are preliminary and confidence intervals are wide. A single rater scored every policy. Incident counts only capture cases that were caught. Socioeconomic and demographic confounders were not modeled, so these results are correlations, not causal effects.

Future research

Repeat the study with a larger, statewide or national sample, add inter-rater reliability checks, and follow districts over time as their policies change to test whether the relationships are causal.

District data

Policy scores and incident counts as reported in the paper. District names are withheld.