IMPACT & EVIDENCE · HISTORICAL EVALUATION
A notable Alabama pattern in ITMESA’s 2019 two-state analysis
ITMESA reported positive adjusted pre/post estimates for all four participant-survey constructs in the Alabama implementation led by Life That Counts. Three Alabama estimates were statistically significant as reported, and the Alabama point estimate was larger than Indiana’s in every row.
This historical evidence strengthens Life That Counts’ record of delivering structured peer mentoring in rural Alabama schools. ITMESA compared within-state estimates; it did not report a direct statistical test of the state-to-state differences.
ITMESA · High-school peer-mentoring survey analysis · January–May 2019
REPORTED ALABAMA PATTERN
Four positive Alabama estimates
Knowledge, attitude, skill, and behavioral intention all moved in the favorable direction; three within-state pre/post estimates were statistically significant as reported.
Study at a glance
Alabama · Life That Counts
434 pre-survey records · 343 post-survey records · Rural community profile
Indiana · Separate implementation
500 pre-survey records · 379 post-survey records · Urban/suburban community profile
Program period: January–May 2019 · Population: high-school mentees and mentors
Scope of this page
This page focuses on ITMESA’s January–May 2019 state table because it reports the same four constructs for both states. ITMESA’s later 2020–2021 materials involved different cohorts and delivery conditions and reported a mixed, predominantly nonsignificant pattern; those findings remain available on the main Impact & Evidence page.
What ITMESA reported
ITMESA reported state-specific, covariate-adjusted pre/post estimates for four principal constructs. A larger positive value represents a more favorable within-state estimate on the report’s 0–100 scale. These values are model estimates; they are not percentages of students who improved.
Reported 2019 adjusted pre/post estimates
State-specific model estimates in points on the report’s 0–100 scale · axis shown from -1 to 12 · ns = not statistically significant as reported
Alabama — teal
Indiana — gold
−1 | 0 ┃ 1 2 3 4 5 6 7 8 9 10 11 12
Risk-avoidance attitude
Alabama +6.6 · p=.0003
Indiana +4.6 · p=.0018
Risk-reduction knowledge
Alabama +10.9 · p=.0001
Indiana +6.9 · p=.0001
Risk-reduction skill
Alabama +4.8 · p=.0002
Indiana −0.1 · ns
Risk-avoidance behavioral intention
Alabama +3.2 · ns
Indiana +1.4 · ns
Alabama’s reported estimate was larger on all four constructs. ITMESA marked three Alabama and two Indiana within-state pre/post estimates statistically significant. The source did not report a direct statistical test of the difference between the state estimates.
Chart data
Risk-avoidance attitude | Alabama +6.6, p=.0003 | Indiana +4.6, p=.0018
Risk-reduction knowledge | Alabama +10.9, p=.0001 | Indiana +6.9, p=.0001
Risk-reduction skill | Alabama +4.8, p=.0002 | Indiana −0.1, not statistically significant
Risk-avoidance behavioral intention | Alabama +3.2, not statistically significant | Indiana +1.4, not statistically significant
A consistent reported pattern—with one result kept in view
ITMESA reported +6.6 points for Alabama risk-avoidance attitude and +4.6 for Indiana. For risk-reduction knowledge, the reported estimates were +10.9 in Alabama and +6.9 in Indiana. For risk-reduction skill, ITMESA reported +4.8 points for Alabama, statistically significant at p=.0002, and −0.1 points for Indiana, which was not statistically significant. For risk-avoidance behavioral intention, Alabama’s estimate was again larger, +3.2 compared with +1.4, but neither state estimate was statistically significant. This keeps the favorable Alabama direction visible without suggesting that every estimate met the same statistical threshold.
REPORTED PATTERN
Alabama’s adjusted estimate was larger across all four principal constructs. Its within-state pre/post estimates for knowledge, attitude, and skill were statistically significant as reported by ITMESA.
CURRENT LIFE THAT COUNTS PROGRAM DESIGN—NOT AN ITMESA FINDING
Why the model is built around peers
Life That Counts’ current peer-mentoring design is informed by social cognitive theory and social norms theory, applied through social-norming strategies. Together, these frameworks recognize that young people learn not only from information, but also from what they observe, practice, and believe their peers expect.
Program-design source: Life That Counts, BELIEVE Project Narrative (2026), Conceptual Framework and “Why Peer Mentoring Matters.”
Observational learning
Theory: Social cognitive theory. Prepared peer mentors can model constructive decision-making, future-focused thinking, and appropriate help-seeking.
Practice and self-efficacy
Theory: Social cognitive theory. Guided discussion, rehearsal, and reinforcement can help mentees build confidence to navigate pressure, relationships, and future goals.
Healthy norms made visible
Theory and practice: Social norms theory and social-norming strategies. Structured peer mentoring can make responsible choices, supportive peer expectations, and asking trusted adults for help more visible.
Structure and adult oversight
Delivery practice: Supervised peer mentoring. Mentor preparation, defined boundaries, repeated sessions, adult supervision, and clear escalation pathways turn the theory into responsible school-based practice.
ITMESA measured reported pre/post differences in attitude, knowledge, skill, and behavioral intention. The analysis did not test whether social cognitive or social-norm mechanisms caused or mediated the findings, and it did not isolate peer mentoring from the broader mentor-coach implementation.
What this historical analysis adds
Structured school-based peer mentoring involves prepared mentors, defined roles, repeated program contact, adult supervision, responsible data collection, and attention to local school context. ITMESA analyzed the 2019 Life That Counts Alabama implementation and reported favorable adjusted pre/post estimates for knowledge, attitude, and skill. The evidence does not settle every research question. It contributes a credible historical foundation for continuing work in peer mentoring, program refinement, and rigorous evaluation.
Designed for the school setting — structured sessions, defined roles, and coordination with school adults
Built on peer influence — modeling, practice, reinforcement, and healthier perceived norms
Examined through evaluation — all four principal constructs remain visible, including the nonsignificant behavioral-intention estimate
How to read the findings
Research source: ITMESA · Principal researcher: Harry Piotrowski · Period represented: January–May 2019 · Population: High-school mentees and mentors · Evidence type: Adjusted pre/post participant-survey associations · Analysis described by the source: Multilevel repeated-measures linear regression adjusted for gender, race, Hispanic ethnicity, age, mentor/mentee role, and state.
Precise source citation: ITMESA reporting materials assembled September 2021: Table 1 (survey records and community context) and Table 2 (2019 adjusted pre/post estimates and p-values); program period January–May 2019.
The supplied materials do not report outcome-specific analytic sample sizes, confidence intervals, standardized effect sizes, complete participant-matching and missing-data procedures, adjustment for multiple outcome testing, an untreated comparison group, or a direct statistical test of the difference between the Alabama and Indiana estimates. The analysis reports favorable historical pre/post associations. It does not estimate a causal program-versus-no-program effect or a statistically tested difference between states.
Review and download the public evidence brief
The 3 vs 2 count summarizes separate within-state significance results as reported by ITMESA. It is not a statistical test of the difference between states.
Life That Counts public evidence summary · Prepared August 2026 from ITMESA reporting materials
Download comparison brief (PNG)
Download infographic (PNG)
Related pathways
Build a responsible peer-mentoring approach for your school
Life That Counts helps schools think through mentor selection, preparation, adult supervision, program boundaries, delivery, and follow-up. Tell us what students currently experience and what you hope peer mentoring can strengthen.