2026 Advanced Teaching Roles Evaluation Report

Advanced Teaching Roles: Annual Evaluation Report cover features educators sitting around a table chatting in a large meeting room of others

Executive Summary

The Advanced Teaching Roles Program (ATR) enables local school administrative units to create innovative compensation models that allow highly effective classroom teachers to impact an increased number of students. Under North Carolina General Assembly General Statutes 115C-310.15, the North Carolina State Board of Education must evaluate ATR’s impact which may be done through a contract with an independent research organization. The North Carolina Department of Public Instruction (NCDPI) selected the Friday Institute for Educational Innovation at North Carolina State University (Friday Institute) to conduct this evaluation.

This evaluation has two primary goals: (1) helping NCDPI and Local Education Agencies (LEAs) assess the academic and instructional impact of ATR models and programs, as well as their impact on the teaching profession; and (2) examining program implementation to identify factors supporting or impeding their success. The 2026 evaluation reports on the statewide ATR implementation data collected through ATR rosters, as well as quantitative estimations of program impact (e.g., student achievement and teacher retention) at the school and teacher levels for all implementing LEAs for the 2024-2025 school year. Additionally, the report builds on prior evaluation efforts by assessing how Advanced Teaching Roles (ATR) district leaders conceptualize and define district-level success after five years of program implementation.

Program Implementation (2025-26)

To evaluate ATR implementation across 32 grant-funded LEAs, this study used a mixed-methods convergent design, collecting quantitative and qualitative data separately (Creswell & Clark, 2026). Additionally, insights from the ATR Leadership Sharing Session helped the evaluation team define and prioritize key indicators of district-level success. Key findings include:

Statewide Summary

LEAs deploy ATs primarily in elementary schools and vary in size, salary supplements, and program staffing. During the 2025-26 school year, 32 LEAs operated ATR programs across 476 schools employing 1,736 Advanced Teachers who supported 4,719 classroom teachers. Elementary schools accounted for 65% of all ATR schools. The number of ATs within a LEA ranged from one to 946, and salary supplements ranged from $2,000 to $21,000. The median salary supplement for an Adult Leadership role was $14,250, while the median salary supplement for a Classroom Excellence role was $7,300. LEAs prioritized and assigned ATR responsibilities differently: elementary schools provided broader instructional support, whereas middle and high schools focused support on specific subject areas.
Instructional support is more general in elementary schools and more subject specific in the middle and upper grades. Over half (52%) of elementary classroom teachers supported by Advanced Teachers received support in multiple core subject areas (either reported as “all subjects” or indicated as multiple subject areas). In middle and high schools, Math and English teachers received support from Advanced Teachers most often.

ATR Leadership Sharing Session

ATR district leaders engaged in a collaborative activity at a June 2026 sharing session entitled, Monitoring Progress: Measuring Success for ATR. This activity addressed the central evaluation question: How do ATR leaders define district success after five years? The research team used conventional qualitative content and descriptive analyses to identify themes and to summarize the characteristics for each category. The findings below were from 36 leaders and teachers across 22 North Carolina districts at various phases of ATR implementation. While the poster-based data collection yielded group summaries, it captured these perspectives at a single point in time rather than tracking the evolution of perceptions longitudinally. Additionally, combining participants from both active and planning districts created variance in implementation experience.

Five core outcome categories emerged: System Coherence, School Culture, Student Achievement, Teacher Efficacy, and Teacher Retention. Student Achievement and Teacher Efficacy were identified as the most more indicators for district success while System Coherence was most often mentioned. Participants identified Teacher Retention as the easiest outcome to measure.
School Culture emerged as a pervasive theme, underscoring its importance to successful ATR implementation. Participants characterized a strong school culture through collaboration, stakeholder buy-in, and embedded practices. Although participants rated School Culture as important, they also identified it as relatively difficult to measure.

ATR Leadership Sharing Session Conclusions

District leaders across participating cohorts drew the following conclusions from the Monitoring Progress: Measuring Success for ATR activity based on participant outcomes and perceived measurement feasibility:

Comprehensive framework: District leaders across participating cohorts emphasized the need for a comprehensive assessment framework to track ATR’s program health and drive continuous improvement.
Key success indicators: Participants identified five common indicators of successful implementation: system coherence, positive school culture, student achievement, teacher efficacy, and teacher retention.
Balanced measurement: Cohort members recommended a balanced mix of qualitative and quantitative metrics to capture these outcomes effectively.
Implementation Support: Because on-going evaluation demands significant resources, districts require systemized monitoring support and standardized measurement tools (i.e., survey) to track progress efficiently over time.

Program Impact (2024-25)

To help NCDPI and stakeholders understand ATR’s impact on students and teachers, the research team assessed program impact using two distinct quantitative analyses. We updated the analysis from the previous report (2023-24) by adding a new year of administrative data to compare outcomes for ATR schools versus a comparison group of similar but non-ATR schools. Then, we analyzed outcomes for students taught by Advanced Teachers (ATs) and the teachers they support (which we call Supported Teachers or STs), relative to students taught by teachers in the same school who are not part of ATR. For this second analysis, data identifying ATs and STs are available only for the 2023-24 and 2024-25 school years. Although these data can identify ATs and STs statewide, the two years of administrative data limits our ability to make causal claims about the effect of ATs. Results from our second set of analyses are correlational and should be interpreted with caution.

School-Level Impacts of ATR

To estimate schoolwide effects, we compared student outcomes in ATR schools before and after program implementation against outcomes in similar, non-ATR schools. We examined student achievement using state-mandated end-of-grade (EOG) and end-of-course (EOC) exams. To measure teacher effectiveness, we used the SAS Education Value-Added Assessment System (EVAAS) and NC Educator Effectiveness System (NCEES). Our results suggest that:

ATR schools produced positive and statistically significant schoolwide effects in ELA, math, and science. The strongest impact was observed in math (0.05 SD). This effect size is comparable to nearly a month of additional learning between third and fourth grade math. Similarly, the effect in ELA (0.02 SD) and science (0.04 SD) are also positive and statistically significant, but smaller in magnitude than in math. In comparison to our previous reports, these results are similar in magnitude but more consistently significant across different subject areas. This suggests that the additional year of data has improved our confidence in the positive effect of ATR on student test scores.
Schools that began implementing ATR in 2018-19 showed the largest positive effects on student test scores with the greatest impacts in math. While the inaugural 2017–18 cohort showed positive math and science trends that were smaller and lacked statistical significance, the strongest and most consistent gains occurred amongst schools in the second cohort that began implementing ATR in 2018–19. This 2018-19 cohort demonstrated statistically significant improvements in ELA (0.07 SD) and math (0.12 SD), alongside positive and marginally significant gains in science (0.09 SD). Recent cohorts, specifically those starting in 2023–24 and 2024–25, also exhibited consistently positive and significant score improvements across all three subjects, although they were smaller in magnitude than the 2018–19 cohort. Cohorts that began implementation during COVID (2019-20 through 2022-23) showed the greatest implementation challenges.
ATR increases test scores over time but at a higher rate in math and science than in ELA. Improvements in math yielded significant gains starting in year one, grew in the first five years, and then remained steady. Science scores demonstrated the largest growth over time reaching (0.12 SD) by year eight. Effects in science are less precise and sometimes fluctuate across years, which may be due to fewer test scores in this area compared to ELA and math. ELA results reached statistical significance after one, four, and six years of implementation, though these gains were smaller compared to math and science.
ATR effects do not vary systematically by grade level. In science, these effects were somewhat larger in elementary than in middle and high school. The results provide suggestive evidence that the ATR effect in math is largest in high school (0.08 SD). In ELA, the strongest evidence for positive results appears in middle schools (0.04 SD). Overall, we find no evidence that ATR is more or less effective for any specific grade level.
Overall, the characteristics of teachers in ATR schools have remained stable between results in last year’s report and current results, which include data up to 2024-25. Teachers hired into ATR schools tend to have higher average EVAAS scores, driven mostly by EVAAS scores in math, relative to teachers hired into comparison schools. There were no significant differences in teacher NCEES ratings or with graduate degree attainment between ATR and comparison schools. On average, teachers in ATR schools have slightly more experience and a one percentage point lower likelihood of being beginning teachers, relative to teachers in comparison schools.
There was moderate evidence of positive ATR effects on teacher effectiveness. ATR is understood as a strategy to improve the overall teacher effectiveness through a combination of enhanced in-service professional learning and talent recruitment. Increases in math EVAAS scores could be explained by different mechanisms. Mentoring from advanced teachers may have improved class instruction, while lower alternative licensure rates point to better support for teachers completing requirements. Alternatively, these gains may reflect changes in school composition, as ATR enables schools to recruit more effective teachers.
ATR schools recruited more effective and qualified teachers; however, there was no significant effect on teacher turnover. Teachers hired into ATR schools possess significantly higher prior year math and science EVAAS scores, higher NCEES (observation) scores, and lower rates of alternative licensure. On the other hand, ATR did not have a statistically significant effect on schoolwide teacher turnover, nor did ATR change the characteristics of teachers who exit these schools.

Teacher-Level Impacts of ATR

The findings below compare outcomes for students taught by Advanced Teachers (ATs) and Supported Teachers (STs) with outcomes for students taught by non-ATR teachers in the same school. Because statewide, roster-linked data identifying ATs and STs are available for only 2023-24 and 2024-25, these results provide correlational evidence and should not be interpreted as causal estimates.

Students taught by ATs have higher test scores in ELA and math compared to peers taught by non-ATR teachers in the same school. In ELA, the results are smaller in magnitude and not significant when comparing ATs to non-ATR teachers in the same school with similar experience, demographics, and licensures to ATs. Science results are not statistically significant. In science, we do not find significant differences between students taught by ATs and students taught by non-ATR teachers in the same school.
When comparing students taught by Supported Teachers (STs) to students taught by non–ATR teachers in the same school, our analysis shows non-significant estimates that fluctuate between positive and negative associations with test scores. In ELA, students taught by STs showed lower test score gains compared to matched non-ATR teachers in the same school. Math showed a positive but not statistically significant relationship when comparing against non-ATR teachers in the same school, while science outcomes were not consistent and not statistically significant.
In general, students taught by ATs demonstrated higher test score gains than students taught by their non-ATR peers. The gains were strongest and most consistent in math. ELA results were positive but less consistent across comparison models. Outcomes were generally similar for students taught by Adult Leadership (AL) and Classroom Excellence (CE) teachers in ELA and math. In science, however, outcomes differed substantially by role: students taught by ALs demonstrated gains while students taught by CEs did not.
We found mixed results when comparing student test scores across ATs with no, partial, and full release time. In ELA, the results were not statistically significant, regardless of AT release time. In math, the results for students taught by STs whose ATs had no release time were positive and significant; however, the math results for students taught by STs whose ATs had partial or full release were not significant. Full release time for ATs in science resulted in lower test scores for Supported Teachers’ students; whereas no release time was linked to higher test scores.

Recommendations

The following recommendations are intended to provide actionable steps for principals, administrators, and state leadership to promote successful strategies and address implementation issues highlighted above:

School Principals: Define support expectations and monitor model implementation. Principals should collaborate with ATs to create annual support plans that establish the focus, frequency, and format of coaching. Because quantitative findings reveal mixed results regarding release time, principals must actively track how ATs time is allocated. Furthermore, qualitative findings show that maintaining consistent implementation records (e.g., coaching logs, classroom observation data) helps gauge instructional impacts.
District Leaders: Broaden success measures and document implementation decisions. District leaders must evaluate ATR using a balanced set of indicators rather than a single measure. During the sharing session, success was viewed as multidimensional and needed to encompass system coherence, school culture, teacher efficacy, and retention. Districts are also encouraged to invest in principal readiness, and peer leadership networks, while systematically tracking staffing structures, recruitment strategies, and coaching models. This documentation will allow future program evaluations to identify what makes performing cohorts successful.
State Leadership: Align program expectations with empirical evidence and strengthen LEA guidance. State policy makers should align statutory expectations for ATR with the empirical evaluation findings, ensuring that program goals reflect realistic, demonstrated outcomes for teacher growth and student achievement. Additionally, it is important for state leaders to study how ATR recruitment impacts teacher movement across schools (i.e. drawing top talent into ATR roles from neighboring districts). Finally, state leadership should develop common measurement instruments for districts and continue investing in principal preparation and professional networks.
Institutes of Higher Education (IHEs): Align leadership preparation and research with ATR models. Higher education institutions should partner with districts to integrate ATR frameworks into principal preparation programs. Specifically, faculty must train future school leaders to manage distributed leadership, navigate coaching relationships, and supervise differentiated staffing roles. IHEs should also participate in multi-state research efforts to build a broader evidence base around advanced teaching roles.
Future Evaluation: Investigate deeper, unexamined questions. Eight years of implementation data and a growing qualitative base now enable researchers to investigate complex questions beyond statutory reporting requirements. Pursuing these deeper research questions will require dedicated resources, time, and broader data access beyond the current evaluation scope.