Teacher burnout: what the data show and how it is measured

Why teacher burnout deserves separate attention

When people say "burnout in helping professions", they usually mean therapists, physicians, social workers. Teachers are a less obvious but no less significant part of that group. Several structural features of the role make teacher burnout a distinct clinical category requiring its own normative data and its own interventions.

Key role-specific factors: sustained emotional involvement with a large group of developing people (rather than a few clients per hour), poorly bounded work hours (extra-classroom work, grading, parent meetings), chronic role ambiguity (teacher / educator / emotional regulator / administrative worker), and the post-pandemic reality of a sharp rise in digital workload and emotional load from hybrid teaching.

The data: pre/post-COVID prevalence

Empirical summary. Pre-COVID baseline — meta-analysis by García-Carmona, Marín & Aguayo (2019, *Social Psychology of Education*, 49 samples, n=14,410 secondary teachers) showed 30–40% of teachers with burnout symptoms. This is the baseline against which the pandemic effect should be evaluated.

The post-COVID picture shifted. Ozamiz-Etxebarria et al. (2023, *Int J Environ Res Public Health*) in a meta-analysis of 9 cross-sectional studies across 8 countries reported a pooled prevalence of 52% (95% CI 33–71%) — this is the figure modern reviews now cite. The authors note explicitly that this level is higher than typical values for medical professionals in the same period. In parallel, Ma et al. (2022, *Work*, n=256,896 teachers across 22 countries) found 59.9% of teachers with depressive symptoms post-COVID — a related but conceptually distinct measure.

An important nuance for intellectual honesty. Westphal et al. (2022, *Frontiers in Psychology*, K-12 systematic review, n=9,874) analysed pandemic-period data and noted that teacher burnout levels are not dramatically different from other professions in the same window. The post-COVID picture rose across the helping cluster — the teacher rise is part of a general trend, not a unique anomaly. Comparison with psychotherapists (54.5%, Simionato & Simpson 2018) and US physicians (43.9%, Shanafelt et al. 2019) provides the needed context: teachers now sit in the upper part of the helping cluster, but not anomalously above it.

Specifics: how it differs from other helping professions

Structural differences between teacher burnout and other helping roles — and why this affects instrument and intervention choice.

• Long cycles: relationships with a group last a school year (vs a therapist's series of sessions with one client). Emotional exhaustion accumulates more slowly, but recovery between cycles (vacations) plays a specific regulatory role
• Large object population: 25–40 students per day vs 6–10 clients for a therapist. Depersonalization (cynical attitude toward students) is a structural risk, not "bad character"
• Dual audience: students plus parents. The emotional load from parent interactions is often underestimated but empirically one of the main triggers of exhaustion
• Role fragmentation: a teacher simultaneously is instructor, educator, diagnostician (spotting problems in students), documentation worker, and often emotional regulator for families. Not "many tasks" but competing professional identities
• Post-pandemic intensification of digital workload: online platforms, digital communication with parents outside working hours, hybrid teaching — structural loads absent from pre-COVID models

These role-specific factors mean: instruments developed for the general working population (MBI-GS) or for medical professionals (MBI-HSS) capture the teacher pattern less well. The educational setting needs a variant designed for its specific triggers — and that variant is MBI-ES.

MBI-Educators Survey — the standard for educational settings

MBI-ES (Maslach Burnout Inventory – Educators Survey, Maslach, Jackson & Schwab 1986) is the variant of the Maslach Burnout Inventory adapted for educational settings. It is the only MBI variant with normative data specific to teachers. Structurally it preserves the three-factor MBI model (emotional exhaustion, depersonalization, personal accomplishment), but item text is adapted: oriented toward students, educational situations, school context.

22 items, Likert 0–6 (frequency from "never" to "every day"). Completion 5–10 minutes. Cronbach α across subscales is consistently 0.71–0.90 in international samples (emotional exhaustion usually at the upper bound α≈0.90; depersonalization at the lower bound α≈0.71–0.76 — a historical feature of the MBI depersonalization subscale, not specific to the educator adaptation).

A core methodological principle shared with all MBI variants: the three subscales are scored separately, not summed. "Classical" teacher burnout is high EE + high DP + low PA. Partial patterns (e.g., high EE with intact PA) require different interventions — this is "overworked but engaged", a more reversible condition than full burnout.

• 22 items, Likert 0–6, ~5–10 min
• 3 subscales: emotional exhaustion, depersonalization (attitude toward students), personal accomplishment
• Cronbach α 0.71–0.90 across subscales in international samples
• Specifically adapted for the teaching population (Maslach, Jackson & Schwab 1986)
• Licensed by MindGarden Inc. — paid license for clinical and research use
• Not in the Soveria catalogue — acquired separately from MindGarden

Russian adaptations and the data gap

A Russian-language adaptation of MBI-ES exists in several working translations used in teaching communities and in school psychology services. However, a formally published PubMed-indexed psychometric validation of MBI-ES specifically on a Russian teacher sample has not been found at the time of writing. This means: clinically, the international Maslach et al. 1986 norms apply, and interpretation requires adaptation to the Russian educational context. This is a structural research gap, not a clinical failing of the instrument.

An alternative for Russian-language practice is V.V. Boyko's "Diagnostic of Emotional Burnout Level" (1996), widely used in school psychology work. Boyko provides a phasic picture (tension / resistance / exhaustion), which is especially useful in teacher work: you can distinguish a chronic-tension phase from established exhaustion and choose the corresponding intervention. Kotova et al. (2024) confirmed convergence between MBI and Boyko in a Russian sample — both instruments capture the same phenomenon through different lenses.

What scales do NOT measure and mandatory parallel screening

The most important clinical work in recent years — Schonfeld, Verkuilen & Bianchi (2019, *Psychological Assessment*, n=734 US teachers) — applied ESEM bifactor analysis to a teacher sample. The result: PHQ-9, GAD-7, CES-D-10 and MBI-EE all load similarly on a single general factor of "nonspecific psychological distress". In other words: statistically, in teachers, burnout (at least its emotional exhaustion component), depression and anxiety are barely distinguishable. This gives a direct clinical mandate for parallel screening.

This is consistent with the broader burnout literature. Bianchi, Schonfeld & Laurent (2015) argued the overlap is fundamental. Koutsimani, Montgomery & Georganta (2019) in a meta-analysis found overlap r = 0.52 while constructs remained factor-analytically distinct. Meier & Kim (2021) in a meta-regression of n=46,191 refined overlap to r = 0.49. The practical implication: a positive burnout-scale score in a teacher is not grounds for a "burnout" diagnosis, but a signal to run parallel screening for depression (PHQ-9) and anxiety (GAD-7) and to evaluate which actually dominates the clinical picture.

Differential diagnosis is a required part of the assessment. Teacher burnout scales do not distinguish burnout from: clinical depression (overlap r ≈ 0.49–0.52, Meier 2021; parallel PHQ-9 is mandatory), chronic anxiety (GAD-7 required), secondary traumatic stress (teachers working with trauma-exposed children often show secondary traumatic stress signs — not burnout, requires ProQOL), or adjustment disorder after acute life events. A positive MBI-ES is the beginning of a differential conversation, not its end.

How to use in school and individual practice

Five typical workflows for working with teacher burnout.

• School-wide annual monitoring: MBI-ES or Boyko at the start of the school year + mid-year + before summer break. Anonymized aggregate report for administration. Goal — track the school's trajectory, not "diagnose" individual teachers
• Individual work with a teacher seeking help: MBI-ES + PHQ-9 + GAD-7 as a starter battery. High MBI without depression → targeted burnout work. High MBI + depression → treat depression first; burnout addressed in parallel
• Preventive work with "second-stage" teachers (years 2–5 of practice): regular screening in the at-risk group. Meta-analytic data points to this as the peak of leaving the profession — Madigan & Kim (2021, *Teaching and Teacher Education*) showed a stable correlation between burnout and intent-to-leave in pre-COVID samples
• Teacher self-screening: Boyko or research-friendly variants of MBI-ES, with mandatory follow-up discussion with a school psychologist or external specialist. Self-screening without follow-up conversation is methodologically incorrect
• Does not replace: structural workload change (load, schedule, classes), organizational intervention (administration support, peer supervision), and medical evaluation at high scores (sleep, endocrinology)

The core principle — teacher burnout is a systemic problem, not "personal weakness" on the teacher's part. Systematic monitoring gives the school system data for structural intervention (workload, support, resource infrastructure), not grounds to blame an individual teacher. Per Schonfeld 2019, at the statistical level teacher burnout overlaps with depression and anxiety so strongly that a single MBI-ES without parallel PHQ-9 + GAD-7 screening yields an incomplete clinical picture — this is a structural methodological requirement, not an add-on.