The Mirage: Confronting the Hard Truth about Our Quest for Teacher Development

10,507 teachers and 566 school leaders surveyed, plus 127 interviews, across three large districts and one midsize charter network

Teachers in three large, geographically diverse US public school districts (together >20,000 teachers, ~400,000 students, 69% low-income) plus one midsize charter management organisation. Districts are anonymised.

Practitioner report by an advocacy organisation, not a peer-reviewed study and not a causal design - grade D is right. Method, in TNTP's own words: "our method was not to implement a particular development strategy and then track its results. Instead, we identified teachers whose performance appeared to improve substantially and worked backward to find any experiences, mindsets or environments they had in common." That is retrospective correlational contrast on observational data - 2 to 4 years of district evaluation ratings, observation scores and value-added, joined to a self-report survey of development time. Its value is the denominator (what is actually spent and delivered), not an effect estimate. Two caveats the numbers need. First, the $18,000 figure is the MID-RANGE of a three-tier sliding-scale estimate that TNTP constructed itself, and 77-87% of it is teacher and staff TIME and salaries rather than professional-development line items - it is an imputed opportunity cost, not a budget line. Second, `independence: developer-involved` is set from the report's own disclosure page: "The districts studied for this report are among the more than 60 school systems with which TNTP is currently engaged as a consultant and/or service provider." TNTP is itself a teacher-training provider (50,000+ teachers recruited or trained) publishing on the effectiveness of teacher development, and its recommendations advance its standing policy platform. The report states no district held editorial control and that it was independently funded. Advocacy-organisation report with no DOI; ERIC ED558206.

The cost side of the professional-development question, plus a null pattern. Districts spend ~$18k per teacher per year (mostly imputed staff time) and teachers give ~19 days a year, and no variety, dose or combination of development separates the 19-30% of teachers who improved from those who did not. Read as an audit of the spend, not as causal evidence that development cannot work - TNTP is explicit that it did not test any strategy and that it opposes cutting the investment. The honest counterweight, which the headline usually drops: one of the four systems studied (the charter network) did produce substantial growth in ~7 of 10 teachers, at nearly double the per-teacher spend - and even there, TNTP could not identify which development activities caused it.

Not applicable - a teacher-level descriptive audit, not a student-outcome study.

Descriptive and not designed to be replicable; districts are anonymised, so the spending and improver figures cannot be independently recomputed. The "nothing distinguishes improvers" claim is correlational and rests on comparing self-reported development time against district evaluation data. The report itself points at the relevant experimental literature - two federally funded experimental studies of sustained, content-focused, job-embedded PD that found no lasting significant change in teacher practice or student outcomes.