How to measure professional development effectiveness.

By Yechiel · Founder, PD That Works. Yechiel builds PD That Works — the workshops, the daily AI coach, and the Day 30/60/90 reports — and wrote every page on this site.

Every PD purchase ends with the same question from a board or a superintendent: did it work? Most schools answer with a sign-in sheet and a satisfaction survey, which is a bit like measuring a diet by whether people enjoyed the cookbook. Here's what those instruments actually measure, what the research field measures instead, and what evidence a school can realistically produce at Day 30, 60, and 90.

What most schools measure — and what it tells you

Attendance. Who was in the room. It's data about the workshop, not about teaching, and it is the single most common "evidence" in a PD file.

The satisfaction survey. How the day felt: was the presenter engaging, was the room comfortable, would you recommend it. Useful for choosing next year's presenter; silent on whether anything changed on Tuesday.

PD hours. How much time was logged, often for a certification requirement. A count of inputs.

None of these is wrong to collect. All three measure the event. The question the board is asking is about the classroom, and the event ends before the classroom begins.

The standard framework, and where evaluation usually stops

The field has had a working framework for a quarter century: Thomas Guskey's five levels of professional-development evaluation — participants' reactions, participants' learning, organization support and change, participants' use of new knowledge and skills, and student learning outcomes (Educational Leadership, 2002). The levels build on each other: you can't get practice change without learning, and you can't get student outcomes without practice change.

Read honestly, the satisfaction survey is Level 1 and the sign-in sheet isn't on the ladder at all. Most PD evaluation stops there. The level that answers "did it work" for a teaching purchase is Level 4 — did teachers use the new knowledge and skills — and Level 5, student outcomes, is the destination that takes longer than any single cycle to reach and is entangled with everything else a school does. A measurement plan that skips Levels 3 and 4 and reaches for test scores usually finds nothing, and concludes wrongly that the PD failed.

What to measure instead: four signals

Here's the evidence a daily follow-up system can actually produce, what each is and isn't, and where it sits on the ladder.

Engagement. The honest number: how many days teachers actually showed up to the follow-up — not perfection, pattern. On the sample dashboard, that's the week-by-week line and the "needs your attention" list of teachers who've gone quiet. It measures whether the support was used, which is the precondition for everything below it. Isn't: proof of change. Level: the bridge from 3 to 4.

Practice change, over cycles. Each teacher sets one goal in their own words, rates their confidence on it at Day 1, and the rating moves — or doesn't — across the cycle. It's a self-rating from teachers, on a focus area the school chose. It's not a test score. It's not a classroom observation. It's the same kind of evidence a coach would gather in a sit-down — captured every day, in five minutes, and added up. Isn't: an external observation rubric. Level: 4.

Themes. What five or more teachers' reflections have in common this week — "getting student-to-student talk going," "tightening the first five minutes." Below five, a theme stays blank on purpose; individual reflections are never shown to leaders. Themes tell you what the faculty is actually working on, which is Level 4 in the aggregate. Isn't: anyone's words.

Wins. Moments teachers chose to share, by name or anonymously. Qualitative, self-selected, and the only place in the evidence where a leader hears a teacher's own voice. Isn't: representative. Level: 4, in one teacher's classroom at a time.

Put together, the four answer the board's question at the level it can honestly be answered in a semester: teachers used the support, their practice moved against goals they set, here is what they were working on, and here is what it felt like when it worked.

A measurement plan that fits a school year

Day 1 — the baseline. Every teacher leaves the workshop with one goal in their own words and a first confidence rating on it. Without a baseline, nothing later is movement.

Days 2–29 — the daily record. Five minutes a day. Engagement accumulates; the goal sharpens ("I want more voices in discussion" becomes "two student voices before mine, in every discussion").

Day 30 — the first report. Engagement, confidence growth, themes, wins, and recommended next steps for the cohort, in a PDF built for a board meeting. See the sample school report. The teacher gets a private letter of her own the same day — a fictional sample shows the shape.

Day 60 — the comparison. The same report, against Day 30. Now there's a trend, not a snapshot.

Day 90 — the full picture. Three cycles. This is the document that makes the renewal case without anyone having to write it.

When you take this to a board meeting or a superintendent, the chart isn't decoration. It's the answer to "did practice change."

The honest limits

• Self-ratings are self-ratings. They measure how confident a teacher feels about a move, not an observer's judgment of it. They're the right instrument for daily capture and the wrong one to call an evaluation.
• The privacy floor hides small groups. Themes need five teachers. A three-person department produces engagement and growth data but no themes, by design.
• Student outcomes take longer than a cycle. Level 5 is real and worth pursuing, but a 30-day report that claimed test-score gains would be lying. Real change in a school takes years; this is the daily work that compounds into it.

A leader's measurement checklist

1. Is there a baseline — each teacher's goal and starting point, in their own words — before the PD begins?
2. Does the evidence cover use in practice (Guskey's Level 4), not just reactions to the event?
3. Is there a record over cycles — 30, 60, 90 — rather than a single survey?
4. Are teachers' individual words protected, so the record is honest?
5. Would the Day 30 artifact survive a board meeting — and does someone have to write it by hand?

Common questions

How do you measure the effectiveness of professional development?

Measure use in practice, not reactions to the event: a baseline goal for each teacher, engagement with follow-up across the cycle, movement on the goal by Day 30, 60, and 90, and the themes a faculty is actually working on. Attendance and satisfaction surveys measure the workshop; effectiveness shows up in the classroom weeks later.

Are satisfaction surveys a good measure of PD effectiveness?

They're a good measure of the day and a poor measure of effectiveness. In Guskey's five-level model a satisfaction survey is Level 1, participants' reactions; whether teachers used what they learned is Level 4. A workshop can score well on Level 1 and change nothing.

What is Guskey's model for evaluating professional development?

Five levels that build on each other: participants' reactions, participants' learning, organization support and change, participants' use of new knowledge and skills, and student learning outcomes. Most schools stop at Level 1 or 2; the question "did it work" is answered at Level 4 and, over years, Level 5.

How long does it take to see whether PD worked?

Practice change is measurable within one 30-day cycle if there's a baseline and daily follow-up; a trend needs Day 60; a defensible picture needs Day 90. Student-outcome effects take longer than a cycle and are entangled with everything else a school does, so a report claiming them at Day 30 should be treated with suspicion.