"Did it work?" is a fair question to ask of any money a school spends, and incursions rarely get asked it properly. The default measure, whether the students were happy, tells you almost nothing about learning, because students can be delighted by an afternoon that taught them very little. Measuring real impact is not hard, but it requires deciding in advance what you are looking for and capturing it deliberately. Done well, it gives you something far more useful than a warm feeling: defensible evidence that survives a sceptical budget conversation.
Decide the outcome before you book
You cannot measure an impact you never defined. Before the session, name the specific thing students should be able to do or understand afterwards: explain why a heavier object does not fall faster, set up a fair test independently, use the word "evaporate" correctly in context. That named outcome becomes your yardstick, and everything you measure afterwards is measured against it.
Without a defined outcome, you are left judging the day by how loud the enjoyment was, which is not the same thing as learning and is easily faked by spectacle. The discipline of naming the outcome up front also improves the session itself, because it forces you to choose what the incursion is for and to prime students toward noticing the things that matter. Measurement and design are linked: deciding what success looks like shapes how you set the whole thing up.
Use a light before-and-after
The simplest honest measure is a quick check of the same thing before and after the session. A one-question entry slip, "draw what you think happens to water when it boils", followed by the same question a week later, shows movement you can actually see across a class. It takes five minutes at each end and produces evidence considerably more convincing than any survey of how students felt about the day.
You are not running a controlled study, and you do not need to. You are looking for a visible shift in understanding across the class, and a rough before-and-after captures it well enough to be genuinely useful. The before-slip also does double duty as priming, surfacing what students already think so the session can build on it or challenge it. A small amount of structure at each end turns an experience into something you can demonstrate worked.
Watch for unprompted vocabulary
A reliable signal that a concept has landed is students using the correct terms without being asked, in contexts you did not set up. When "friction" turns up unbidden in a student's account of why the playground slide felt slow on a particular day, the concept has generalised beyond the lesson it was taught in, which is exactly what you want. Listening for spontaneous, accurate vocabulary in the fortnight after a session is one of the most honest impact measures available.
It also costs nothing but attention. Unlike a formal test, you are not setting anything up, you are simply noticing when the language of the session reappears in unprompted student talk. Vocabulary that students reach for on their own, correctly, in unrelated moments is strong evidence of durable understanding, far stronger than the same words produced to order on a worksheet completed the same afternoon.
Track the usually-disengaged students specifically
Whole-class averages can hide the most important effect of all. Often the real value of a hands-on session is in reaching the students who had checked out of the written unit, and that value is invisible in a class-wide tally where their gains are averaged in with everyone else's. Pick two or three of those students before the session and watch them specifically: did they participate, contribute, and retain in a way they had not been?
Their response is frequently where the strongest case for the program lives. A program that lifts the engaged students slightly is worth less than one that re-engages the students who had given up, because re-engagement opens a door that was closing. Tracking named individuals, rather than only the class mean, surfaces this effect and gives you the most compelling evidence you can take to anyone questioning the spend.
Feed it into your existing reporting
Impact evidence is most useful when gathering it does not create extra work. A photograph of student work, a recorded explanation, a kept prediction slip: these slot into the assessment records you already maintain and double as evidence of curriculum coverage. Measured this way, a science program is not a cost you have to justify with vibes, it is a contribution to the assessment file you were building anyway.
This is the framing that wins budget conversations. "The science day was fun" is easy to cut. "The science program produced documented evidence against three curriculum outcomes and re-engaged four students who had disengaged from written science" is not, because cutting it now means losing something concrete and visible rather than just a treat. Tie the measurement to your existing reporting and you turn a soft expense into a defensible part of how you deliver and document the curriculum.
