Compare Predicted to Actual

The strategy, as published

Exactly as it appears on the card

Part of Self-Reflection & Evaluation in 06 Metacognition & Self-Regulation, where it sits alongside its neighbouring strategies, the infographic and the thinking for the whole area.

Why you might use it

The problem it may help with

Pupils are usually most confident about the material they met most recently, not the material they actually know. That miscalibration drives bad revision decisions all the way to the exam hall.

Prediction makes calibration visible. Write the score you expect, see the score you got, and the gap between the two becomes evidence a pupil can learn from, week on week.

What it might look like

One version of it, in a classroom

Before handing in a low-stakes quiz, every pupil writes a predicted score in the corner. When the quizzes come back, they log predicted against actual and note the direction they missed in. After four quizzes, one pupil can see he always overestimates on definitions and underestimates on application, which quietly rewrites his revision plan.

Try it

A way to begin

Add the predicted-score corner to your next low-stakes quiz. The comparison takes one minute when the quiz is returned, and the value builds over repetitions rather than arriving in week one.

What to notice

Reading what happens

• The direction of error, which matters more than its size. Systematic overconfidence and systematic underconfidence need different conversations.
• Whether gaps narrow over time. The card's phrase is the goal: monitoring corrected by evidence.

Things to think about

Before and while you use it

• Keep the stakes low. The moment a prediction affects anything that counts, pupils manage the prediction instead of learning from it.
• This works best attached to a regular quizzing routine that already exists, rather than as a reason to invent one.

Related strategies

Where this connects

• 🎯 Confidence Ratings Dunlosky and Metcalfe showed students are most confident about things most recently read, not things they actually know. Rate confidence after each attempt, then reveal the answer. That gap is the target. 01 Retrieval Practice · Metacognition & Calibration
• 🔬 Teach the Illusion of Knowing Show students their own illusion of knowing. Ask them to rate confidence before revealing the answer. When they were certain and wrong, that is the precise moment to intervene. 01 Retrieval Practice · Metacognition & Calibration
• 👀 Check, Don't Just Feel Winne: a novice's sense of understanding is unreliable. Pair every self-check with an external test, a quiz or worked answer, so monitoring rests on evidence rather than confidence. 06 Metacognition & Self-Regulation · Performance & Monitoring
• 🔄 Turn Reflection Into a Plan Butler: feedback only regulates learning if it changes the next action. Have students convert each reflection into one concrete adjustment they will make on the very next task. 06 Metacognition & Self-Regulation · Self-Reflection & Evaluation

The wider thinking

Where the evidence sits

Winne's work on calibration sits behind this, alongside Dunlosky and Metcalfe's finding that confidence tracks recency rather than knowledge; the strand's further reading takes both further. The card’s attribution stands as published: Winne.

The fuller picture, sources included, is in further reading and evidence for Metacognition & Self-Regulation.