“Check your work” sounds simple until a student asks what checking actually means. Without a strategy, learners often reread the same answer in the same way and unsurprisingly see what they expected to see.

Checking is a second task

Producing an answer and evaluating an answer require different attention. Immediately after solving, the original reasoning remains highly available, making it easy to repeat the same assumptions during the check.

Check high-risk features

Use error history to decide where checking has the highest value. A Mathematics student may repeatedly lose signs or units. A writer may omit required task details. A Science student may state facts without causal links. A comprehension student may answer without sufficient evidence.

Change the checking method

Do not merely reread. Estimate a numerical answer independently. Substitute a result back into an equation. Compare the response with the command word. Trace every claim to evidence. Read writing specifically for one error class at a time.

Separate accuracy from perfection

Students have limited examination time. Checking should target likely, consequential errors rather than become an anxious attempt to redo the entire paper. Efficient checking uses risk and mark value.

Build checking into practice

A routine introduced only before the examination is unlikely to become automatic under pressure. Require the same targeted checks during homework and timed practice, then measure whether the relevant error category decreases.

Teach students to distrust familiarity

An answer can look right because it is the answer the student just produced. Where possible, check through an independent route. Different evidence is more useful than repeating the original thought process.

The repair

Turn “check your work” into specific, risk-based procedures. Use error history, alternative verification methods and repeated practice until checking becomes a reliable performance skill. Good checking is not looking again. It is looking differently.