MYP Physics: Evaluating Experiments & Error

Structure your answer in three parts: identify a specific limitation (not "human error"), explain how it affected your specific results using physics reasoning, then propose a realistic improvement. Vague answers like "be more careful" score zero — MYP Criterion C rewards named sources of error linked directly to your data.

A systematic error shifts every reading the same way — a wrongly zeroed balance adds the same offset each time — and averaging repeats won't fix it. A random error scatters readings above and below the true value from unpredictable factors like reaction time on a stopwatch, and repeats reduce its effect.

"Evaluate" is an IB command term meaning you make a judgement backed by evidence — not just describe what happened. For a method evaluation, weigh its strengths and weaknesses, say whether it was fit for purpose, and back every claim with a specific detail from your own procedure or data.

Percentage error compares your result to an accepted value using: percentage error = (|experimental value − accepted value| / accepted value) × 100. It tells you how far off your result was, but on its own it doesn't explain why — you still need to link the number to a specific source of error.

Accuracy means how close your result is to the accepted or true value; precision means how close your repeated readings are to each other. You can be precise but inaccurate — five tightly clustered readings that are all wrong because of a systematic error, like a miscalibrated scale.

A good improvement names the exact piece of equipment or step you'd change, explains the specific error it removes, and stays realistic for a school lab. "Use a data logger with a light gate instead of a stopwatch to remove reaction-time error in timing the trolley" earns marks; "do it more carefully" doesn't.

An anomaly is a result that doesn't fit the trend of your other data — always identify it by comparing it to a graph or table, never by guessing. Explain it with a plausible physical cause specific to that trial, such as a dropped stopwatch reading or a slipped ruler, then say whether you excluded it and why.

Criterion C — Processing and evaluating — is marked out of 8 across two strands in the current MYP guide: interpreting data and processing/evaluating validity. At the highest achievement level (7–8), you must evaluate the validity of the method and explain its impact on results with specific, well-reasoned detail, not general comment.

"Evaluate" needs a judgement with evidence for and against. "Explain" needs reasons and mechanisms, not just description. "Justify" needs you to give valid reasons for a decision, like why you excluded an anomaly. Mixing these up — describing when asked to evaluate — is the single most common mark loss.

Length isn't the marker's concern — specificity is. A tight 3–4 sentence evaluation naming one real error, quantifying its effect, and giving one realistic improvement will outscore a full page of general comment about "trying to be more accurate next time."

Cover three things: how valid your results were against known values or trends, which specific errors (systematic and random) affected your data and how, and one realistic, technically-justified improvement per major error. Reference your actual data — a graph, a table, a specific outlier — throughout.

It's not conceptually hard — it's a writing-precision problem. Most students lose marks not from misunderstanding error but from writing generically instead of specifically. Once a student learns to name, quantify, and justify, this becomes one of the more reliably scorable parts of the criterion, not the hardest.

The fastest gains come from practising the evaluation section in isolation, not redoing whole experiments. Give your child a set of results and ask them to write just the error/improvement paragraph, then compare it against a mark scheme — this trains the specific-language habit examiners reward without needing a full lab each time.

Look for resources with worked, annotated example evaluations against the actual MYP Criterion C strands — not generic "scientific method" explainers. Revision Notes that model full high-mark answers, paired with topical practice questions, help far more than reading about error types in the abstract.