Reliability and evaluation at Higher Level: Same skills as SL: the skills section is common to both levels. HL Paper 1B has asked students to comment on the reliability of results from ten repeats.
Practise this as you read
- Comment on reliability with a reason.
- Distinguish random from systematic error.
- Evaluate a claim: support, then limitations.
Free preview
This is the free notes preview
You're reading the free notes. Aimnova Pro unlocks the full study experience — and you can try it with your first topic free to keep:
- FlashcardsLock in vocabulary and key terms with spaced repetition.
- Practice questionsAnswer exam-style questions and get instant AI marking.
- Mock exams & past-paper vaultSit full mocks and see exactly how examiners award marks.
- Personalised study planA daily plan built around your exam date and weak areas.
The big idea: No single measurement can be fully trusted — every measurement carries some random error, and any one reading could be an anomaly (a freak result).
- Replicate
- One repeat of a measurement made under the same conditions. Several replicates let you calculate a mean.
- Reliable / repeatable
- A result is reliable if repeating the same method gives a very similar result — the repeats agree with each other.
- Outlier (anomalous result)
- A value that lies far from the other repeats. Repeats let you spot it; leave it out of the mean only if you can give a reason, such as a counting mistake.
- Random error
- Small unpredictable variation that makes repeats differ. Taking a mean of many repeats reduces its effect.
- Mean
- The average of the replicates: add the values and divide by how many there are. The single best estimate from a set of repeats.
See it drawn, step by step
In My Learning, handwritten diagrams build one step at a time. Play, pause and replay each explanation until it clicks. Create your free account to try it.
The points to remember
- Repeat each measurement, at least three times, and use the mean.
- An outlier lies far from the other repeats; say why before you leave it out.
- A small spread of repeats means the result is reliable.
- More repeats give a more reliable mean; more levels show the trend.
Replicates vs range of the variable: These are two different kinds of 'more': More replicates = measure the SAME setting several times → makes each point more reliable.
More values of the variable = test MORE settings (e.g. more temperatures) → shows the trend across the range.
A good improvement answer often needs both.
| A RELIABLE method | An UNRELIABLE method | |
|---|---|---|
| Repeats | Each measurement is REPEATED several times (replicates) | Each value is measured only ONCE |
| Spread of repeats | Repeats are CLOSE together (small spread) | Repeats are WIDELY scattered (large spread) |
| What you do with them | Calculate a MEAN; spot and re-check anomalies | No mean possible; an anomaly is impossible to detect |
| If you repeated the whole experiment | You'd get a very SIMILAR result (repeatable) | You might get a very DIFFERENT result |
| Confidence in the result | HIGH — the value is trustworthy | LOW — the value could be a one-off fluke |
Learn what examiners really want
See exactly what to write to score full marks. Our AI shows you model answers and the key phrases examiners look for.
Four words, four questions: Accurate: is the reading close to the true value? Precise: do repeated readings agree closely? Reliable: would the investigation give the same result again? Valid: does the method really test the research question?
| Term | What it asks | How you improve it |
|---|---|---|
| Reliable / repeatable | Do REPEATS of the SAME method agree with each other? | More replicates; reduce random error |
| Valid | Does the method actually TEST what it claims (fair test, control, only one variable changed)? | Add a control; control all other variables; test the right range |
| Accurate | Is a measurement CLOSE to the true value? | Calibrate / use a better instrument; reduce systematic error |
| Precise | Do repeated readings agree CLOSELY with each other? | A finer instrument; read it the same way every time |
Random and systematic error: Random error scatters readings both ways, for example judging a colour by eye. More repeats and a mean reduce it.
Systematic error shifts every reading the same way, for example a balance that was not set to zero. Repeats do not remove it: calibrate the instrument instead.
Worked example: five repeats, one outlier
Mean of all five
bubbles per minute: higher than four of the five counts.
Spot the outlier
64 lies far outside the cluster of 30 to 33, so it is the outlier.
Mean without it
bubbles per minute.
Judge the spread
Range without the outlier: 33 − 30 = 3. The four repeats agree, so 31.5 is the more reliable mean.
Leaving out an outlier: You may leave an outlier out of the mean if you can justify it: a counting mistake, a leaking tube, a contaminated plate. Say so in your answer. Never drop a value only because it is inconvenient.
How it is asked: Paper 1B and the Paper 2 data question end with judgement: suggest how to make results more reliable, explain why repeats are needed, comment on reliability, suggest a limitation, or evaluate whether the data support a claim or a hypothesis. Most parts are worth 1 or 2 marks.
A gardener grew 3 basil plants with a new plant feed and 3 without it for two weeks. Final heights (cm): fed 18, 20, 19 (mean 19); unfed 14, 22, 12 (mean 16). The gardener concludes that the feed makes basil grow taller. Evaluate this conclusion and suggest how the investigation could be improved.
Model answer plan
See the answer broken into clear points, with guidance on how each point earns marks.
Why this scores full marks: It is balanced: it first says what the data show (the means support the claim), then weighs that against the limitations (overlap/spread and tiny sample), and finishes with matched improvements (more plants for reliability, controlled variables for validity).
| Limitation in the method | Matched improvement | Why it helps |
|---|---|---|
| Only measured once at each setting | Take more REPLICATES and average them | A mean evens out random error and reveals anomalies → more reliable |
| Only a few values of the variable tested | Test MORE values across a WIDER range | Shows the full trend / locates a true optimum, not just two points |
| No control group | Add a CONTROL with the variable absent | Shows the effect is really caused by your variable, not something else → valid |
| A confounding variable could differ between groups | Keep all other variables CONSTANT (control them) | A fair test → any difference is due to your one variable |
| The measuring method is rough / subjective | Use a more PRECISE instrument / a fixed objective rule | Smaller measurement error → repeats agree better |
| Sample is small or from one place/time | Use a LARGER sample / repeat over time / more sites | More representative → the conclusion generalises better |
Comment on reliability: give the reason: 'Low' or 'reliable' alone scores nothing. Say why: few repeats or a small sample, no control group, no error bars or standard deviation to judge the spread, or many repeats with a small spread. A study on yeast or mice may not apply to humans, and an optimum may lie between the values tested.