Science skills

Accuracy vs precisionWhy a precise result can still be wrong

In everyday speech, accurate and precise mean much the same thing. In science they mean different things, and exam questions test the difference all the time. A set of results can be very precise and still be wrong.

  • Years 7 to 12
  • Physics, Chemistry, Biology, Science
  • Updated November 2026

The short answer

Accuracy

How close is the result to the true value?

Poor accuracy usually comes from a systematic error, such as a balance that isn’t zeroed.

Precision

How close are repeat measurements to each other?

Poor precision comes from random errors, which make the readings scatter.

Resolution

What is the smallest change the instrument can show?

A balance that reads to 0.01 g has a resolution of 0.01 g. It is not the same thing as precision.

The four possibilities

Each dot is one measurement. The centre is the true value.

Accurate and precise

Precise, not accurate

Accurate on average, not precise

Neither

Precise results sit close together. Accurate results sit close to the centre. Precision on its own doesn’t mean the result is right.

The second target is the one students forget. Precise results look convincing because they agree so well, but if something pushes every reading the same way, they agree on the wrong answer.

The same idea with real numbers

Four groups measure the acceleration due to gravity, g. The accepted value is 9.80 m/s².

Accepted value: 9.80 m/s²

Group A

  1. 9.79
  2. 9.81
  3. 9.80

Accurate and precise

Close together and close to 9.80.

Group B

  1. 9.51
  2. 9.52
  3. 9.50

Precise, not accurate

Very close together, but all about 0.3 m/s² too low. Something in the method is pushing every result down.

Group C

  1. 9.5
  2. 10.1
  3. 9.8

Accurate on average, not precise

The mean is 9.8, but the results are spread over 0.6 m/s². Any one of them on its own could be well off.

Group D

  1. 8.9
  2. 10.6
  3. 9.3

Neither

Widely spread, and the mean (9.6) is still off.

Precision is not the same as resolution

Resolution is the smallest change an instrument can display. A digital stopwatch that shows 0.01 s has a resolution of 0.01 s. That doesn’t make your timing precise to 0.01 s.

When you time something by hand, your reaction time at the start and the stop is around 0.2 s, and it varies each time. Repeat readings will spread by much more than 0.01 s. The extra digits on the screen don’t add any real precision.

How to improve each one

To improve accuracyTo improve precision
Main cause of the problemSystematic errorsRandom errors
What to doRemove the systematic error: zero or calibrate the instrument, read scales at eye level, reduce heat lossReduce the scatter: better technique, an instrument with finer resolution, timing many swings instead of one
Does repeating help?No. Every repeat has the same error.Repeating shows the spread, and the mean is a better estimate than one reading.

Our systematic and random errors guide goes through the common errors one by one.

How precision relates to reliability

In NSW science, you will hear both words. Precision describes how closely a set of measurements agree. Reliability describes whether the whole experiment gives consistent results when it is repeated. Precise measurements are a big part of reliable results. Our validity, reliability and accuracy guide explains reliability in full.

Practice: which is it?

  1. Boiling water at sea level is measured as 100.1, 99.9 and 100.0 °C. (True value: 100.0 °C.)

  2. The density of aluminium is measured as 2.95, 2.96 and 2.95 g/cm³. (Accepted value: 2.70 g/cm³.)

  3. g is measured as 9.3, 10.4 and 9.7 m/s². (Accepted value: 9.80 m/s².)

  4. The speed of sound is measured as 301, 362 and 280 m/s. (Accepted value at 20 °C: 343 m/s.)

  5. A ruler with a worn end measures a 12.0 cm strip as 12.3, 12.3 and 12.3 cm.

Common mistakes

Thinking more decimal places means more accurate.

Instead: Extra digits only show resolution. Accuracy is about closeness to the true value.

Saying precise results must be correct.

Instead: Precise results can all share the same systematic error.

Using “accurate” when you mean “precise”.

Instead: If you are talking about the spread of repeats, the word is precise (or reliable).

Judging accuracy without a true value.

Instead: Accuracy needs something to compare against. Without it, you can only comment on precision.

Exam-style questions

  1. Distinguish between accuracy and precision, using an example. 2 marks

    Show a model answer

    Accuracy is how close a measurement is to the true value. Precision is how close repeated measurements are to each other. For example, a balance that reads 0.15 g too high gives precise readings that are all inaccurate.

  2. A student measures a desk by moving a 30 cm ruler along it four times, and gets 120.4, 119.6 and 120.9 cm. Using a tape measure, she gets 120.2, 120.3 and 120.2 cm. Which set of results is more precise? Can you tell which is more accurate? Explain. 3 marks

    Show a model answer

    The tape measure results are more precise: their range is 0.1 cm, compared with 1.3 cm for the ruler. Moving the ruler four times adds a small error at each step, which varies from trial to trial. You can’t tell which is more accurate without knowing the true length of the desk.

  3. Explain why a digital stopwatch that reads to 0.01 s does not give times that are precise to 0.01 s when an experiment is timed by hand. 2 marks

    Show a model answer

    Human reaction time when starting and stopping the watch is about 0.2 s and varies each time. This random error is much larger than 0.01 s, so repeat readings will vary by far more than the stopwatch’s resolution.

Questions students and parents ask

Can something be precise but not accurate?

Yes. This happens when a systematic error affects every reading in the same way. The readings agree closely with each other but not with the true value.

Is precision the same as resolution?

No. Resolution is the smallest change an instrument can display. Precision is how closely repeat measurements agree, which also depends on the method and the person using the instrument.

Does averaging improve precision?

Averaging doesn’t make the individual readings any less scattered. But the mean of several readings is a better estimate of the true value than any single reading, as long as there is no systematic error.

Which matters more, accuracy or precision?

Both matter. Accuracy is the goal, but you can only trust an accurate result if it is also precise. A single accurate result from scattered data may just be luck.

Keep going