Direct comparison
Accuracy vs precision — what is the difference?
Accuracy and precision are distinct dimensions of measurement quality. Accuracy is how close measurements are to the true value; precision is how close repeated measurements are to one another. A measurement can be precise but inaccurate, accurate but imprecise, both, or neither — they are independent qualities.
Side-by-side comparison
| Dimension | Accuracy | Precision |
|---|---|---|
| What it measures | Closeness to the true value | Closeness of repeated measurements to each other |
| Question answered | "Is it right?" | "Is it consistent?" |
| Related error | Affected by systematic error (bias) | Affected by random error (variability) |
| Target analogy | Shots centred on the bullseye | Shots clustered tightly together |
| Needs a reference | Yes — requires a known true or accepted value | No — judged from the spread of repeats alone |
| Improved by | Calibration; removing systematic bias | Better instruments; reducing random variation |
| Can have one without other | Accurate but imprecise: scattered around truth | Precise but inaccurate: tight cluster, wrong place |
| Quantified by | Error or bias relative to the true value | Standard deviation or repeatability of repeats |
| Example | A scale that reads the correct weight on average | A scale that gives the same reading every time |
Common questions
FAQ
Can a measurement be precise but not accurate?+
Yes — this is a common and important case. Precise measurements cluster tightly together but may all sit far from the true value because of a systematic error, such as an uncalibrated instrument. Tight agreement among repeats says nothing about whether they are correct.
Which matters more, accuracy or precision?+
Both matter, and a trustworthy measurement needs each. Without accuracy your results are biased; without precision they are unreliable and hard to reproduce. The good news is that a precise-but-inaccurate system can often be corrected by calibration, whereas pure random scatter is harder to remove.
How do systematic and random error map onto these?+
Systematic error (bias) shifts results consistently in one direction and damages accuracy. Random error scatters results unpredictably around their average and damages precision. Identifying which type dominates tells you whether to recalibrate (for bias) or to refine the method and instruments (for variability).
Going deeper
