Accuracy
Definition
Measurement accuracy describes the closeness of agreement between a measured quantity value and the true quantity value of the measurand. It is a qualitative concept and is not itself expressed as a numerical quantity.
Bias
Definition
Measurement bias is an estimate of systematic measurement error, commonly evaluated as the difference between the average observed result and an appropriate reference value.
Formula
Bias = Mean Measured Value - Reference Value
Illustrative Example
Reference Value
50.0 units
Measured Values
50.4 units
50.5 units
50.3 units
Mean Measured Value
50.4 units
Estimated Bias
+0.4 units
Interpretation
The average measured result is 0.4 units higher than the stated reference value.
Evaluation Methods
Comparison with suitable reference materials.
Comparison with calibrated reference instruments.
Analysis of control-sample results.
Review of measurement trends.
Interlaboratory comparison where appropriate.
Important Distinctions
Bias concerns systematic differences.
Precision concerns agreement among repeated results.
Accuracy concerns closeness of agreement with a true or reference value.
A measurement system may be precise but biased.
Limitations
The reference value has its own uncertainty.
Observed differences may include random variation.
Bias estimates depend on the measurement conditions and method.
Knowledge Check
| Question | Answer |
|---|---|
What does a positive bias indicate under the stated formula? | The mean measured value is higher than the reference value. |