Related Experiment Videos
"Precision" and "accuracy": two terms that are neither
David L Streiner1, Geoffrey R Norman
1Baycrest Centre for Geriatric Care, Department of Psychiatry, Kunin-Lunenfeld Applied Research Unit, University of Toronto, 3560 Bathurst Street, Toronto, Ontario M6A 2E1, Canada. dstreiner@klaru-baycrest.on.ca
Journal of Clinical Epidemiology
|March 22, 2006
Summary
The terms "precision" and "accuracy" are often misused as synonyms for "reliability" and "validity." This article clarifies that these terms are not interchangeable, advocating for the traditional terminology in scale development.
Area of Science:
- Psychometrics
- Measurement Theory
- Scale Development
Background:
- The terms "precision" and "accuracy" are frequently conflated with "reliability" and "validity" in scientific literature.
- This imprecise usage can lead to misunderstandings in the interpretation of measurement scales.
Purpose of the Study:
- To critically evaluate the interchangeable use of "precision" and "accuracy" with "reliability" and "validity."
- To differentiate the distinct meanings of these psychometric concepts.
- To advocate for the appropriate use of established terminology in scale development and evaluation.
Main Methods:
- Conceptual analysis of psychometric terms.
- Examination of scale properties related to reliability and validity.
- Review of historical and contemporary measurement theory.
Main Results:
- Scales can exhibit high reliability (precision) but still be contextually unreliable.
- "Accuracy" as a synonym for validity represents an outdated conceptualization.
- Current validity frameworks emphasize the interpretability of test results for drawing conclusions about individuals.
Conclusions:
- The terms "precision" and "accuracy" are not adequate substitutes for "reliability" and "validity."
- Adopting traditional terminology enhances clarity in scale development and application.
- Correctly distinguishing these concepts is crucial for robust scientific measurement.