測量工具的效度與信度

Validity and Reliability of an Instrument

李中一(Chung-Yi Li)

《台灣公共衛生雜誌》 23卷4期 (2004/08) Pp. 272-281

https://doi.org/10.6288/TJPH2004-23-04-02

摘要

目標：介紹測量工具的信度與效度概念與評估的方法，討論使用缺乏信度與效度的工具對研究結果的影響，並針對某些測量實務上的議題提出建議。方法：透過文獻探討，摘要其內容，列舉實例或假想之數據，說明測量工具的信度與效度概念及其評估方法。結果：測量工具效度的高低取決於測量所牽涉系統誤差大小，而信度大小則與隨機誤差有關；評估測量工具信效度的方法依測量資料的屬性而定，而流行病學研究將會因為所使用測量工具的信效度不完善而產生訊息偏差，此偏差的程度與方向則與測量誤差的本質有關。結論：研究者在使用測量工具前有必要選擇適當的方法評估其相關之效度與信度，以了解使用該測量工具對研究結果的可能影響，如此方能對研究結果作正確的闡釋。

關鍵字

信度 ； 效度 ； 流行病學 ； 偏差 ； 測量誤差

並列摘要

Objectives: This paper illustrates the concept of validity and reliability associated with an instrument and how the validity and reliability are assessed. We also included a discussion on how an instrument with unsatisfactory validity and reliability may affect study results, and provide suggestions for certain practical problems encountered by investigators.Methods: We reviewed the literature and provided real-world or hypothetical examples.Results: The level of validity of an instrument is related to the magnitude of systematic errors associated with that instrument, while the level of magnitude of reliability is determined solely by the degree of random errors involved in the measurement. The choices of methods used for the assessment of validity and reliability depend on the attribute of research data. Results from epidemiological studies that used an instrument with non-perfect validity and reliability might entail certain degrees of bias, for which the direction and magnitude are associated with the nature of measurement errors.Conclusions: Researchers should assess, using appropriate methods, the validity and reliability of an instrument before it can be used. This would help to appreciate the potential effects on the study results caused by measurement errors, and lead to correct interpretations of the study results.

並列關鍵字

Reliability ； Validity ； Epidemiology ； Bias ； Measurement error

參考文獻

全中妤、杜宗禮、葉文裕、李中一(2002)。台灣旅館業餐飲人員工作動作特性與肌肉骨骼傷病之橫斷式研究。台灣衛誌。21，140-9。

Bland JM,Altman DG(1986).Statistical methods for assessing agreement between two methods of clinical measurement.Lancet.1,307-310.

Cohen J(1960).A coefficient of agreement for nominal scales.Educ Psychol Meas.20,37-46.

Cohen J(1968).Weighted kappa: nominal scale agreement with provision for scaled disagreement or partial credit.Psychol Bull.70,213-220.

Copeland KT,Checkoway H,McMichael AJ,Holbrook RH(1977).Bias due to misclassification in the estimation of relative risk.Am J Epidemiol.105,488-495.