Test-retest reliability assessment is used
to establish that an instrument is capable of measuring a variable with
consistency….If the test is reliable, the subject’s score should be similar on
multiple trials. In terms of reliability theory, the extent to which the scores
vary is interpreted as measurement error [1].
Test-retest reliability is the stability of
a measuring instrument over time [2].
Test-retest reliability indicates stability
of the assessment and determines the consistency of measures from one testing
occasion to another, on the assumption that the behavior being scored does not
change over time [3].
Reliability can be defined as the
consistency of measurements, or of an individual’s performance, on a test; or “the
absence of measurement error” [4].
Reproducibility evaluates whether an
instrument yields the same results on repeated assessments. Examples of
reproducibility are parallel-forms, rater and test-retest reproducibility [5].
1. Portney
LG and Watkins MP. Foundations of Clinical Research: Applications to Practice.
3rd ed2009, Upper Saddle River, New Jersey: Pearson Education.
2. Hobart JC, Lamping DL, and Thompson AJ.
Evaluating neurological outcome measures: the bare essentials. J Neurol
Neurosurg Psychiatry 1996;60:127-130.
3. James C, Mackenzie L, and Capra M.
Test-retest reliability of the manual handling component of the WorkHab
functional capacity evaluation in healthy adults. Disabil Rehabil
2010;32:1863-1869.
4. Atkinson G and Nevill AM. Statistical
methods for assessing measurement error (reliability) in variables relevant to
sports medicine. Sports Med 1998;26:217-238.
5. Hobart JC, Riazi A, Lamping DL,
Fitzpatrick R, and Thompson AJ. Improving the evaluation of therapeutic
interventions in multiple sclerosis: development of a patient-based measure of
outcome. Health Technol Assess 2004;8:iii, 1-48.
就我自己對test-retest reliability的概念，test-retest reliability可看二個方面，一個方面是重複施測結果的一致程度，另一方面是重複施測之測量誤差程度。
在FDT論文的test-retest reliability，我著重於重複施測結果的一致程度，而為要強調測量測量誤差，因為在此篇是著重於RCIp，列於practice effect底下，所以在test-retest reliability中，沒有刻意提到要看RCI。
looks good!
回覆刪除1. 您覺得最好的定義為何？
2. RCI-p 是因為 RCI 需校正 practice effect, 主旨還是在 RCI. 目前的想法似乎「反客為主」。
謝謝老師的意見，以下為我的回覆
刪除1. 我覺得最好的定義應為包括重複施測結果的一致程度和測量誤差程度的說法。例如：Atkinson et al. (1998)所下的定義 Reliability can be defined as the consistency of measurements, or of an individual’s performance, on a test; or “the absence of measurement error”.
2. 瞭解，我會再思考要如何重新架構文章結構，把RCI納入test-retest reliability的解釋中。