9/2 record (已更新)

9/2主要任務	進展與困難
問題釐清	ICC(2,1)和 ICC(1,2)找到的例子有點類似，但卻用不同算法，會有點困惑，可能因為目前只懂基本概念。    理解錯誤，已修改範例。 需要從ANOVA的角度思考，依研究設計來做選擇。
9/3主要任務	目標
寫背景前言?
流程	問題確認與修改段落

                                                 Intra-class correlation coefficients

There’s six different formulas for calculating the ICC which depend on the purpose of the study, the design of the study and type of measurements taken. The first number designates the model, and the second number designates the form.

“Models” of the ICC

Model 1 – each subject is assessed by a different set of randomly selected raters. This is rare in reliability studies.	每一受測者之施測者不一定相同，但數目一定，由一群施測者中隨機選出，每一受試者的施測者數小於施測者總數。此設計較不普遍，此分析較不實際也較不符合臨床研究目的。使用one-way ANOVA分析。[1]
Model 2 – each subject is assessed by each rater, and raters have been randomly selected.	每一受試者之施測者相同，每一受試者的施測者由一群施測者中隨機選出，數量小於施測者總數；算出之信度可反映施測者間同意度，施測者間是否可互換。使用two-way ANOVA分析。[1]
Model 3 – each subject is assessed by each rater, but the raters are the only raters of interest.	每一受試者之施測者相同，每一受試者的施測者數等於施測者總數，算出之信度反映特定施測者施測時之準確度，但無法代表其他施測者。使用two-way ANOVA分析。[1]

 “Form” of the ICC

The form reflects whether the reliability is to be calculated on a single measurement or by taking the average of 2 or more measurements taken by different raters. In most cases, the form will be 1, however if you want to test whether taking an average of 3 raters’ scores improves reliability, you might use form 2,3,4,etc.

Single measurement = 1

Average of 2 measurements = 2

Average of 3 measurements =3….etc

http://www.uvm.edu/~dhowell/methods9/Supplements/icc/More%20on%20ICCs.pdf

ICC公式之選擇

採用之公式	舉例
ICC(3,1)	(A於第一天評估2次，A於第二天評估2次，B於第二天評估1次) 施測者有A與B。施測者A之第一次評估結果和同天施測者B之評估結果相比(施測者間信度) [1]
ICC(2,1)	(一位評估者C於第一天評估1次，於第二天評估1次) 將第一天評估結果和第二天結果相比。[2]再測信度常用。
ICC(1,1)	(A於第一天評估2次，A於第二天評估2次，B於第二天評估1次) 施測者有A與B。分別將第一天A第一次之評估結果和和A第二次之評估結果相比，及將第二天A第一次之評估結果和和A第二次之評估結果相比(施測者內信度) [1]
ICC(1,2)	將第一天兩次評估結果之平均和第二天兩次平均結果相比。[1]

Equations (1,k), (2,k) and (3,k) are used when the unit of analysis is the mean measurement obtained either from more than one measurement or from more than one rater (k in this situation does not always refer to the number of raters)

If the aim is general application in clinical practice or research trials,

ICC (2,1) is appropriate, and a greater number of raters than in the present study would be required. If testing is only to be performed by a small number of raters who are the same raters used in the reliability study, ICC (3,1) is the choice, as is the case with the present study.

1.      Rankin G, Stokes M. Reliability of assessment tools in rehabilitation: An illustration of appropriate statistical analyses. Clinical Rehabilitation. 1998;12:187-199

2.      Lee P, Li PC, Liu CH, Hsieh CL. Test-retest reliability of two attention tests in schizophrenia. Archives of Clinical Neuropsychology. 2011;26:405-411