研究目的：本研究目的為驗證中風復健動作評估量表(Stroke Rehabilitation Assessment of Movement , STREAM)之再測信度。研究方法：研究者徵召50名復健科門診慢性期中風病人，接受治療師於前後問隔一週共2次的STREAM測量。研究者利用組內相關係數(intra-class correlation coefficient, ICC)驗證再測結果之一致性。再依據ICC計算測量標準誤(standard error of measurement, SEM)及最小真正改變量(smallest real difference SRD)，以利臨床及研究人員解釋個別病患或群體病患動作能力恢復是否超過測量誤差。研究結果：STREAM再測信度ICC值高達0.989（95%信賴區閒為0.981~0.994），顯示2次測量結果之一致性良好。STREAM總分之SEM數值代表受試群體改變值(within-group change)是否具備真實變化（或超過測量誤差）之閾值為2.00。STREAM總分之SRD數值代表單一個案改變值(within-individual change)是否為真正改變之閾值為5.54。結論：本研究結果證實STREAM具有良好的再測信度與測量誤差，臨床治療師和研究人員可根據STREAM之SEM及SRD數值判斷中風病人之動作功能進步量是否超過測量誤差。

Purpose: The purpose of this study was to examine the test-retest reliability of the Stroke Rehabilitation Assessment of Movement (STREAM). Methods: We recruited 50 chronic stroke patients undergoing outpatient therapy at rehabilitation department of the Medical University Hospital. The STREAM was administered twice, with the tests conducted one week apart, by a specially trained therapist. Intra-class correlation coefficient (ICC) was used to examine the level of agreement between test and retest. According to ICC value, the standard error of measurement (SEM) and the smallest real differences (SRD), were used for clinicians and researchers to explain whether the extent of improvement within-group change or within-individual change beyond the measurement error. Results: ICC was 0.989 (95% CI=0.981~0.994), indicating good test-retest agreement. The SEM of the STREAM was 2.00, representing the smallest change threshold that indicates a real improvement within-group change. The SRD of the STREAM was 5.54, exhibiting the smallest change threshold that indicates a real improvement within-individual change. Conclusion: The results indicate that the STREAM has high agreement between the test-retest measurements with acceptable measurement errors. The SEM and SRD of the STREAM can be used for therapists and researchers to judge whether the changes of movement function beyond the measurement errors of the stroke patients.