本研究主要目的是以閱讀的自動化理論為基礎，為國小二至六年級朗讀流暢度做篩檢準 確度及最佳切截點分析。研究中，以閱讀理解答對率作為效標變項，朗讀流暢度z分數作為預 測變項，然後採用「接受者操作特徵」（Receiver-Operating-Characteristic, ROC）分析，根據 「ROC 曲線下面積」之大小，來檢測採用朗讀流暢度篩檢因認字效能不足導致閱讀理解困難 之學童的準確度。此外，也根據χ 2 值、命中率、敏銳度、錯誤拒絕率及錯誤接受率等數據， 對朗讀流暢度做最佳切截點的分析。研究結果顯示，以朗讀流暢度來篩檢因認字效能不足導 致閱讀理解困難的學童，具有中等的準確度。而且以二至六年級上下學期的資料而言，朗讀 流暢度z分數最佳切截點都是-0.5。本研究也提供國小二至六年級上下學期的朗讀流暢度平均 值資料，以供教育研究者及教師瞭解二至六年級學童認字效能平均成長情形之參考。

This study evaluated the assessment accuracy of oral reading fluency. Additionally, this study focused on determining the cut-off point for identifying poor reading comprehension caused by low decoding efficiency in students of grades 2-6. Receiver operating characteristic (ROC) analysis was adopted to evaluate the assessment accuracy of oral reading fluency. The area under the curve (AUC) of the ROC indicated that oral reading fluency exhibited medium accuracy in screening. Based on analyses of χ 2 , the hit rate, sensitivity, false negative rate, and false positive rate, a z score of oral reading fluency below -0.5 was identified as the optimal cut-off point for screening students with poor reading comprehension caused by low decoding efficiency. In addition, the mean scores of oral reading fluency for students in grades 2-6 are provided to inform teachers and researchers of the level of oral reading fluency in elementary-school students.