本研究選取南投縣某國小兩班三年級學童為研究對象,研究者任教的班級擔任實驗組,採閱讀理解推論策略融入國語科教學;另一班為研究者同學年教學夥伴教師任教的班級擔任對照組,採傳統講述式的國語科教學。實驗課程實施十週二十節課,教材為選用國語科文本,進行「課文本位」閱讀理解教學,探討經由閱讀理解推論策略融入國語科的教學後,對國小三年級學童閱讀理解以及運用推論策略的能力是否有影響。在進行教學前、後以黃堯香(2007)編製的「國小四年級學童中文閱讀理解測驗」作為前測、後測工具,透過統計假設檢定,統計測驗得分,進行量化分析,再輔以質性描述學童學習成效。本研究結果如下: 一、 關於「題目類型」部分,實驗組學童在「文本理解」、「字義理解」、「推論理解」以及「摘取要旨」類型的閱讀理解成績上有顯著進步。對照組學童在「文本理解」、「字義理解」、「推論理解」類型的閱讀理解成績上有顯著進步,在「摘取要旨」類型的閱讀理解成績上並無顯著的進步。二、 整體顯示,比較實驗組和對照組學童前後測成績,顯示閱讀理解能力均有顯著進步,且實驗組成績優於對照組成績。 根據上述結果,本研究對閱讀理解推論策略教學提出建議。 The participants are third-grade students from class A and class B in an elementary school in Nantou County. The students in class A were chosen as the experimental group while the students in class B were chosen as the control group. The researcher integrated inferential strategies into Chinese language arts instruction for the students in the experimental group while the students in the control group were taught with traditional expository strategies for Chinese language arts instruction. The experiment for text-based reading comprehension was allocated in 20 classes within ten successive weeks and the Chinese textbook was the teaching material for both the experimental and control group students. The study aimed at exploring the effects of integrating inferential strategies into Chinese language arts instruction on the third-grade students’ performances in reading comprehension as well as their abilities in using inferential strategies while reading. The Chinese Reading Comprehension Test for the Fourth Graders designed by Huan Yao-Hsiang was adopted in both the pretest and the posttest. The score of the test, obtained via statistical hypothesis testing, was quantively analyzed and qualitatively illustrated. The results of the research indicate that:1. The students in the experimental group made significant progress in the tests of text comprehension, literal comprehension, inferential comprehension and summarizing. The students in the control group also made significant progress in text comprehension, literal comprehension and inferential comprehension; however, they did not make significant progress in summarizing.2. The results indicate that although both the experimental and control group students made significant progress in reading comprehension, the test score of the experimental group was higher than that of the control group. The results above could be used as a reference for the researchers who take interests in inferential strategies for reading comprehension.