本研究目的有二：1.分析加、減法應用題中的語文知識對問題難度之影響。2.探討動態評量在評估學生解不同語文知識情境之文字應用題的效果。 研究一，以282位小學二年級的學生為受試者，採受試者內研究設計。每一位受試者都要回答不同的語文知識情境的加、減法應用題。研究中的應用題包含改變問題類型和比較問題類型，為兩個獨立的2（語意結構）×3（語意經驗）×3（語意陳述）三因子重複數量設計。研究結果發現：語文知識裡的語意結構、語意經驗、語意陳述三個因素都會影響問題的難度，且三個因素之間有交互作用。 研究二，根據研究一的結果，設計不同難度的問題（近遷移題、中遷移題、遠遷移題），選取解題能力達一定標準的60名小學三年級的學生（排除解題能力最好與最差的學生）為受試者，將他們隨機分派至實驗組及控制組。實驗處理分為兩個階段，實驗組學習階段接受語意結構圖示教學，遷移階段則接受漸進教學提示之動態評量。控制組在學習階段則實施一般傳統教學，遷移階段則提供受試者解題是否正確的回饋，如果解題失敗，則給予解題示範說明，研究結果發現： 1.實驗組在學習階段後的中測，及遷移階段完畢後的後測，再解題能力表現並未優於控制組。 2.簡單消極回饋的提示、問題轉譯的提示、工作記憶的提示、提示解題重要關鍵等四類教學提示占全部教學提示量87%，而提供語意圖示教學僅占全部教學提示量的8%。

　　　　　There are two purposes in this study. First, it tries to shed some light on the effect of the language knowledge on the difficulty of problems in simple additive/subtractive word problem sloving. Second, it also attemptes to examine the effectiveness of dynamic assessment on students' ability in solving additive/subtractive word problems of different language context. This study comprises two parts. Research one is conducted in terms of completely repeated-measure design. The subjects of the study are 282 second grade students in primart school. The subjects are required to answer the additive/subtractive word problems in different context of language knowledge. The word problems used in this study include the patterns of change and comparative problems. This is done in two independent case studies of 2 (semantic structure) 53 (semantic experience) 53 (semantic statement)factors design. It is found that in additive/subtractive word problem solving, all three factors, the sematic structure, semantic experience and semantic statement, are interactive and all have significant effects on the difficulty of problems. Research two, based on the results of research one, is designed in terms of the different levels of difficult in problem solving (near-transfer, mid-transfer, and far-transfer). Sixty third grade students with average ability of problem solving are selected and divided at random into the control and the experimental group. Students in the experimental group received the schematic drawing instruction in training stage and are instructed with graduated prompting assessment in transfer stage. As for students in the control group, however, they are treated in general instruction in the learning period. In the stage of transfer, their answers to the word problems are provided with proper feedback, such as correct demonstration was then given, if they fail in answering the question. The results of study indicate: 1)in the mid-test after the learning stage and the post-test after the transfer stage, the experimental group students' performance in problem solving was not superior to that of the control group. 2)The prompts of instruction in simple negative feedback, problem translation, working memory refresher, and key-note inspiration, constitute 87% and the schematic drawing instruction constitutes only 8% of whole instructions.