本研究目的在協助彩虹（化名）國小數學低成就學生，提高他們對3年級數學文字題的解題能力，並比較傳統錨式與強化錨式策略的成效差異。本研究從96-100年度花蓮縣學測題庫抽取3年級數學文字題各20題，編成前測、後測與延宕後測複本試卷，並從前測篩選出30名低成就學生，隨機分派至實驗組與控制組，各20節的補救教學於2013年1月寒假實施，教材針對前測答對率低於60%的6題文字題而設計。本研究利用t考驗和共變數分析，以瞭解實驗的組內與組間效果。結果顯示：（1）傳統錨式與強化錨式策略都有助於改善數學低成就學生的文字題解題能力；（2）包括實物操作、角色扮演和類推練習等的強化錨式策略，比傳統錨式策略有更好的學生解題能力改善效果。

The first purpose of this study is to help improve the abilities of low-achievers in mathematics of a Hualien aboriginal primary school with their third-grade math word problems. It also aims to compare the effects on problem-solving strategies between traditional anchored instruction (AI) and enhanced anchored instruction (EAI). The pretest, posttest, and delayed alternate posttest used in this study contain 20 third-grade math word problems from Academic Competence Test Bank of Hualien County (2007-2011). The participants in this study are 30 low-achievers from the pretest. They are randomly assigned to the experiment and control groups for a 20-session remedial course during the winter vacation in January 2013. The materials designed are based on the items from the pretest where students failed to answer correctly, e.g., with poor performance under 60%. The anchored animations were used, for both groups, as supplement materials to help learners’ comprehension during the instruction. This study employed t-test and one-way ANCOVA to investigate both within-groups and between-groups effects. The results show that: 1) The AI and EAI are effective for improving low-achievers’ comprehension of word problems as well as their problem-solving abilities; 2) Applying EAI strategies (e.g., hands-on practices, role play, and analogy exercises) are more effective approaches to improve learners’ word-problem solving abilities.