本研究旨在探討 Dash 機器人學習融入 清華DDMT 教學模式所設計的 STEAM 課程，對國小學生問題解決能力與學習態度的影響。隨著資訊科技快速發展，教育機器人逐漸成為教育現場進行資訊教育的重要工具之一。其中 Dash 機器人以其視覺化的APP應用程式介面與機器人本身的造型具有高互動性，適合作為國小學童初次接觸程式設計與培養初階問題解決能力的媒介。本研究以Dash機器人學習為核心，融入清華大學DDMT教學模式(發現、定義、模型與建模、遷移)做為課程設計主軸，結合 STEAM 教育理念與連結 SDGs(聯合國永續發展目標)情境，以培養學生跨領域理解、合作學習與探究行動歷程，進一步提升其問題解決能力。
在課程實施歷程中，學生會依循清華DDMT教學模式的四個階段進行，包括在「發現」階段從真實情境中辨識問題線索；在「定義」階段釐清問題並形成解決方向；於「模型與建模」階段運用 Dash 機器人進行程式設計與反覆測試；最終在「遷移」階段進行成果分享與反思，並嘗試將解決方案延伸至其他情境。此流程強化學生在課堂中逐步建構問題理解、方案規畫與實作驗證的能力基礎。
本研究採用準實驗單組前後測研究設計，研究對象為新竹市某國小一至六年級208名學生，課程為期十週，每週進行一次40分鐘課程。量化資料包含「Dash 機器人學習態度量表」及「STEAM 課程學習態度量表」之前、後測，以檢視學生學習態度的改變；質性資料則包含教師觀察記錄、學生學習歷程、學習單內容與開放性填答等，以補充量化結果並呈現學生在 DDMT 四階段中展現之問題解決歷程。
研究結果顯示，以清華DDMT教學模式建構之STEAM課程，能提升多數學生在Dash機器人學習及STEAM課程學習態度上的正向表現，其中以中年級學生最為明顯；低年級在實作階段投入度高，但於語文表達與書寫上較需協助；高年級因課程連續性受外部因素影響，學習態度呈現較不穩定的變化。此外，從質性資料可看出學生能依照清華DDMT教學模式四階段逐步建構問題理解、方案規畫、模型測試與反思遷移等問題解決行為，並在小組合作中展現學習動機與參與度。整體而言，本研究設計之課程具有促進國小學生學習態度與問題解決歷程的正向價值，並呈現出年段差異對課程成效的不同影響。

This study aims to investigate the effects of integrating Dash robot learning into a STEAM curriculum designed based on the Tsinghua DDMT teaching model on elementary school students’ problem-solving abilities and learning attitudes. With the rapid development of information technology, educational robots have become important tools in school-based information education. The Dash robot, with its visual programming interface and highly interactive design, is suitable for introducing elementary students to programming and supporting the development of early problem-solving skills. In this study, Dash robot learning serves as the core instructional medium, while the curriculum is structured according to the Tsinghua DDMT model—Discover, Define, Model & Modeling, and Transfer—and incorporates STEAM education principles as well as SDGs related real-world scenarios to promote interdisciplinary understanding, cooperative learning, and inquiry-based engagement that fosters students’ problem-solving processes.
During the curriculum implementation, students followed the four stages of the Tsinghua DDMT teaching model. In the Discover stage, they identified problem clues from authentic scenarios; in the Define stage, they clarified problems and formed solution directions; during the Model & Modeling stage, they used the Dash robot for visual programming, hands-on testing, and iterative refinement; and in the Transfer stage, they presented results, reflected on their learning, and explored how solutions could apply to new contexts. This process strengthened students’ progression in problem understanding, planning, and practical verification.
A quasi-experimental one-group pre- and post-test design was adopted. Participants were 208 students from grades 1 to 6 at an elementary school in Hsinchu City who engaged in a 10-week course consisting of weekly 40-minute sessions. Quantitative data were collected using the Dash Robot Learning Attitude Scale and the STEAM Curriculum Learning Attitude Scale to examine changes in learning attitudes. Qualitative data, including teacher observation records, students’ learning artifacts, DDMT learning sheets, and open-ended responses, were analyzed to supplement quantitative findings and reveal students’ problem-solving behaviors across the DDMT stages.
The results indicate that the DDMT-based STEAM curriculum enhanced learning attitudes toward Dash robotics learning and STEAM learning for most students, with the most pronounced improvement observed in middle-grade students. Lower-grade students showed high engagement during hands-on activities but required support with reading and writing, while upper-grade students exhibited more fluctuating attitudes due to interruptions in course continuity. Qualitative data further revealed that students gradually demonstrated problem understanding, solution planning, model testing, and reflective transfer aligned with the DDMT stages, and actively engaged in group collaboration. Overall, the curriculum positively contributed to students’ learning attitudes and problem-solving processes, revealing meaningful grade-level differences in their learning performance.

The study anticipates that, under teacher guidance, students will gain foundational programming knowledge and skills through Dash robot activities. By engaging in SDG-themed tasks, students will exhibit high learning motivation, collaborative performance, self-reflection and the ability to generate innovative solutions using design-thinking processes. These outcomes are expected to manifest in interdisciplinary knowledge integration, logical reasoning, computational thinking, and creative problem-solving capabilities.