探討以「建模導向探究」及「專題導向探究」的教學策略融入「探究與實作｣課程設計下之學生學習成果

Abstract

隨著「探究與實作」課程被列為部定必修課程，大考中心規劃將重視情境脈絡且能運用科學上的方法論的素養導向試題列為大學考試的命題方向。另一個受到矚目的是學生在課程產出的實作作品，將成為未來申請大學時的重要資料。本研究將兩種教學設計－「專題導向探究」及「建模導向探究」的教學策略融入「探究與實作｣課程。兩組課程大致分成兩個階段。第一階段先由教師利用三周的教學分別協助不同組的學生搭設探究歷程或建模歷程的鷹架，第二階段學生再利用三周的時間進行開放式問題的探究。同時，本研究改編PISA試題及歷年全國科展作品，設計科學素養導向測驗。並探討學生在課程前後的科學素養變化以及在課程中產出之實作作品的情形作為學生學習的成果表現。此外，有鑑於科學教育領域對於模型與建模的日漸重視，本研究將比較學生在兩種模式下對於模型本質的了解與建模能力的培養情形。 研究結果顯示，專題導向探究組在科學素養測驗在總分、提出可驗證觀點、尋找變因或條件、分析資料和呈現證據、提出結論或解決方案及表達與溝通等項目後測優於前測，且達顯著差異。而建模導向探究組僅在總分、分析資料和呈現證據、及表達與溝通等項目後測優於前測，且達顯著差異。在運用Johnson - Neyman法比較兩組探究模式在科學素養測驗的表現時，在總分、論證與建模、表達與分享等項目，皆得到專題導向探究教學適合素養評量前測較高的學生學習，而建模導向探究教學適合素養評量前測較低的學生學習的推論。而在發現問題與規劃與研究則以共變數進行分析，兩組無顯著差異。 而在模型本質問卷的施測結果，建模導向探究組在模型本體面向、認識面向與方法面向，前後測的t-檢定顯示後測優於前測，且達顯著差異。專題導向探究組的問卷前後測t-檢定顯示僅在模型本體面向後測優於前測，且達顯著差異。進一步兩組前後測共變量的分析顯示，在模型本體、認識與方法等三個面向，建模導向探究組的表現皆優於專題導向探究組，且達顯著差異。 在建模能力測驗的表現上，使用費雪精確檢定結果顯示，兩組的學生在模型選擇、模型效化兩項度的表現有差異，而在模型建立及模型應用與調度皆未達顯著差異。其中又以建模組在高層級(Level4與Level5)的人數多於探究組，可知建模教學有助於提升模型選擇與模型效化之建模能力。 在學生實作作品方面，建模導向探究組的學生分為8組，其中2組的成果能夠利用科學理論驗證模型並應用於其他情境，有4組可以建立模型的延伸關係，有2組可以建立模型的成分關係。對於開放性探究問題，建模導向探究組的學生能夠掌握選擇正確變因、並且能透過實作找到變因之間的定性關係，但是距離科學社群能夠形成模型解釋實作成果仍有一段差距。 在專題導向探究組將學生分為8組，有7組完成作品，但有1組未完成實驗驗證。評量結果顯示，在有7組在「提出適合科學探究問題」、有6組在「提出符合研究問題的設計」、有5組在「能分析數據」、有5組在「會製作並應用圖表」、有7組在「提出結論或建立模型」方面達到3分(可)以上。對於開放性探究問題，專題導向探究組的學生能夠掌握控制變因實驗方法的設計、認為本課程在實驗設計幫助最多，但仍會體會到問題的複雜與不確定性。 整體而言，以｢專題導向探究」及「建模導向探究」的教學策略融入「探究與實作｣課程，學生有豐富的學習成效也多能產出實作作品。｢專題導向探究」設計在培養學生科學素養的成果較佳，而「建模導向探究」則在培養學生對模型本質的概念與建模能力有較好的效果。兩組學生對於開放式的真實問題探究，皆體會到問題的複雜與不確定性，必要時教師需要協助對學生的探究調查技巧以及科學背景知識搭設鷹架，才能引導學生克服真實探究問題的挑戰。

This study integrates two instructional designs—“project-based inquiry” and “modeling-based inquiry”—into the “Science Inquiry and Practice” curriculum. The two sets of curricula are divided into two stages. In the first stage, teachers spend three weeks scaffloding students the inquiry or modeling process. In the second stage, students use another three weeks to conduct research for investigating open-ended questions. With the “Science Inquiry and Practice” curriculum listed as a required course stipulated by the Ministry of Education in Taiwan, the College Entrance Examination Center has planned to apply literacy-oriented test questions, which are regarded as a context-focused and scientific methodology, to designing test items for national examnination for university admissions. Another focus is the practical works produced by students in their hands-on courses which will serve as an important portfolio used for their future university application. This research has designed scientifically literacy-oriented tests adapted from the PISA assesement and students’ works of the annual national science exhibitions. It will also explore the differences of students’ scientific literacy before and after the course and analyze students’ learning performance in light of their practical works produced in the course. In addition, this study will compare the extent to which students understand the essence of models and how teachers cultivate students’ modeling capability under the two modes of inquiry, given an increasing emphasis on models and modeling in the field of science education. The results of this study illustrated that the performances of the project-based inquiry group in posttests were far superior to those in pretests with significant differences in these performances. These performances included “overall scores,” “abilities to give a verifiable perspective,” “the search for variables or conditions,” “analysis of data and presentation of evidence,” “ability to draw conclusions or provide solutions,” and “expression and communication.” Applying the Johnson-Neyman procedure to comparing the two groups of inquiry models in the scientific literacy test, this study discovered that in terms of overall scores, argumentation and modeling, as well as expression and sharing items, the project-based inquiry pedagogy were suitable for students with higher levels of achievement in the scientific literacy test. In addition, this study also found that modeling-based inquiry pedagogy were suitable for students with higher levels of achievement in the scientific literacy test. However, after identifying problems and planning and research were analyzed by covariates, this study found that there was no significant distinction between the two groups. In the nature of model questionnaire, the t-test of the modeling-based inquiry group showed that its posttests were better than pretests in terms of the model ontology, cognition and methodology with significant differences among these tests. Based on the questionnaires of the project-based inquiry group carried out by the pre-and-post t-tests, the results showed that the posttests were better than the pretests only in terms of the model ontology, and they reached a significant difference level. However, there is no significant difference in cognitive perspective and methodological perspective. A further analysis of the pre-and-post-test covariates of the two groups revealed that the modeling-based inquiry group outperformed the topic-based inquiry group on model ontological perspective, cognitive perspective, and methodological perspective with significant differences in these three aspects. In the performance of the modeling ability test, Fisher’s accurate test results showed that the two groups of students produced different performances in model selection and model validity, but there was no significant difference in model construction, model use and deployment. Furthermore, students in the modeling-based inquiry group at the higher level (Level 4 and Level 5) outnumbered those in the project-based inquiry group. Therefore, it was apparent that modeling instruction facilitated the modeling ability associated with model selection and model validity. In terms of students’ practical works, students in the modeling-based inquiry group were divided into 8 groups. Among these 8 groups, 2 groups could be able to examine models using scientific theories and apply them to other scenarios. Four groups could be able to build an extended relationship of the model while another 2 groups could be able to form the component relationship of the model. As far as the open-ended questions were concerned, students could select the correct variables ; besides, through hands-on courses, they could find the qualitative relationship among the variables. Nevertheless, there was a large discrepancy between students’ practical works and the models constructed by scientific communities. In addition, students were divided into 8 groups in the project-based inquiry group. 7 groups completed the inquiry works and submitted reports, but 1 group failed to complete the experimental verification. In general, students perform better in “raising questions suitable for scientific inquiry” and “being able to come up with designs that meet research questions,” whereas there is room for growth in “being capable of analyzing data” and “being able to produce and apply charts.”As far as the open-ended questions were concerned, students in the project-based inquiry group could conduct experiments with selecting the correct variables but experience these questions’ complexity and uncertainty. Students feel that this curriculum helps a lot when they design an experiment. Overall, the students had an abundance of learning outcomes and produced practical works under the “project-based inquiry” and “modeling-based inquiry” teaching strategies integrated in the “Science Inquiry and Practice” curriculum. “Project-based inquiry” designs achieved better results when applied to cultivating students’ scientific literacy, while “modeling-based inquiry” was more effective in cultivating students’ concepts of the essence of models and developing their modeling ability. Dealing with the inquiry of open-ended questions, the two groups of students experienced these questions’ complexity and uncertainty. Teachers need to assist students in scaffolding the skills of investigation and prior knowledge of science wherever necessary so that they can instruct students how to overcome the challenge of investigating authentic problems.