探討中、低科學解釋評鑑表現的大學生其評鑑科學解釋的自我調節學習歷程

Abstract

學生能力國際評量計畫在2015年指出評鑑科學解釋是有效提升學習者科學素養的方法之一，且當學習者具備評鑑科學解釋的能力時，有助於提升學習者省思與評鑑的能力。諸多研究指出，學習者的自我調節學習表現(例如策略的類型或序列)在進行困難任務如：科學閱讀、科學探究或問題解決時，會影響其學習成果，然而自我調節學習歷程是否影響且如何影響評鑑科學解釋的成果，卻少有研究探討。故本研究目的在於比較科學解釋評鑑表現不同的學習者，其在評鑑科學解釋時的自我調節學習歷程是否不同?以及歷程的不同如何影響學習者評鑑科學解釋的表現。 本研究採質量混合設計，設計一則國中生物的科學解釋評鑑任務，以大學生於評鑑任務之表現進行分組。本研究先檢視學習者的先備概念多寡、有關科學認識論的知識以及科學解釋評鑑表現之關聯。參與者進行評鑑科學解釋任務的自我調節學習歷程，以放聲思考的方式收集學習者在閱讀探究情境與問題、實驗數據，及評鑑兩則科學解釋的過程，並進行自我調節學習策略的編碼。除了比較不同評鑑表現的學習者使用自我調節學習策略的類型和頻率之差異，亦以質性分析之方式提出兩組策略序列模式的不同。 研究結果顯示，先備概念、科學認識論及評鑑表現三者無顯著相關性。比較中、低表現的學習者使用各項策略之類型及頻率，在閱讀實驗數據時，表現不佳的學習者使用較多分析策略，而在評鑑第一則科學解釋時，表現中等的學習者則使用較多認知策略，前述之差異接近顯著但未達.05之水準。其餘任務階段中，兩組學習者使用的認知和後設認知策略其類型和頻率趨勢相似，使用頻率皆無顯著差異。自我調節學習策略的序列分析和質性分析結果顯示，表現中等的學習者較常出現閱讀→分析、分析→監控、推論→評估等策略序列，質性資料則顯示此組學習者會分析指出關鍵變因，並根據探究問題和數據趨勢推論自變項和應變項的關聯、監控分析的結果。在評鑑科學解釋時則會閱讀待評解釋的內容，分析並比較待評敘述與推論的結果，判斷待評解釋的完整性。表現不佳者則較常出現閱讀→監控、監控→閱讀、分析→閱讀等策略序列，質性資料則指出：此組學習者雖會分析並指出探究情境中的關鍵變因和依數據的趨勢推論變因間的關係，但皆未監控分析結果的品質，評鑑科學解釋時則僅分析待評敘述的文意是否與數據相符，即判斷待評解釋的完整性。 本研究結果指出，不僅自我調節學習策略的使用頻率，使用認知和後設認知策略的序列組合亦對科學解釋評鑑表現有重要影響。本研究的另一個貢獻在於以序列分析和質性分析捕捉自我調節學習的複雜歷程，對自我調節學習研究有重要啟示。

The PISA 2015 draft science framework indicates that providing opportunities to evaluate scientific explanations is one of the effective ways to improve scientific literacy. Many studies have indicated that process of self-regulated learning may influence learning outcomes when learners underwent challenging tasks, such as reading and understanding scientific texts or accomplishing an inquiry or a problem solving task. However, research that investigates the process of scientific explanation evaluation is rare. The purpose of this study is to examine whether and how the learners with different evaluation performances of scientific explanations differ in the process of self-regulated learning, as well as how the difference in the evaluation processes influence their outcomes. This study took a mixed-method approach. Nineteen undergraduates were assigned into either the moderate or the low performance group according to their performance on a scientific explanation task. The participants were to think-aloud while undergoing the task of scientific explanation evaluation. The cognitive and metacognitive strategies used during the task were coded for further analyses. A sequential analysis was conducted, supplemented with qualitative data, in order to understand how the evaluation process influences the evaluation performance. No correlation was observed between that the score of prior knowledge, the level of scientific epistemology, and the evaluation performance. The analysis of strategy use indicates that the low performing students use analyzing strategy more frequently while reading the data table, whereas the moderate performing students demonstrated more cognitive strategies collectively. Neither difference in use of strategies reached a significant level. No difference was observed upon the use of other strategies. Findings from the sequence analysis and qualitative analysis indicated that the moderate performers were able to analyze and infer relationship between independent and dependent variables according to the question of inquiry and the data trend. The to-be-judged explanation statements were carefully read, analyzed, and compared with their mental model for making quality. Monitoring was performed to ensure the quality of the thinking process and outcomes. Although the low performers would analyze and infer the relationship between variables according to data trend, they failed to monitor the result of thinking. They made quality judgment by simply determining whether the literal meaning of the words aligned with the data table. This study shows that learners who can effectively integrate cognition and metacognition strategies will perform better on the tasks of scientific explanation evaluation.