以德菲法評析國際運算思維挑戰賽題型之研究

Abstract

運算思維能力已日趨重要，世界各地的教育家也開始發展關於運算思維能力培養的課程，而如何有效地評量學生的學習成效也十分重要，目前有許多測驗聲稱能夠評量學生的運算思維能力，但未能說明評量到哪一項概念。因此，本研究以廣為人知的國際運算思維挑戰賽（International Challenge on Informatics and Computational Thinking）為例，利用德菲法進行三回合問卷調查，蒐集專家們的想法，統整出挑戰賽題與運算思維核心概念之關係，並分析各類型專家進行題目評析時的差異，包含哪一類型專家較不受他人影響及其想法較具代表性。 本研究礙於題目樣本數及專家人數的關係，將專家分為兩群，評析兩份不一樣的問卷，作為對照，以探討各類專家的差異。 實驗結果得知，三回合德菲法問卷調查，可將挑戰賽題依據運算思維五大概念分類，且發現未有任何一題被分類為「分解」這一概念。另外，本研究亦發現Bebras Challenge之命題者在評析題目時也容易受他人影響，並且其填答也未必具代表性，主要是德菲法的匿名性使得評析結果更加客觀。

Computational thinking has become increasingly important. Educators around the world have also begun to develop courses on the development of computational thinking skills. The measurement of students' learning effectiveness is also very important. There are many tests that claim to be able to measure it. But most of them can not be explained which concept is evaluated. Therefore, this study takes the “International Challenge on Informatics and Computational Thinking” as an example. We use Delphi method to conduct a three-round questionnaire survey, collects the ideas of experts, and integrates the core concept of computational thinking and the tasks. We analyzed which types of experts are less affected by others and which are more representative. In this study, due to the number of questions and the experts, the experts were divided into two groups with two different questionnaires that were evaluated as a comparison to explore the differences among various experts. The experimental results show that the three-round Dephi questionnaire can classify the tasks according to the concepts of computational thinking, and find that no one is classified as "decomposition". In addition, this study also found that the Bebras Challenge's propositions are also susceptible to others when evaluating the questions, and their answers are not necessarily representative, mainly because the anonymity of Dephi makes the results of the evaluation more objective.