Implementing Bayes' Theorem to Analyze Interactive Web-Based Scientific Inquiry Assessments: Inquiring Time-Temperature Graph on Atmospheric Climate Change Issue

Abstract

noneFrom an early age, students in formal schools have been taught how to interact with nature through natural science lessons. However, have we made sure the lesson is appropriate to teach interaction with nature? Climate change is happening and has already been confirmed by scientists; teachers need to take action in class to spread awareness. Education is a way to educate people and can be a good way to prevent climate change before the tipping point (+1.5° C) is exceeded. Even worse, today, humans face many crises in many fields along with the development of human civilization; this thesis will discuss three crises around the field: the climate crisis, the 21st-century learning crisis, and the replication crisis. Three of them are demonstrated in a study of Scientific Inquiry Assessment in this thesis which included two studies. The First Study discusses Revolutionizing Scientific Learning using Innovating Interactive Web-based Assessment for Scientific Inquiry. This study aims to analyze the process of creating an interactive web-based evaluation that provides complete information about students' scientific inquiry abilities. Science education reform is transpiring worldwide, facilitating scientific inquiry (SI) ability which aims for students to understand how scientists do their work. Interest has rapidly expanded in integrating scientific inquiry into computer-based assessments, which can measure complex inquiry skills more effectively. According to Bayesian inference, there are differences between some demographic characteristics among eleventh-grade students from natural science classes in Southwest China’s public schools in terms of students’ scientific inquiry ability, in which “being” male and schools located in the city have evident to have higher probability on understanding climate change issue. The Second Study discusses Bayesian Multilevel Model Analysis on Scientific Inquiry Ability. Understanding the replication crisis is a hot topic that is widely discussed among scientists. This replication crisis is characterized by the difference between the latest results and the previous results. There is a need for alternatives for hypothesis testing in science education. This study aims to demonstrate and apply Multilevel Bayesian statistics to analyze the scientific inquiry ability of students in China. Finally, a truncated Poisson regression-based approach was used, clustered by Student ID, which was a good fit for the data. In summary, results varied widely between districts and schools but less so between classrooms. Moreover, adjusting for gender (which is an important control) is taken into account; students who take longer to take the test score about two points higher on average. In conclusion, this thesis answers three main keywords on the design of this thesis: Technology (Computer-based Assessment) as the answer to the 21st-century learning crisis, Bayesian (Statistical analysis) as the answer to Replication Crisis, and Climate change awareness content as the answer for Climate Change Crisis.