A Mixed Assessment for the Science Learning via a Bayesian Network Representation

Abstract

This study explored an alternative assessment model to examine Chemistry learners’ progress. “The Assessment of Problem-Solving in Chemistry Learning” as a model represented students’ mastery of chemistry study. The data were from journaling narratives and analyzed through cognitive task analysis. Based on the analyses, a student model was established, which represents the qualitative information in a structure, and provides a potential framework of the assessment model for the quantitative representation—a Bayesian network assessment model. The student’s performance was assessed via the Bayesian network assessment model, and classified into three categories: low level, middle level, and high level. The mastery level should be at least scored at and above 90.51/100 for Declarative, Procedural, and Strategic Knowledge respectively.

Mapping the Problem Statement and Study Purposes to the Quantitative Analysis: Insights From Mixed Methods Perspectives

“…BayesiaLab was used to develop the thematic network, building upon the structure established through Semantic Network Analysis. The directed graph techniques in BayesiaLab facilitated the exploration of connections within the data (Zhang, 2022a(Zhang, , 2022bZhang & Guanzon, 2022;Zhang & Zhang, 2020). By employing these techniques and tools, the authors conducted a comprehensive analysis that integrated both quantitative and qualitative perspectives.…”

Consistency Examination of a Systematic Review and Data Analysis Report for a Social Justice Study—A Mixed Research Model

“…In the second phase, four items were incorporated to evaluate students' access to academic resources, including tutorial assistance in subjects like math, writing, physics/chemistry, as well as other services offered within the college environment (Zhang & Garcia, 2022;Zhang & Guanzon, 2022). Phase three of the questionnaire examined the involvement of students in various College Student Outreach Programs.…”

The Exploratory Sequential Design With Generalized Linear Models for the Learning Strategy Analysis

“…There is a diversity of different mixed methods and data analysis models, which may be applied to this study. Creswell and Plano Clark (2018) introduced the six mixed methods designs, which were methodological candidates and may be applied to the cooperative learning (Zhang, 2007a(Zhang, , 2007b): a) convergent parallel design, b) explanatory sequential design, c) exploratory sequential design, d) embedded design, e) transformative design, and f) multiphase design (Zhang, 2007a;Zhang, 2007b;Zhang & Zhang, 2020;Zhang, Guanzon, 2022). Among these mixed methods designs, two of them are possibly used in this study: the explanatory sequential design and the exploratory sequential design.…”

“…Even though we cannot decide exactly a research questions should be responded to by a mixed research framework, some better suggestions to select a mixed methods design is still necessary. Some research questions are very obvious that should be responded to by using a qualitative or a quantitative research design; some research questions can use both qualitative methods design and quantitative methods design (Zhang, 2022).…”