The Influence of Personality on Epistemic Network in an Open-Ended Question Discussion Scenario

Students' cognitive pattern of scientific reasoning has an important influence on learning effectiveness. It is generally believed that the mutually complementary student cognitive patterns of scientific reasoning can benefit every participant in a discussion group on an open-ended scientific question. However, due to the absence of necessary and convenient means, it is rather difficult for teachers to predict students' cognitive patterns in scientific reasoning before grouping, resulting in an intuitive-only grouping schema which is short of scientific basis. In this paper, we examined whether personality was a proper predictor for students' cognitive pattern in the process of scientific reasoning. More specifically, we explored the influence of personality on epistemic network in an open-ended question discussion. The dataset used in this paper came from 39 graduate students who took the course of Statistical Analysis on Educational Data. The dataset included the audio-to-text transcripts of their online voice discussion, and their responses to the Big Five personality scale. Following Fischer et al.'s conceptual framework of scientific reasoning and argumentation in educational context, eight epistemic activities (problem identification, questioning, hypothesis generation, evidence generation, evidence evaluation, generating solutions, communicating/scrutinizing, and drawing conclusions) were adopted to label the audio-to-text transcripts. The Big Five personality scale involves five dimensions: extraversion, neuroticism, conscientiousness, agreeableness and openness. The results show that the student's epistemic network is significantly different only in the conscientiousness dimension. More specifically, low conscientious students have a significantly stronger connection between evidence generation and evidence evaluation than the high conscientious ones. Whereas high conscientious students are more significantly likely to connect communicating/scrutinizing with drawing conclusions than the low conscientious ones. These findings shed light on how to group students to improve learning effectiveness.