Empirical Research: Emerging Research: Using Automated Detectors to Examine the Relationships Between Learner Attributes and Behaviors During Inquiry in Science Microworlds

点击翻译按钮获取中文摘要

The research study will examine the relationship between student attributes and behavior while working through a series of science projects on computer. This project seeks to advance understanding of how middle school student behaviors within using educational software influence science learning. The investigators will measure how student attributes such as poor goal orientation, low self-efficacy, and low perseverance are associated with learning outcomes. They will investigate how these attributes lead to negative behaviors such as gaming the system rather than learning the material, off-task behavior, haphazard inquiry, and help avoidance. The study will use survey measures of student attributes, automated detectors of student behavior during use of educational software, and measures of student learning to develop models of how different student behaviors lead to improved learning outcomes. The models are chosen to test which behaviors mediate the relationships between student attributes and learning and how the behaviors do so. The researchers will use statistical models, such as path analysis, to test their hypotheses. The study will be conducted in 500 ethnically and economically diverse middle school students who are part of a science assessment project in those schools. The findings are intended to serve as a basis for creating new educational software and classroom practices that identify in real-time students with low grit, low self-efficacy, and non-productive goal orientation. Through research with diverse populations, and in a domain where many American students are falling behind, the project has the potential to benefit learners currently disenfranchised from science learning.