Understanding How Students Learn from Multiple Sources: A Latent Variable Approach to Assessing Comprehension

• In Event: Virtual Poster Session
  In Virtual iPresentation Only: SIG-Cognition and Assessment Virtual Poster Session

Abstract

Today’s students encounter information from many sources—news, websites, textbooks—and must integrate it meaningfully. This study introduces and validates a new way to assess students’ ability to comprehend, evaluate, and synthesize across multiple texts. Working with 72 high school students, we used a series of comprehension tasks to test a “latent comprehension construct,” a statistical model that represents students’ deeper understanding across different types of reading tasks and topics. Despite modest sample size and task reliability, structural equation modeling results support a unified comprehension factor that generalizes across topics and correlates with reading and vocabulary skills. These findings offer a theoretically grounded approach for assessing complex comprehension, helping educators and researchers evaluate students' ability to learn from multiple sources.

Authors

• Zhaocheng Wu, Mississippi State University

• Andrew Potter, Arizona State University

• Danielle S. McNamara, Arizona State University