Groups in Flux: A Novel Mixed-Methods Approach for Linking Narrative Data to Quantitative Analysis

• In Event: AERA Roundtable Session Sunday 7:45 am - Gold 4
  In Roundtable Session: Behavior and Achievement Across Contexts: Analyzing the Complex Web of Influence Through Mixed Methods

Abstract

This study presents and validates a novel mixed-methods approach for linking narrative data to statistically testable constructs, improving alignment between student experiences and intervention strategies. It evaluates whether students’ interpretations of past academic challenges more effectively predict STEM motivation than demographic groupings such as gender. Undergraduates across a large university system (N = 40) were grouped into “Failure,” “Neutral,” or “Success” categories based on open-ended responses. Bayesian classification validated these groups by applying posterior probabilities to confirm the statistical plausibility of the qualitative sort. Inferential testing results showed that perception group membership significantly predicted motivation scores, while gender did not. Subsequent qualitative analysis of the perception groups revealed distinct motivational hallmarks in each group, informing the design of targeted interventions.

Author

• Camila Torres Rivera, Guttman Community College