Uncovering Learner Heterogeneity Through Latent Class Analysis: Evidence from Elementary Pupils in the Philippines

Abstract

Introduction: Learners from economically disadvantaged communities are often treated as a homogeneous population despite differences in their academic performance and household circumstances. Identifying distinct learner profiles based on academic and socio-demographic characteristics provides a more comprehensive understanding of learner heterogeneity and supports the development of targeted educational interventions.

Objectives: This study aimed to identify latent class profiles among elementary pupils based on their academic performance and household socio-demographic characteristics using Latent Class Analysis (LCA).

Methods: A cross-sectional quantitative design employing a person-centered analytical approach was utilized. The participants comprised 76 elementary pupils from a public elementary school in Margosatubig, Zamboanga del Sur, Philippines. Academic performance data and household socio-demographic information were obtained from official school and barangay records. Latent Class Analysis was performed using Latent GOLD 6.0 to identify homogeneous learner profiles based on academic performance and household characteristics.

Results: The analysis supported a five-class solution, demonstrating excellent model fit and classification quality. Five distinct latent class profiles were identified: Academically Stable Learners (38.8%), Supported Learners (22.9%), Resilient Learners (17.3%), Highly Vulnerable Learners (14.2%), and Exceptional Achievers (6.8%). The profiles revealed substantial heterogeneity among learners, showing that socioeconomic disadvantage did not uniformly correspond to lower academic achievement. While one profile exhibited multiple household disadvantages alongside lower academic performance, other profiles demonstrated consistently high academic achievement despite experiencing socioeconomic constraints.

Conclusions: The findings demonstrate the usefulness of Latent Class Analysis in uncovering meaningful learner heterogeneity that may remain undetected using conventional variable-centered approaches. The identified latent class profiles provide evidence for designing profile-responsive educational, and social interventions tailored to the diverse needs of elementary learners. Future research involving larger and more diverse populations is recommended to further validate the identified learner profiles.