Informative Tools for Characterizing Individual Differences in Learning: Latent Class, Latent Profile, and Latent Transition Analysis

Abstract

Highlights• learning is often multidimensional, heterogeneous, and discontinuous• traditional statistical analyses are limited in capturing this complexity• latent class and latent profile models identify subgroups of learners• latent transition models characterize discontinuous, non-linear, learning paths• these models contribute to our understanding of learning and individual differences 3 Informative Tools for Characterizing Individual Differences in Learning: Latent Class, Latent Profile, and Latent Transition AnalysisThis article gives an introduction to latent class, latent profile, and latent transition models for researchers interested in investigating individual differences in learning and development. The models allow analyzing how the observed heterogeneity in a group (e.g., individual differences in conceptual knowledge) can be traced back to underlying homogeneous subgroups (e.g., learners differing systematically in their developmental phases). The estimated parameters include a characteristic response pattern for each subgroup, and, in the case of longitudinal data, the probabilities of transitioning from one subgroup to another over time. This article describes the steps involved in using the models, gives practical examples, and discusses limitations and extensions. Overall, the models help to characterize heterogeneous learner populations, multidimensional learning outcomes, nonlinear learning pathways, and changing relations between learning processes. The application of these models can therefore make a substantial contribution to our understanding of learning and individual differences.

The relation between the control-of-variables strategy and content knowledge in physics in secondary school

“…Next, we added again students' school grade and the latent variable representing their content knowledge to the model, in order to examine whether and how students showing the different profiles differ on these variables. Adding the covariates directly into the model did almost nothing to change the level and shape patterns of the CVS profiles, which sometimes happens, particularly in smaller samples (Hickendorff et al, 2018). Similar to the confirmatory factor analysis-based model, school grade exhibited positive predictive value.…”

“…The second type of analysis was a person-centered approach, namely, a latent profile analysis (Hickendorff, Edelsbrunner, McMullen, Schneider, & Trezise, 2018). This analysis does not model a single CVS skill that overarches the four sub-skills, but rather searches for subgroups of students who show different answer patterns across the four sub-skills (Harring & Hodis, 2016).…”

“…Specifically, these methods assume linear patterns in students' achievement across the four sub-skills and that across all students these skills adhere to the same difficulty pattern (e.g., students who solve more difficult items also solve less difficult items) and thus are ill-equipped for detecting non-linear patterns in students' CVS sub-skills. Latent profile analysis on the other hand can be applied to model both linear and non-linear patterns and can reveal complex patterns across the four sub-skills that would stay hidden in the former types of analyses (Edelsbrunner & Dablander, 2019;Hickendorff et al, 2018). This analysis therefore can provide additional information over and beyond the confirmatory factor analysis.…”

“…This test has been shown to be quite reliable in simulations (Nylund, Asparouhov, & Muthén, 2007: Peugh & Fan, 2013. Furthermore, we considered that CVS profiles should each describe at least about five percent of students (Hickendorff, et al 2018).…”

Detecting Strategies in Developmental Psychology

“…An alternative approach is based on latent-class analysis (also called latent-mixture analysis; Bouwmeester and Sijtsma 2007;Bouwmeester and Verkoeijen 2012;Hickendorff et al 2018;Huizenga et al 2007;Jansen and van der Maas 2002;Nylund et al 2007a). Although latent class analysis can be used in a confirmatory way, the exploratory way is more common.…”