Illustration of Step-Wise Latent Class Modeling With Covariates and Taxometric Analysis in Research Probing Children's Mental Models in Learning Sciences.

Dimitrios Stamovlasis,Georgios Tsitsipis,Themistoklis Tsikalas,George Papageorgiou,Julie Vaiopoulou

doi:10.3389/fpsyg.2018.00532

Dimitrios Stamovlasis, Georgios Tsitsipis + Show 3 more

Open Access

https://doi.org/10.3389/fpsyg.2018.00532

Copy DOI

Abstract

This paper illustrates two psychometric methods, latent class analysis (LCA) and taxometric analysis (TA) using empirical data from research probing children's mental representation in science learning. LCA is used to obtain a typology based on observed variables and to further investigate how the encountered classes might be related to external variables, where the effectiveness of classification process and the unbiased estimations of parameters become the main concern. In the step-wise LCA, the class membership is assigned and subsequently its relationship with covariates is established. This leading-edge modeling approach suffers from severe downward-biased estimations. The illustration of LCA is focused on alternative bias correction approaches and demonstrates the effect of modal and proportional class-membership assignment along with BCH and ML correction procedures. The illustration of LCA is presented with three covariates, which are psychometric variables operationalizing formal reasoning, divergent thinking and field dependence-independence, respectively. Moreover, taxometric analysis, a method designed to detect the type of the latent structural model, categorical or dimensional, is introduced, along with the relevant basic concepts and tools. TA was applied complementarily in the same data sets to answer the fundamental hypothesis about children's naïve knowledge on the matters under study and it comprises an additional asset in building theory which is fundamental for educational practices. Taxometric analysis provided results that were ambiguous as far as the type of the latent structure. This finding initiates further discussion and sets a problematization within this framework rethinking fundamental assumptions and epistemological issues.

Highlights

Research on children’s mental conceptions of everyday reality, before they acquire the science view, is an interdisciplinary area where psychometrics take the dominate role providing sophisticated tools to access these intangible entities or latent variables in question
The one-step latent class analysis (LCA) besides the disadvantages mentioned in a preceding section (Vermunt, 2010) has a weak point associated with the external variables
The nature of children’s mental model and the role of psychometric predictors, are crucial issues, for which state-of-the-art and specialized methodologies are demanded for building contemporary theories for learning and development, and what is more for the pedagogical practices

Summary

Introduction

Research on children’s mental conceptions of everyday reality, before they acquire the science view, is an interdisciplinary area where psychometrics take the dominate role providing sophisticated tools to access these intangible entities or latent variables in question. Similar enquiry subsists in psychopathology research, where, considering latent disorders as continua or discrete kinds has important implications for diagnosis procedures and treatment (Rezai et al, 2010; Edens et al, 2011; Haslam et al, 2012) The challenge at this juncture concerns the measurement theory, which studies such latent variables by means of mathematical structures and suggests the proper formalism and modeling procedures (e.g., Rust and Golombok, 2009; Trendler, 2009). If the children knowledge, reflected by certain mental models, was directly observable, this categorization process would be easy and straightforward Even though this does not happen, psychometric theory postulates that this difficulty could be overcome if these latent entities are assumed to be responsible for behaviors that are observable (Markus and Borsboom, 2013). These misclassification probabilities can be used to correct the W - relationship to get the relationship between X and (Bolck et al, 2004)

Objectives

Methods

Results

Conclusion