Dimensionality of Morphology Skills in Elementary Students

• In Event: Mapping Morphology and Its Multidimensionality, Assessment, Relations to Outcomes, and Relation to Covariates

Abstract

Purpose:
The development of morphological knowledge is important to students’ language and literacy growth. Knowledge of morphemes supports students’ reading achievement (Carlisle & Goodwin, 2014) by influencing their ability to decode and access the meanings of multisyllabic words (Carlisle, 2004), which then supports their reading comprehension (Carlisle, 2000; Nagy, Berninger, & Abbott, 2006). Recent work suggests morphological knowledge is multidimensional (Goodwin et al., 2017; Goodwin et al., 2020), and this has important implications for assessment and instruction.
However, dimensionality for elementary students is less clear. Multiple studies suggest potential multidimensionality for elementary students by showing for different morphology measures differing performance (Apel, Diehm, & Apel, 2013) or growth trajectories (Berninger et al., 2010) or relations to outcomes (Lan, Chen, & Deacon, 2020; Levesque et al., 2017, 2019). However, the single study that has examined the dimensionality of morphology for elementary students found the best fit of a unidimensional model (Spencer et al, 2015). Yet because the focus was on the relationship between morphology and vocabulary, this work did not consider different morphological dimensions, leading to the need for the current study that evaluates the dimensionality of morphological skills for 4th grade elementary students.

Method:
A 42-item morphological knowledge test intended to measure eight domains within the broader morphological construct was administered orally by researchers to 2,000 4th graders from five school districts, some rural and others urban, across the Southwest. The eight content domains were as follows: recognizing prefixes, recognizing suffixes, recognizing roots, comprehending meaning of prefixes, comprehending meaning of suffixes, comprehending meaning of roots, problem solving with prefixes, and problem solving with suffixes. Dimensionality was explored via confirmatory factor analysis, investigating the fit of six models ranging from unidimensional models, to variations on multidimensional models, to testlet models that control for nuisance latent traits related to type of task required by each item. Correlations with measures of reading were then explored.

Results:
Results indicate multidimensionality as absolute and relative fit results pointed to the best fit of the model in which three morphological skills were measured (i.e., recognition of morphemes, comprehension of morphemes, and problem solving with morphemes), while controlling for variance from the four unique item types used in the measurement tool. Results for the correlations to reading measures will be presented in the conference symposium.

Significance:
Given that morphological knowledge is important to students’ language and literacy growth, it is critical to have a deep understanding of the nature of the construct in upper elementary aged students. Such understanding can serve as foundational knowledge underlying morphological instruction, curricula, and assessments. Studying and clarifying the dimensionality of morphological knowledge for elementary students is a necessary step toward these deeper understandings of the construct.

Authors

• Tae Yeon Kwon, University of Florida

• Corinne Huggins-Manley, University of Florida

• Amanda P. Goodwin, Vanderbilt University

• Amber Elizabeth Benedict, Arizona State University - Tempe