The Developmental Relationships Between Middle School Students’ Ability Judgments and Academic Interest: A Person-Centered Approach

• In Event: Considering the Whole Student: The Use of Person-Centered Approaches to Characterize and Study Motivation

Abstract

Research has consistently shown a decline in adolescents’ perceptions of competence and interest through the middle school years (Jacobs et al., 2002; Marsh et al., 2005; Wigfield & Eccles, 2002). Studies also evidence relatively systematic gender differences suggesting that, compared to boys, girls, in general, hold a less positive view of themselves as learners but value school more. Less is known about how the developmental trends in ability judgments and academic interest are related to each other, and how the different parameters of change are associated with achievement and gender. In order to gain a better insight into the trajectories of middle school students’ motivational development, this study builds on the framework of latent growth curve modeling. More specifically, in order to capture the potential heterogeneity that corresponds to qualitatively different development, we used growth mixture modeling, which assumes that the population of interest consists of homogenous subpopulations with varying parameters of change, and thus uses a combination of continuous and categorical latent variables (e.g., different latent growth models for each latent trajectory class).

The participants were 626 middle school students (307 girls, 319 boys), who completed a motivation questionnaire once a year during their middle school years (from 7th to 9th grades). Ability judgments were assessed with five Likert-scaled items reflecting students’ perceptions of their learning ability and sense of competence-related agency (e.g., “I have the ability to learn in school”), and general academic interest was assessed with three items focusing on students’ experiences of interest related to school and schoolwork (e.g., “In my opinion, schoolwork is really boring” [reversed]). Prior (6th grade GPA) and later (9th grade final GPA) academic achievement was also recorded.

Based on a series of general growth mixture models, a three-class solution fit the data best, thus indicating the presence of three homogenous groups of students with qualitatively different types of developmental trends. High onset, increasing ability judgments and high onset, stable interest characterized group I (46 % of participants), low onset, stable ability judgments and low onset, decreasing interest characterized group II (28%), and high onset, increasing ability judgments and low onset, decreasing interest characterized group III (26%). Those with lower level of school achievement were likely to be members of group II (low onset, stable ability judgments and low onset, decreasing interest), and boys were more likely members in group III (high onset, increasing ability judgments and low onset, decreasing interest). Prior school performance predicted initial levels of ability judgments and interest in all groups, and the trajectory class predicted later school achievement; after controlling for the effects of gender and prior GPA, students in the first group (high onset, increasing ability judgments and high onset, stable interest) had higher later GPA than students in the other groups.

In sum, the present methodological approach enabled us to identify homogenous groups of students whose developmental trends varied in terms of shape and level as well as predictive effects, thus demonstrating the importance of capturing qualitatively different developmental trajectories in students’ motivation.

Author

• Markku J. Niemivirta, University of Helsinki