Motivational Profiles of Upper-Elementary School Students: Stability and Change in Relation to Academic Engagement

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

Abstract

Prior research suggests that motivational processes combine in different ways to predict academic engagement and achievement (e.g., Braten & Olaussen, 2005; Lau & Roeser, 2008; Shell & Husman, 2008). Building on this work, we conducted a short-term longitudinal study to identify motivational profiles, examine their stability across the school year, and investigate these profiles as predictors of academic engagement. We focused on two broad categories of motivation: perceived competence and reasons for engagement (task value, achievement goal orientations). The consideration of multiple motivational profiles may help to resolve the mixed pattern of findings for some variables. Moreover, the inclusion of a variety of motivational constructs allows a more comprehensive examination of the multiple ways that students utilize motivational resources to support engagement and achievement (Pintrich, 2003).

Fifth-grade students (n=151) participated in the study during the fall and spring semesters. Students reported on their motivation and behavioral and cognitive engagement in academics; teachers also provided ratings of student engagement. Using an i-states technique, we identified motivational profiles using hierarchical cluster analysis with Ward’s method and squared Euclidean differences. A 4-cluster solution was identified as the most reasonable representation of the data across both time points. Students in cluster 1 (n1=62, n2=25), highly motivated by any means necessary, reported high levels of all five motivational indicators. The second cluster (n1=31, n2=63), intrinsically motivated and competent, included students motivated through intrinsic means (mastery goals, task value) and high perceived competence. Students in the third cluster (n1=40, n2=32), performance-focused, strongly endorsed both performance-approach and performance-avoidance goals, but had low efficacy, mastery goals, and task-value. Finally, students in the fourth cluster (n1=17, n2=31), amotivated, reported low levels of every form of motivation. Cluster membership and the relative size of the cluster shifted across time, with 59% of students shifting clusters. Cluster 1 (highly motivated) was the least stable cluster; only 25% of the students remained in this cluster from fall to spring, while about 50% remained in the other clusters. Interestingly, cluster 2 (intrinsically motivated) doubled in size from fall to spring to become the largest cluster, with students moving primarily from cluster 1 (highly motivated) or 4 (amotivated) into cluster 2.

There were significant differences in student-reported behavioral and cognitive engagement based on cluster membership at both time points. As expected, membership in the highly motivated and intrinsically motivated clusters was equally beneficial in supporting student-reported engagement, while the amotivated and performance-focused clusters had the lowest engagement. We observed a marginally significant difference in teacher-reported engagement. Teachers perceived students in the intrinsically motivated cluster as the most engaged relative to the performance-focused cluster (time 1) and the highly motivated cluster (time 2). Thus, teachers did not view students motivated by performance goals, even in combination with more adaptive forms of motivation, as willing to engage in class. Results suggest that multiple motivational profiles support academic engagement and identify shifts in motivational profiles over time. Implications for theory and practice will be discussed.

Authors

• Lisa Linnenbrink-Garcia, Duke University

• Jan J. Riggsbee, Duke University

• Nancy E. Hill, Harvard University

• Kate E. Snyder, University of Louisville

• Adar Ben-Eliyahu, University of Pittsburgh