ABSTRACT
Models of achievement goals suggest that different tasks and contexts influence the goals students adopt at a given time. However, many studies of achievement goals rely on measures assessed at the class level, analyze results with a variable-centered approach, and employ self-report questionnaires, which may reduce understanding of the contextual factors that arise as students interact with tasks. We compared a behavioral, task-level measure of middle school students' achievement goals to a self-reported, class-level measure and analyzed both measures' relations to task performance and quarterly grades using both variable-centered and person-centered approaches. Task-level goals predicted task performance but not quarterly grades, while class-level goals predicted grades but not task performance. The two measures of achievement goals were not related. This work demonstrates the importance of measuring goals at multiple levels and suggests opportunities for changing students' behaviors and achievement through motivation at the task level.
Acknowledgements
We thank Paul Ronevich and Brian Rose for their insightful feedback and assistance in developing materials and conducting this research.
Funding
This research was supported by Grant SBE00354420 from the National Science Foundation to the Pittsburgh Science of Learning Center (http://www.learnlab.org).