Hide/Show Apps

Predicting Science Engagement with Motivation and Teacher Characteristics: a Multilevel Investigation

Kiran, Dekant
Sungur, Semra
Yerdelen, Sundus
The purpose of this study was to investigate the student and teacher-level predictors of Turkish middle school students' engagement in science classes. Students' engagement was examined in terms of agentic, behavioral, cognitive, and emotional engagement. The participants of the study were 134 Turkish science teachers and their 3394 grade 7 students. Separate multilevel models were specified for each dimension of students' science engagement. Results of the HLM analyses indicated that dimensions of students' science engagement were significantly predicted mostly by the student-level variables including science self-efficacy, mastery approach and avoidance goals, and performance approach goals. Teacher-level variables were influential only on the cognitive and emotional engagement. There were also cross-level interactions in predicting science engagement. Results were discussed in the light of related literature.