A construct validity study of the student engagement instrument
Lovelace, Matthew Davis
MetadataShow full item record
School failure is a substantial problem in the United States, particularly for various at-risk demographics. In dropout prevention efforts, student engagement has emerged as an area of focus. While generally considered to be a behavioral, cognitive, and affective multifactor construct, studies of student engagement primarily center on its behavioral aspects. However, indicators of cognitive and affective engagement may have distinct or additive value if incorporated into multiple-factor risk models. In order to investigate such possibilities, however, the construct validity of measures of such indicators must be sufficiently supported. The present study evaluated the Student Engagement Instrument (SEI), a self-report measure of cognitive and affective engagement. The results of confirmatory factor analyses supported the use of the SEI with various at-risk populations. Furthermore, results of rate and proportion analyses suggested considerable and highly consistent associations in expected directions between extreme scores on the SEI and indicators of behavioral disengagement.