Multigroup Construct Validity Evidence of the HOPE Scale: Instrumentation to Identify Low-Income Elementary Students for Gifted Programs

Abstract

Students with exceptional academic potential who come from low-income families are frequently not identified for and consequently are underrepresented in gifted and talented programs. Because of this, new means of identifying such children must be developed. This article presents the findings of exploratory and confirmatory factor analyses conducted on the HOPE Scale, a 13-item teacher-rating instrument designed to identify academic and social components of giftedness in elementary-aged students. Participants included 349 teachers who completed HOPE Scales on 5,995 ethnically and economically diverse students from three rural and two metropolitan school districts in the Midwest. Multigroup confirmatory factor analysis was also used to evaluate measurement invariance between income groups. Findings suggest a two-factor model represents good fit for the data while remaining loyal to the latent constructs of academic and social giftedness. Although showing some legitimate mean differences, invariance test results suggested equivalence of model form, factor loading, and factor variances across income groups.

Putting the Research to Use

This research has important implications for practice. Frequently, traditional measures of achievement or aptitude under-identify children from low-income families. Concerns exist about the usefulness of teacher-ratings forms or scales in gifted and talented student identification. However, when given specific items or descriptors, teachers can provide useful information concerning student performance. The HOPE Scale provides items that teachers can use to rate specific social and academic behaviors of their students. Findings from this study revealed that teaches can effectively nominate low-income students for gifted programs. Further, items on the HOPE Scale were not biased against low-income students as rated by their teachers, meaning that the social and academic scales provided similar information concerning students in either income group. However, mean scores for student from low-income families on both scales were lower than their non-low income peers, providing evidence that instruments need to be normed on the specific groups for which their use is intended. Practitioners, should not shy away from using teacher nomination instruments, but they should consider the psychometric information available concerning the use of these instruments or rating forms as used with students in demographic groups of intended use. The HOPE Scale, as developed with indicated revisions, will provide educators and researchers with a simple, psychometrically sound instrument to help with identification of underrepresented student for gifted education services.