Filling a gap: initial evidence for reliable and valid measures of students' self-concept, self-efficacy, and interest in science with elementary students traditionally underrepresented in STEM

American students continue to perform poorly on national and international assessments of Science, Technology, Engineering, and Mathematics (STEM) competencies, and achievement gaps spanning racial/ethnic and socioeconomic lines emerge early and widen over time. Scholars and practitioners agree that expanding access to high-quality STEM education has the potential to improve students' performance and reduce inequalities. Research has elucidated the critical role that students' self-perceptions play in driving academic achievement, which has spurred the development of many educational programs and initiatives aimed at increasing students' confidence, self-efficacy, and interest. However, our capacity to determine what programs and initiatives are effective and for whom is limited by our lack of psychometrically sound measures that assess science-related self-perceptions and interests of elementary students from diverse populations. To address this gap, we developed and tested measures of Science Self-Efficacy, Science Self-Concept, Interest in Science Activities, and Attitudes toward STEM careers in a sample of third-grade students (8-9 years) traditionally underrepresented in STEM careers (94% of our sample identified as either African American or Hispanic). We present initial evidence, from a preliminary pilot study, for the reliability and validity of these measures and reveal the multi-dimensional nature of students' self-perceptions and interests in science. We discuss how such measurement tools will inform our understanding of the nature of young students' science self-perceptions, how the utilization of such tools can inform educational practice, and highlight the critical importance of conducting measurement development research with diverse populations.