Elementary Teachers' Self-efficacy and Interest in Teaching Data Science

• In Event: Supporting Teachers to Engage in Meaningful Data Literacy and Data Science Practices Across Disciplines

Abstract

Objectives
Our research investigated elementary teachers’ self-efficacy and interest in data science after professional development (PD) aimed at teaching data science and computational thinking (CT) to elementary children. Drawing on the CT-STEM taxonomy of practices (Weintrop et al, 2016), we posed these research questions:
To what extent does data science PD impact elementary teachers' self-efficacy and interest in teaching data science aligned with CT-STEM practices?
After participating in data science PD, what do elementary teachers attribute their increased self-efficacy and interest in teaching data science to?

Perspectives
Weiland and Engeldowl (2022) emphasize the need for a comprehensive approach to build K–12 data science capacity, including curriculum development, school-based and preservice PD, and supportive policies fostering data literacy. However, research on effective data science PD remains limited. Enhancing teachers’ STEM self-efficacy is critical for improving confidence, job satisfaction, and promoting learning (National Academies, 2023); it's essential to understand how elementary teachers can develop confidence and interest in teaching data science.

Theoretical Framework
Bandura’s self-efficacy theory (1977) guided our analysis of teacher confidence and interest. The four sources of self-efficacy - mastery experiences, vicarious experiences, social persuasion, and emotional/physiological states help us explain how PD can influence teachers’ beliefs in their ability to teach data science.

Methods
We employed a mixed-methods design (Creswell & Plano-Clark, 2011) to examine how PD affects teachers’ confidence and interest in data science and CT. Using a concurrent explanatory approach, we analyzed quantitative data, followed by qualitative data to enrich interpretation and inform more effective PD strategies. To quantify their self-efficacy and interest, teachers completed surveys before and after the PD. The survey was originally developed to assess elementary school teachers’ self-efficacy and attitudes toward teaching STEM content areas (Friday Institute, 2012). We adjusted items to focus on CT and data science. Qualitative data were collected through teachers’ reflective journals and interviews. All 29 teachers wrote journal entries daily during the PD to reflect on their learning experiences.

Results/Conclusions
We observed clear gains for each cohort and for the combined group on all three constructs. Paired t-tests indicated significant gains for CT teaching self-efficacy (M = 10.1, t(26)= 5.8, p<.001, 95% CI [6.5, 13.7]), data science teaching self-efficacy (M = 8.2, t(26)= 5.3, p<.001, 95% CI [5.0, 11.4]), and CT and data science teaching interest (M = 3.3, t(26)= 2.3, p =.027, 95% CI [0.4, 6.1]). The largest gains were for CT and data science teaching self-efficacy, with average gains of 10.1 and 8.2, respectively. While CT and data science interest gains were statistically significant, they were not as large because teachers were highly interested as the PD began.

Our qualitative analysis produced four themes: feelings of mastery and success with new skills and familiar instructional practices; observing others and valuing on-site support systems; productive feedback from peers and researchers; and positive self-reflection from accomplishing beneficial tasks. We will also detail teachers’ perceived challenges and suggestions to improve the PD.

Authors

• Danielle C. Herro, Clemson University

• Jeremiah Nosakhare Akhigbe, Clemson University

• Matthew J. Madison, University of Georgia