Future Elementary Teachers Gain Confidence in STEM, But Worries Linger

A large new study of future elementary school teachers reveals a striking paradox at the heart of science education reform: after a semester of methods coursework, aspiring teachers report substantially greater confidence in teaching integrated STEM, yet many of their anxieties persist, and some even grow. The research, published in the International Journal of STEM Education, tracked 465 preservice elementary teachers across five U.S. institutions and offers one of the most detailed pictures yet of how teacher preparation shapes, and fails to shape, the self-beliefs of the educators expected to deliver interdisciplinary science, technology, engineering, and mathematics instruction to young children.

The research team, led by Deepika Menon of the University of Nebraska-Lincoln together with colleagues at Southern Methodist University, Towson University, and Indiana University Southeast, used a mixed-methods design combining a validated self-efficacy survey with open-ended questionnaires administered at the beginning and end of science or STEM methods courses. The survey instrument, known as SETIS, measures three dimensions of confidence: personal capability, the ability to connect disciplines and engage students, and the capacity to manage materials and technology. Statistical analyses, including repeated-measures multivariate tests and paired comparisons, revealed significant gains across all three dimensions by semester’s end, with the largest improvement in teachers’ beliefs about their own abilities.

That growth matters because self-efficacy, a concept rooted in social cognitive theory, is strongly linked to whether teachers adopt innovative practices, persist through difficulties, and remain in the profession. Teachers with low confidence in STEM tend to avoid reform-based instruction and give up more easily when lessons falter. The finding that a single semester of hands-on, inquiry-based methods coursework, whether embedded in a science methods course or a dedicated STEM semester, produced medium-to-large gains suggests that teacher preparation programs can move the needle on the beliefs that ultimately shape classroom practice.

Yet the qualitative side of the study tells a more complicated story. Analyzing roughly 1,300 coded response segments, the researchers identified six broad categories of perceived challenges: teacher affect and experience, student-related concerns, content and curricular demands, pedagogical difficulties, support from colleagues and parents, and time and resources. By the end of the semester, participants reported significantly fewer overall challenges, particularly around science content knowledge and pedagogical confidence, but two concerns intensified rather than faded: worries about teaching engineering and doubts about integrating technology.

The researchers interpret this pattern through two complementary frameworks. Windschitl’s model of teacher dilemmas, spanning conceptual, pedagogical, cultural, and political tensions, helps explain how future teachers wrestle with what integrated STEM knowledge even means and how to enact it in real classrooms. Ertmer’s distinction between first-order barriers, such as limited time, materials, and institutional support, and second-order barriers rooted in personal beliefs and confidence, maps closely onto the challenges participants described. Content knowledge gaps and pedagogical uncertainty, the internal barriers, shrank after coursework. External constraints, especially time for planning and teaching STEM amid crowded elementary schedules, grew more salient as participants gained a clearer-eyed view of classroom realities.

Several findings stand out for their implications. Concerns about meeting state standards and assessment demands barely budged, with participants observing that elementary curricula leave little room for STEM when reading and mathematics dominate. One participant at a STEM integration school noted that despite the label, STEM instruction was rare. Meanwhile, structural equation modeling revealed that teachers who still felt unprepared in science or STEM content at semester’s end scored significantly lower on self-efficacy, confirming that content preparedness remains a critical lever. Intriguingly, identifying technology access as a challenge was associated with higher self-efficacy, suggesting that more confident teachers may simply be more aware of the resource constraints they will face.

The study also surfaced demographic patterns that warrant attention. Participants who had spent five or more years in college reported lower self-efficacy, possibly because their extended, full-year student teaching exposed them to a more sobering view of classroom challenges. Hispanic/Latino participants reported lower self-efficacy than their White peers, a result the authors connect to broader literature on inequities in STEM access and representation, and one they argue demands further investigation across more diverse populations.

The authors are careful about limits. Because data were self-reported at only two time points, and because most participants had not yet completed full-time student teaching, self-efficacy estimates may be somewhat inflated. The analysis also pooled data across five quite different programs, so it cannot isolate which specific course features drove the gains. Still, the breadth of the sample strengthens confidence that the overall pattern, rising efficacy alongside persistent and evolving concerns, reflects something real about how teacher preparation works.

The practical implications are pointed. The researchers recommend that preparation programs increase dedicated attention to engineering design and technology integration, precisely the areas where confidence lagged or declined. They suggest strategic school and informal-education placements so future teachers can witness successful integrated STEM instruction in action, something many mentor elementary classrooms rarely model. They also call for partnerships between universities and school administrators to address resource gaps, and for collaboration between STEM content faculty and education faculty to build better curricula. Longitudinal follow-up studies, the authors argue, should track whether the challenges these future teachers anticipate actually materialize once they enter elementary classrooms, closing the loop between preparation and practice.

Subject of Research: Preservice elementary teachers' integrated STEM teaching self-efficacy and perceived challenges

Article Title: “I learned a lot, but…”: preservice elementary teachers’ integrated STEM teaching self-efficacy and perceived challenges

Article References: Menon, D., Wieselmann, J. R., Haines, S., Asim, S., & Johnson, A. (2026). “I learned a lot, but…”: preservice elementary teachers’ integrated STEM teaching self-efficacy and perceived challenges. International Journal of STEM Education, 13(1), Article 56. https://doi.org/10.1186/s40594-026-00645-8

Image Credits: AI Generated

DOI: 10.1186/s40594-026-00645-8

Keywords: integrated STEM, self-efficacy, preservice teachers, elementary education, teacher preparation, STEM challenges, engineering education, methods coursework, mixed methods, structural equation modeling, educational psychology, K-12 STEM