Play-Based Physics Training Boosts Early Childhood Educators’ Confidence

When a four-year-old watches a marble race down a track and asks why it went faster the second time, an educator stands at a crossroads. That moment can become a rich physics lesson or fizzle into a vague reassurance. New research from Germany and Switzerland suggests that a targeted professional development program can tip the balance toward the lesson, at least in the minds of the educators themselves. The study, published in Frontiers in Education, examined how participation in the PhysikSPIEL program changed early childhood educators’ self-reported professional competencies, and its findings carry implications for how the early childhood workforce might be equipped for the science classrooms of tomorrow.

The research team, led by Michaela Gelewski, Simone Lehrl, and Stefanie Schnebel at the University of Education Weingarten, focused on a persistent gap in early science education. Decades of evidence show that experiences with science in early childhood play a central role in fostering children’s scientific knowledge and inquiry skills, and that young children frequently encounter physical phenomena such as movement, balance, light, sound, and material properties in their everyday lives. Yet many early childhood educators report limited confidence and knowledge in science, with the challenges particularly pronounced in physics, a domain often perceived as abstract and difficult to translate into developmentally appropriate learning experiences. Professional development is widely regarded as a key strategy to address this deficit, but empirical evidence on play-based physics programs specifically has remained scarce.

PhysikSPIEL was designed to close that gap. The program, which ran in approximately three-month cycles in 2024 and 2025, is grounded in the concept of play-based science learning within Guided Play, a framework in which children’s agency is combined with an intentionally structured learning context. In this approach, the play remains largely child-directed, but educators deliberately create the conditions for learning through the selection of materials, the arrangement of the environment, and pedagogical prompts. The program’s designers emphasize that play-based science learning extends beyond hands-on manipulation of materials to include minds-on engagement, meaning observing, questioning, comparing, reasoning, and developing emerging explanations. Children’s independent exploration alone does not ensure meaningful science learning; educators must recognize emerging opportunities and connect with children’s ideas and questions.

Structurally, the program combined three half-day sessions with a six-week implementation phase. The first two sessions centered on hands-on exploration of physics phenomena related to magnetism and forces and motion, with participants generating questions, exploring possible explanations, and reflecting on how children could be supported in play-based contexts. The selected concepts were not treated as content to be transmitted through predefined experiments but as scientific perspectives that could be connected to children’s experiences and emerging ideas. During the implementation phase, participants designed and implemented a play-based learning environment in their own settings, supported by a resource called the MiniBook, which presented six core physics concepts in accessible language along with prompts and reflective questions. Participants also had the opportunity to be observed by a PhysikSPIEL trainer, followed by a coaching session focused on joint analysis of the implemented learning situation. The final session centered on collective reflection through structured group discussions and video-based examples.

The study assessed five competence facets organized into two dimensions: motivational orientations, comprising domain-specific self-efficacy expectations, self-efficacy expectations regarding sustained shared thinking, domain-specific academic self-concept, and domain-specific enthusiasm; and pedagogical beliefs, represented by co-constructive teaching and learning beliefs. Sustained shared thinking, or SST, is an interactional format in which two or more individuals work together in an intellectual way to solve a problem, clarify a concept, or extend a narrative, with both parties contributing to the thinking. Previous research, including the EPPE study involving approximately 3,000 children and the subsequent REPEY study, identified SST as a key feature of high-quality educational interactions that promote children’s cognitive development. The researchers hypothesized that participation in PhysikSPIEL would be associated with increases in all five self-reported facets.

Participants were 42 early childhood educators from Germany and kindergarten teachers from Switzerland, drawn from an initial pool of 69 registrants, with only complete and matched pre-post datasets included in the analyses. The final sample had a mean age of 45.16 years, ranging from 23 to 62. Data were collected using a standardized self-report questionnaire administered immediately before and after the program, with all competence dimensions assessed on four-point Likert scales. The scales showed acceptable to high internal consistency, with Cronbach’s alpha ranging from 0.74 for the co-constructive beliefs scale to 0.90 for the domain-specific self-efficacy scale. The SST self-efficacy measure consisted of only two items, which showed a strong interitem correlation of r = .76, and the researchers noted that it captures perceived capability to initiate SST rather than the broader range of interactional processes involved.

The results revealed a differentiated pattern of change. Domain-specific self-efficacy expectations showed the largest increase, rising from a mean of 2.83 to 3.25 on the four-point scale, a difference of 0.42 points, with similar patterns in both countries. Mixed ANOVA analyses confirmed a statistically significant time effect for this scale, F(1,39) = 73.07, p < .001, with a partial eta squared of .65, indicating a substantial change. Self-efficacy expectations regarding SST also increased significantly, from 3.19 to 3.46, F(1,39) = 13.02, p < .001, with a partial eta squared of .25. Domain-specific academic self-concept rose from 2.67 to 2.99, F(1,21) = 17.17, p < .001, with a partial eta squared of .30, and domain-specific enthusiasm increased from 3.29 to 3.44, F(1,40) = 6.19, p = .017, with a smaller but significant partial eta squared of .13. In contrast, co-constructive teaching and learning beliefs remained largely stable, shifting only from 3.27 to 3.19, F(1,38) = 1.26, p = .270, with a negligible partial eta squared of .03.

Correlational analyses added a structural dimension to these findings. The motivational orientations were strongly interrelated, forming what the researchers describe as a motivational subsystem within professional competence. The strongest association emerged between domain-specific self-efficacy expectations and self-efficacy expectations regarding SST, r = .60, p < .001, which appears conceptually coherent given that the latter represents a domain-specific application of the broader self-efficacy construct. Domain-specific academic self-concept and enthusiasm were also strongly related, r = .56, p < .001. Notably, co-constructive teaching and learning beliefs showed no significant associations with most other variables, including only a weak, non-significant correlation with SST self-efficacy, r = .23, p = .15. This pattern provides empirical support for treating generalized pedagogical beliefs and domain-specific self-efficacy judgments as distinct facets, even though both concern dialogic, extending interactions with children.

The differential pattern across dimensions is theoretically meaningful. The researchers argue that motivational orientations may be more sensitive to short-term, practice-oriented professional development than pedagogical beliefs, which are deeply embedded, experience-based dispositions that change slowly over time. The program’s hands-on exploration sessions provided mastery experiences and direct engagement with physics content, mechanisms that social-cognitive theory identifies as primary sources of self-efficacy. The implementation phase, with its observation and coaching components, provided situated feedback on educators’ own enactment of dialogic interactions, which was particularly relevant to SST self-efficacy. By contrast, co-constructive beliefs were addressed more indirectly through collaborative reflection, and the researchers suggest that such broader pedagogical orientations may require more sustained professional learning experiences to change, particularly when baseline levels are already high.

The authors are careful to frame these findings as preliminary rather than definitive. The pre-post design without a control group precludes causal conclusions about program effects, the relatively small and self-selected sample limits statistical power and generalizability, and the reliance on self-report measures leaves open the possibility of social desirability bias and discrepancies between perceived and enacted practice. Deviations from normality were also detected for some variables, particularly the SST scale. Participants who chose to enroll in a physics-related program may have been comparatively motivated from the outset, and the observed changes reflect self-assessed perceptions rather than directly observed behavior. Future research, the team suggests, should employ larger and more diverse samples, include control groups, and combine self-report data with observational and performance-based measures to clarify how self-assessed competencies relate to professional practice.

Even with those caveats, the study offers a technically grounded contribution to a field where evidence has been thin. It demonstrates that different facets of professional competence vary in their responsiveness to practice-oriented professional development, and that motivational orientations, the conditions considered particularly relevant for fostering scientific interactions in early childhood education, can shift within a roughly three-month window. As countries increasingly recognize early science education as a foundation for later STEM learning, programs like PhysikSPIEL suggest a practical pathway: equip the adults who guide children’s play with the confidence and enthusiasm to see physics in a marble run, a magnet, or a balancing act, and the science may follow the child into the classroom and beyond.

Subject of Research: Changes in general and physics-related professional competence facets of early childhood educators

Article Title: Changes in general and physics-related professional competence facets of early childhood educators

Article References: Gelewski, M., Lehrl, S., & Schnebel, S. (2026). Changes in general and physics-related professional competence facets of early childhood educators. Frontiers in Education, 11, Article 1923768. https://doi.org/10.3389/feduc.2026.1923768

Image Credits: AI Generated

DOI: 10.3389/feduc.2026.1923768

Keywords: Changes, general, physics-related, professional, competence, facets, early, childhood, educators, scientific research