Impact of Scientific Inquiry-Based Activities on STEM Interest in Croatian Primary Schools

Abstract

This research examines the impact of practical work and the use of scientific research methods when teaching ten-year-old students interested in the fields of science, technology, engineering, and mathematics (STEM). An increased interest in STEM is encouraging, as it may carry through to higher education institutions and potentially increase students’ ability to contribute to improving Croatian society. The program “Sa STEMom raSTEMo” was approved by the Ministry of Sciences and Education for researching and improving didactic work in STEM fields in elementary schools. A total of 650 participants from four classes in five primary schools were selected. Half of the participants formed the intervention group in which the “Sa STEMom raSTEMo” program was implemented for three months, and the other half formed the control group. Before and after the experimental intervention, a survey to determine interest in STEM fields was conducted in both groups, and the results were compared to verify the impact of the implemented forms of teaching. A questionnaire was then constructed and tested through a pilot study; its understandability and reliability were measured, as well as the validity of the applied measurement scales. Furthermore, a survey of interest in STEM fields was conducted three months after the intervention. All results were analyzed and compared. The results showed that implementing practical work and using scientific research methods in classroom teaching increases students’ interest in STEM. In general, no statistically significant differences in interest in STEM were observed between girls and boys aged 10, and no relevant gender differences were observed in 10-year-olds who participated in the program.

1. Introduction

STEM education (science, technology, engineering, and mathematics) is increasingly recognized as a key tool for developing scientific literacy and fostering long-term interest in scientific fields (Honey et al., 2014; English, 2016). Despite its growing prominence, research consistently shows that students’ enthusiasm for STEM subjects begins to wane in the later years of primary education, with girls particularly affected by this early decline (Maltese & Tai, 2011; Olive et al., 2022). This trend highlights the urgency of implementing early interventions that not only introduce STEM content but also nurture motivation and identity within these domains.

Inquiry-based and integrative pedagogical approaches have shown promise in reversing this decline. Z. H. Wang et al. (2023) emphasize that connecting STEM disciplines through real-world problem solving enhances conceptual understanding and promotes long-term engagement. However, mathematics often remains underrepresented in such interventions, with science and technology receiving disproportionate attention (Park et al., 2020; English, 2017). This disciplinary imbalance is particularly relevant when evaluating programs that aim to foster holistic STEM interest.

Building on this body of research, the present study introduces a short-term, inquiry-based STEM intervention implemented in Croatian primary schools. The study investigates changes in student interest across STEM domains, with a focus on gender differences and the sustainability of motivational effects. While prior research has explored the benefits of educational robotics (Papadakis et al., 2022), design thinking (Santos et al., 2025), and interdisciplinary learning (Yim et al., 2024), few studies have examined how these approaches interact with students’ psychological needs in early education.

To conceptually frame the intervention, we draw on Self-Determination Theory (SDT) (Deci & Ryan, 1985), which posits that intrinsic motivation flourishes when learners experience autonomy, competence, and relatedness. These psychological needs are especially pertinent in STEM contexts, where students must feel empowered to explore, solve problems, and collaborate meaningfully (Ryan & Deci, 2020; Chiu, 2023). Although SDT did not explicitly guide the initial program design, post hoc analysis revealed that many activities—such as open-ended experimentation and peer collaboration—naturally aligned with its principles. Future iterations may benefit from embedding SDT constructs more deliberately into both design and evaluation frameworks.

This study also contributes to the ongoing discourse on gender and STEM engagement. While some studies report persistent gender gaps in motivation and aspiration (M.-T. Wang & Degol, 2017), others suggest that inclusive pedagogies can foster equal or even greater interest among girls in certain STEM domains (Olive et al., 2022). By analyzing gendered responses to a structured intervention, this research adds nuance to our understanding of how early experiences shape STEM trajectories and identity formation.

The importance of fostering STEM interest from early childhood is well-documented. Attitudes, motivation, and self-confidence related to STEM are often formed before adolescence and significantly influence future academic and career choices (Archer et al., 2013; Maltese & Tai, 2011). Yet many children, especially girls, encounter barriers such as limited exposure, prevailing stereotypes, and a lack of relatable role models, which contribute to disengagement.

Active learning and student-centered pedagogies have shown promise in enhancing STEM interest at the primary level. However, most existing programs are conducted outside formal classroom settings, leaving a gap in understanding their effectiveness within structured school environments. Moreover, few studies have simultaneously tracked emotional responses alongside motivational changes, limiting the holistic evaluation of intervention outcomes.

To address these gaps, this study designed, implemented, and evaluated a structured STEM program tailored to fourth-grade students. Grounded in motivational theories (Ryan & Deci, 2000), emotional engagement models, and pedagogical principles that support autonomy, competence, and curiosity, the program aimed to foster positive emotions toward STEM subjects and empower students through hands-on, interdisciplinary exploration.

Recent findings underscore the value of integrated approaches such as STEAM in early education—not only for stimulating scientific curiosity but also for enhancing foundational academic and language skills (Aktulun et al., 2024). These insights support the rationale for implementing inquiry-based programs within formal curricula.

The Current Study

Building on the theoretical and empirical insights, this study aimed to design, implement, and evaluate a structured STEM program for fourth-grade primary school students in Croatia. The program “Sa STEMom raSTEMo” was developed in accordance with principles of active learning and motivational theories, with a particular focus on fostering curiosity, autonomy, and competence through interdisciplinary activities.

A quasi-experimental design was employed, involving both control and intervention groups, and program effects were assessed at three time points: before the intervention, immediately after, and three months later. The study included 650 students from five primary schools. The questionnaire comprised scales measuring interest, motivation, self-efficacy, and emotional responses to STEM content. The goal was to examine changes in STEM interest, with particular attention to gender differences and the sustainability of intervention effects.

The “Sa STEMom raSTEMo” program was implemented over a three-month period, with weekly workshops lasting approximately 90 min. The program was designed to foster curiosity, autonomy, and competence through interdisciplinary STEM activities, integrating natural sciences, mathematics, technology, and computer science.

Each workshop followed a thematic structure, progressing from simple observations to complex inquiry-based projects. Activities were designed to be age-appropriate, engaging, and aligned with motivational principles.

Illustrative Activity Example: “The Chemistry of Air”. In one session, students explored the composition and properties of air. After a brief introduction, they conducted an experiment to produce and identify oxygen using manganese dioxide and hydrogen peroxide. Students observed the reaction, collected gas in test tubes, and tested for oxygen using a glowing splint. They then discussed the role of oxygen in combustion and compared results with everyday phenomena (e.g., why candles go out in closed jars). This activity encouraged the following:

• Autonomy: Students chose materials and predicted outcomes.
• Competence: They applied prior knowledge and interpreted results.
• Relatedness: Collaboration in small groups fostered peer learning.

Other activities included programming Micro:bit devices, exploring the solar system using digital simulations, and conducting biological observations with microscopes.

By providing empirical evidence on the efficacy of structured STEM programming in primary education, this study enriches the existing literature and offers practical implications for curriculum development and educational policy. The findings advocate for the early integration of emotionally responsive, motivationally grounded STEM interventions within formal education systems.

The following hypotheses were set based on the statements in the Introduction and the defined research objectives.

Hypothesis 1. 

Conducting practical work and using scientific research methods in classroom teaching will increase students’ interest in STEM fields.

Hypothesis 2. 

There is no gender difference in interest in STEM fields between ten-year-old girls and boys, in general.

2. Materials and Methods

2.1. Study Design and Ethical Considerations

A structured survey comprising 23 items was developed for the purpose of this study (Appendix A). Prior to the commencement of the research, ethical approval was obtained from all participating institutions. The study design adhered to ethical standards, ensuring the professional execution of the research and the validity of its findings. Participation was entirely voluntary, and informed consent was obtained from all respondents.

2.2. Participants and Group Allocation

The study involved 650 students from randomly selected classrooms across five elementary schools. Participants were evenly divided into control and intervention (experimental) groups. Baseline data were collected via questionnaires administered to both groups prior to the intervention. Follow-up surveys were conducted immediately after the intervention and again three months later to assess both immediate and delayed effects on students’ interest in STEM disciplines.

2.3. Program Description “Sa STEMom raSTEMo”

The intervention, titled “Sa STEMom raSTEMo,” was implemented over a three-month period. Each week, students in the experimental group participated in a 90-min workshop led by external STEM educators. Classroom teachers supported the sessions by facilitating logistics and ensuring continuity. All facilitators underwent introductory training to standardize program delivery across schools.

The program was thematically structured, progressing from basic observational activities to complex inquiry-based projects. Students engaged in a variety of hands-on experiences, including: Using microscopes and basic chemical apparatus; Conducting experiments to detect and confirm the presence of oxygen; Performing Bilstein reactions; Programming with Micro:bit; Exploring scientific concepts via the mozaBook digital platform, including the anatomy of the ear, auditory processes, and cosmic phenomena such as the origin of the universe and the Solar System; Investigating air composition, magnesium combustion, and chemical transformations.

The overarching goal was to foster active participation, independent thinking, and collaborative problem-solving among students.

The control group followed the standard national curriculum in science and mathematics without any additional enrichment or inquiry-based activities. Instruction was delivered by classroom teachers with equivalent professional qualifications to those of the intervention facilitators, in accordance with Croatian educational regulations.

Although formal fidelity monitoring was not conducted, all schools adhered to a shared program outline and timeline. Facilitators maintained regular communication with the research team to ensure alignment and exchange of best practices.

2.4. Data Analysis

Survey responses were statistically analyzed to evaluate changes in STEM interest. Frequencies, arithmetic means, and standard deviations were calculated for all numerical items. Gender distribution was expressed as percentages, and final academic grades were reported as frequencies. Descriptive statistics were used to characterize the sample.

To assess differences between groups, independent samples t-tests and McNemar’s tests were employed. A mixed-design Analysis of Variance (ANOVA) was conducted to examine changes in STEM interest across three time points: baseline, post-intervention, and follow-up. Time was treated as a within-subject variable, while group and gender were treated as between-subject variables. Where significant effects were identified, post hoc analyses (e.g., Tukey’s HSD) were performed to determine specific group differences. ANOVA was selected for its robustness in comparing means across multiple groups and time intervals.

3. Results

A multi-day STEM workshop program was implemented in 2023, and participants (N = 650) were repeatedly surveyed using a questionnaire. The gender structure of the respondents was analyzed; interest in STEM fields was compared in the control and intervention groups before implementation of the program (N_I = 320; N_C = 330); and potential gender differences in interest in STEM fields were examined before the intervention. The control and intervention groups were compared immediately and three months after the intervention. Gender differences were also analyzed during each examination.

3.1. Gender Structure of the Sample

The study included 650 participants: 312 boys (48%) and 338 girls (52%). There were 166 girls and 154 boys in the intervention group and 171 girls and 159 boys in the control group.

3.2. Overview of Analytical Strategy

After identifying no statistically significant difference between the intervention and control groups before the intervention, a multi-day STEM program called “Sa STEMom raSTEMo” was implemented with students in the intervention group. Differences between the groups were investigated immediately and at three months after the program.

The primary analyses were conducted using mixed-design ANOVA and linear mixed models to assess changes in STEM interest across three time points (pre-intervention, post-intervention, and three-month follow-up), with group (intervention vs. control) and gender as between-subject factors. These models allow for the examination of both main effects and interaction effects, accounting for repeated measures and individual variability.

3.3. Main Effects and Interactions

The repeated measures ANOVA revealed that time had a statistically significant main effect on STEM interest scores (F (2, 1278) = 73.65, p < 0.001), indicating that students’ interest levels changed meaningfully across the three measurement points. Post hoc analyses (Bonferroni-corrected) showed a significant increase in interest from baseline (M = 4.01, SD = 0.39) to immediately after the intervention (M = 4.31, SD = 0.23; p < 0.001), with a further increase observed three months later (M = 4.42, SD = 0.20; p < 0.001).

A significant interaction effect between time and group (F (2, 1278) = 58.21, p < 0.001) suggests that the intervention group exhibited a stronger upward trend in STEM interest over time than the control group, whose scores remained relatively stable (M = 4.01 → 4.02 → 4.07).

Linear mixed model analysis confirmed these findings, with the time × group interaction reaching statistical significance (β = 0.35, SE = 0.04, t = 8.75, p < 0.001). The random intercepts for participant ID accounted for within-subject variability (ICC = 0.72), strengthening the reliability of the longitudinal design (Figure 1).

These results support Hypothesis 1 and demonstrate that participation in the “Sa STEMom raSTEMo” program leads to a sustained and statistically significant increase in STEM interest among 10-year-olds.

The findings related to the delayed post-intervention measurement, conducted three months after program completion, provide support for Hypothesis 1. Students in the intervention group continued to exhibit elevated levels of interest in STEM subjects, with mean scores surpassing those recorded immediately after the intervention. Although the increase was not statistically significant between the two post-intervention time points, the consistent upward trend in STEM interest demonstrates sustained engagement. These results confirm Hypothesis 1, indicating that the impact of inquiry-based and practical STEM activities was not transient but, rather, contributed to enduring interest among primary school students.

3.4. Gender Differences

Statistical analysis of differences based on gender was performed in the control and intervention groups. No statistically significant difference was observed.

Since no gender difference was observed in the responses to the questions in the sample of 320 respondents who participated in the multi-day educational program “Sa STEMom raSTEMo”, Hypothesis 2 is confirmed, i.e., there is no difference in interest in STEM fields between girls and boys who participated in the program.

A three-way interaction (time × group × gender) was significant (F (2, 32) = 4.34, p = 0.021), suggesting that the intervention had a particularly strong effect on female students. Girls in the intervention group exhibited the steepest growth in interest, with scores increasing consistently across all time points.

Boys in the intervention group showed moderate gains, while both genders in the control group maintained near-baseline scores throughout the study period. Error bars represent ±1 standard deviation (Figure 2).

3.5. Differences in Interest in STEM Fields Examined Using a Mixed-Design ANOVA (Time × Group, Time × Gender, and Time × Group × Gender)

The graph below (Figure 3) displays average interest scores across three time points (before intervention, immediately after, and three months later), comparing intervention and control groups and distinguishing between male and female participants. This figure illustrates a significant increase in STEM interest among participants in the intervention group, particularly among girls, with effects sustained at follow-up.

Table 1. Summary of mixed-design ANOVA results examining changes in STEM interest over time.

Effect/Interaction	F-Value	p-Value	Interpretation
Main effect of time	F (2,32) = 19.61	p < 0.001	Significant change in interest over time
Main effect of group	F (1,32) = 899.10	p < 0.001	Intervention group shows significantly higher interest
Main effect of gender	F (1,32) = 305.10	p < 0.001	Girls show higher interest than boys
Time × Group interaction	F (2,32) = 21.83	p < 0.001	The effect of the intervention varies over time
Time × Gender interaction	F (2,32) = 0.69	p = 0.508	No significant difference in change over time between genders
Time × Group × Gender interaction	F (2,32) = 4.34	p = 0.021	The effect of the intervention over time differs by gender

F-value—indicates the ratio between the variance between groups and the variance within groups. p-value—indicates the probability that the observed results occurred by chance.

presents F-values and p-values for the main effects and interactions.

The results of this study provide compelling evidence that targeted STEM interventions can significantly enhance students’ interest in science and technology, especially among girls. The mixed-design ANOVA revealed that time had a strong main effect, indicating that interest levels changed significantly across the three measurement points. More importantly, the significant interaction between time and group confirms that these changes were driven by the intervention itself, rather than natural fluctuations or external factors. Girls in the intervention group showed the greatest increase in interest, which continued to rise even three months after the program ended. This suggests not only immediate engagement, but also a lasting motivational impact. The sustained interest may be attributed to the hands-on, inquiry-based nature of the activities, which aligned well with principles of self-determination theory—supporting autonomy, competence, and relatedness.

The significant three-way interaction (time × group × gender) further highlights that the intervention was particularly effective for female students. This finding aligns with previous research indicating that early exposure to STEM in supportive environments can counteract gender stereotypes and foster long-term engagement.

In contrast, the control group showed no meaningful change, underscoring the importance of structured and intentional programming in cultivating STEM interest. The minimal variation among control participants suggests that standard curricula may not be sufficient to stimulate enthusiasm for STEM fields, especially among younger students.

3.6. Impact of the Intervention

In addition to statistically significant group differences (p < 0.001), the calculated effect size using Cohen’s d (d = 1.75) indicates the large practical impact of the intervention. This suggests that the “Sa STEMom raSTEMo” program had a substantial influence on students’ interest in STEM domains, well beyond what might be expected by chance.

3.7. Supplementary Analyses

Item-level t-tests comparing individual survey items across groups and time points are comprehensively presented in the Supplementary Materials (Tables S1–S12; Figures S1–S6). These analyses provide critical granularity, enabling a nuanced understanding of the intervention’s effects beyond aggregate trends. By examining each item in isolation, the supplementary data reveals subtle shifts in attitudes, perceptions, and engagement that may otherwise remain obscured in summary statistics. This level of detail is essential for interpreting the full scope and impact of the educational program and offers valuable insights into future research and practice.

4. Discussion

The findings of this study underscore the importance of early, inquiry-based STEM education in fostering sustained interest among primary school students. Through the implementation of the “Sa STEMom raSTEMo” program, students were exposed to hands-on, interdisciplinary activities that supported key motivational constructs such as autonomy, competence, and relatedness—principles rooted in self-determination theory (Deci & Ryan, 1985; Ryan & Deci, 2020). These elements proved effective in enhancing both cognitive and emotional engagement, as reflected in the longitudinal data collected across three time points.

Importantly, the intervention aligns with broader educational goals outlined by international frameworks such as the Programme for International Student Assessment (PISA). According to OECD (2023), scientific literacy involves the ability to evaluate and plan scientific investigations—skills that were actively developed through the program’s emphasis on experimentation and inquiry. Similarly, mathematical literacy, defined as the capacity to apply mathematical reasoning in real-world contexts, was fostered through activities such as Micro:bit programming and data interpretation. Although these components were not always explicitly framed as mathematics, they contributed to the development of transferable analytical skills.

By situating our findings within the PISA framework, we highlight the relevance of early STEM interventions in preparing students for future academic challenges and global assessments. This connection is particularly pertinent given Croatia’s performance in the 2022 PISA cycle, where students scored below the OECD average in mathematical literacy (NCVVO, 2023). The results of our study suggest that integrating inquiry-based STEM activities into formal curricula could serve as a strategic response to these national trends.

Beyond international benchmarks, the study also contributes to ongoing discourse on pedagogical practices in Croatian schools. Previous research has emphasized the limitations of traditional, content-driven instruction, which often fails to engage students as active participants in the learning process (Rocard et al., 2007; Osborne & Dillon, 2008). Our findings support this critique, revealing that many students had not previously used microscopes or conducted experiments—despite curricular provisions for such activities. This gap between curriculum and practice reflects broader systemic challenges and underscores the need for professional development and instructional support.

The observed increase in student interest, particularly among girls, further validates the program’s inclusive design. While gender differences in STEM engagement are well-documented (M.-T. Wang & Degol, 2017), our results align with studies suggesting that early exposure to inclusive, hands-on learning can mitigate these disparities (Olive et al., 2022; Burušić & Šerić, 2015). The absence of significant gender differences in our sample reinforces the potential of structured interventions to promote equity in STEM education.

Moreover, the sustained interest observed three months post-intervention suggests that the program’s impact was not merely immediate but enduring. This delayed effect may be attributed to the reflective and emotionally resonant nature of the activities, which encouraged students to continue engaging with STEM topics beyond the classroom (Aktulun et al., 2024; Dou et al., 2019). Such findings echo earlier research indicating that attitudes toward science are largely shaped between the ages of 10 and 14 (Ormerod, 1971; Lindahl, 2007; Tai et al., 2006), reinforcing the importance of timely interventions.

Braš Roth et al. (2008) indicate that students who perform creative and independent science-related activities more often (e.g., experiments, student research papers, watching science shows) achieve better results in science literacy.

To increase the STEM literacy of our students, contemporary curricula and teaching of these subjects should be oriented toward inquiry-based learning (Anderson et al., 2002; Abd-El-Khalick et al., 2004; EURYDICE, 2006). Ristić Dedić (2010) emphasized that student participation in inquiry-based activities is an educational goal in the natural sciences and that a shift from a predominantly deductive methodology to an inductive, inquiry-based approach, which enables students to use and develop a wider spectrum of skills, increases students’ curiosity and interest in science, and leads to better educational achievements, is necessary.

Curricula, or documents that prescribe the framework for learning, are only partly to blame for the unsatisfactory success or decreased student interest in particular subjects or areas. There is a gap between educational theory reflected in curriculum goals and everyday school practice. Rocard et al. (2007) and Osborne and Dillon (2008) pointed out that, in practice, the dominant approach in science education is content-oriented and is still limited mainly to the transmission of knowledge.

When experiments are performed in class, they do not place the student in the role of an authentic researcher but are conducted in a directed manner, according to a protocol, and with the main goal of confirming or illustrating the teacher’s claims (Chinn & Malhotra, 2002). This is supported by the results of this research, which found that most of the children had not used a microscope before the implementation of the “Sa STEMom raSTEMo” program, and many had not even had the opportunity to conduct experiments. Our research also showed that many teachers do not perform practical work or assign students tasks that include the performance of practical work, and that the traditional teaching approach dominates. However, based on a comparison of the frequencies of responses to questions 7 and 8 on the use of microscopes, we can conclude that a small number of teachers include microscopy procedures in their teaching process. Bahar (2003) and Lavonen et al. (2005) also encouraged the diversification of teaching methods through their research. They have shown that students in science classes want to experience a greater variety of teaching methods, such as the use of concept maps, advanced knowledge organizers, debates, brainstorming, guest lectures, and educational visits, providing them with access to original scientific information and a better understanding of its importance and application in everyday life. The participants in this research expressed a greater desire to carry out practical work and experimental activities that stimulate their curiosity, and they actively participated in such activities by participating in the multi-day “Sa STEMom raSTEMo” educational program.

Stokking (2000) has pointed out that experiencing and learning about science outside of school directly impacts overall interest in science and natural sciences. This is supported by the results of this study, which found a statistically significant difference in the responses of the intervention and control groups after the intervention to questions that tested the desire to carry out practical work and experimental methods, to claims that knowledge from STEM fields will be useful to them in life, and to questions that tested interest in studying phenomena and processes in STEM fields.

Gardner and Tamir (1989) believe that interest refers to an individual’s choice to engage in one activity rather than another. When we are interested in a phenomenon or activity, we tend to devote attention and time to it. Schreiner and Sjøberg (2005) believe that when a student is interested, they develop a close relationship with the material, and this form of learning leads to deeper understanding. Student interest in a subject is also influenced by gender and developmental age, as well as how the student and his/her immediate environment perceive the importance of the subject, especially for a future career. Boys are more likely to pursue science-oriented careers, while girls are more likely to have intrinsic interests (Gardner, 1985). According to the sample of 320 respondents who participated in the multi-day educational program in this study, no differences in interest in STEM fields were identified between girls and boys who participated in the program.

Research results from the 1970s onward (Ormerod, 1971; Ormerod & Duckworth, 1975; Murphy & Beggs, 2001; Lindahl, 2007) have indicated that the period from 10 to 14 years is critical in the development of positive attitudes toward natural sciences in young people, which are mostly defined by the age of 14. Although the attitudes of 14-year-olds do not necessarily reflect their future professional paths, they can still indicate the professions they want to pursue.

Research conducted by Tai et al. (2006) indicated that students who show interest in science careers at the age of 14 are more likely to choose a professional career in physics or a related engineering field.

The sustained increase in interest observed at the three-month follow-up may reflect the long-term impact of hands-on, interdisciplinary learning. As Aktulun et al. (2024) suggested, STEAM-based activities can ignite lasting motivation for science by engaging students cognitively and linguistically through practical exploration.

How science content is presented is extremely important for shaping students’ attitudes and interests. By applying learning in context, as in the “Sa STEMom raSTEMo” program, students can be practically shown that science can be useful in everyday life, in further learning, or at work, thus increasing their interest in STEM. This is supported by the results of the research reported here, which showed that interest in STEM fields among students who participated in the program was even greater three months after the implementation of the program than it was immediately after its implementation. Students who participated in the multi-day educational program “Sa STEMom raSTEMo” show a greater interest in pursuing STEM fields in the future (a greater number of respondents wanted to study space or explore volcanoes when they grew up) and believe that knowledge from STEM fields, especially computer science, would be useful to them in life.

Although the intervention was conducted within a school setting, it is important to acknowledge the broader ecosystem influencing STEM engagement. Parental awareness and attitudes have been shown to play significant roles in shaping children’s orientation toward STEM fields, particularly during transitional periods (Mercan et al., 2022). Future studies might consider incorporating family-based components to reinforce school-based efforts.

According to the National Research Council (1996), boys and girls of preschool age share very similar interests in STEM and display remarkable curiosity and a desire to learn through play. Gender differences begin to emerge in primary school and can already be identified by the fourth grade. As children grow older and transition to higher grades in primary school, there is a general decline in their interest across all subjects. However, differences between boys and girls become increasingly apparent in terms of both interest and their perception of STEM fields. Several studies have shown that the upper grades of primary school, specifically the period around the age of thirteen, are a crucial time period for forming and structuring interest in STEM fields (National Research Council, 1996).

The workshops conducted as part of “Sa STEMom raSTEMo” and the application of the research approach and discovery learning contribute to greater and more lasting interest, skill acquisition, and procedural knowledge in STEM. The research results reported here indicate the importance of quality and timely STEM education for students at an early age.

4.1. Scientific Contributions of This Study

This research makes a significant contribution to the field of early STEM education by providing empirical evidence on the effectiveness of inquiry-based and practical teaching methods in increasing students’ interest in STEM fields. By implementing the “Sa STEMom raSTEMo” program and using longitudinal measurement techniques, the study accomplishes the following:

It confirms the positive impact of experiential learning on motivation and interest among 10-year-old students (Kelley & Knowles, 2016; Bahar, 2003). Although the constructs of interest and motivation are conceptually grounded in self-determination theory, their operationalization within the study is now made explicit through the survey design. Specifically, student interest in STEM was assessed using items such as Q9 (“I would like to conduct practical work in Nature and Society lessons that will teach me more.”), Q10 and Q11 (“I would like more materials in Nature and Society lessons that spark my curiosity” and “I would like teaching materials that spark curiosity in Nature and Society lessons, even if they are difficult.”), and Q14 (“I find conducting experiments interesting.”), which reflect students’ desire to engage in STEM-related tasks and their perceived competence across domains. These items correspond to the psychological needs of autonomy and competence, as defined by SDT, and provide insight into how the intervention may have supported intrinsic motivation.

It demonstrates gender-neutral outcomes, offering important insights into inclusive curriculum design and combating stereotypes in STEM education (Burušić & Šerić, 2015; National Research Council, 1996).

It validates a quantitative instrument (Likert-scale questionnaire; Cronbach’s α = 0.92) tailored to young learners, which can be replicated in future STEM-related educational research.

It offers a delayed-effect analysis, showing sustained student interest three months after the intervention—an underexplored area in similar educational studies (Tai et al., 2006; OECD, 2023).

The sustained increase in STEM interest observed three months after the intervention suggests that its impact may extend beyond the immediate instructional period. One possible explanation for this delayed effect is the role of cognitive and emotional reflection. After the initial exposure to inquiry-based activities, students may have continued to process and internalize their experiences, leading to a gradual reinforcement of interest. Additionally, the program may have sparked curiosity that persisted beyond the classroom, encouraging students to engage informally with STEM-related topics in their everyday environments.

The absence of significant gender differences in STEM interest is noteworthy, particularly in the context of Croatian primary education, where gender stereotypes in science and mathematics may be less pronounced at younger ages. This finding aligns with previous research indicating that early interventions can mitigate gender-based disparities before they become entrenched (Maltese & Tai, 2011). However, further investigation is needed to determine whether these patterns hold across different regions and age groups.

Beyond the classroom, the broader ecosystem influencing student engagement should be considered. Parental attitudes and awareness of STEM have been shown to play a critical role in shaping children’s orientation toward these fields, especially during key transitional periods such as the shift from preschool to primary school (Mercan et al., 2022). Although the current intervention was school-based, acknowledging the influence of the home environment provides a more holistic understanding of the factors that contribute to sustained interest in STEM. Future research could explore the integration of family-oriented components, such as workshops or take-home activities, to reinforce school-based efforts and foster a supportive learning culture across contexts.

It adds depth to national discourse on declining PISA scores and STEM readiness in Croatia, presenting actionable pedagogical strategies to address gaps (NCVVO, 2023; Strategy of Science, Education and Technology, 2019).

This study introduces a custom-designed and validated questionnaire specifically tailored for measuring STEM interest among 10-year-old primary school students, which fills a notable gap in existing educational assessment tools (Cronbach’s α = 0.92).

Its longitudinal design, including pre-, post-, and delayed post-intervention data collection, allows for a nuanced examination of both the immediate and sustained impacts of practical STEM interventions—an approach rarely used in early education research (Tai et al., 2006; OECD, 2023).

By combining psychometric validation and multi-phase data tracking, the instrument offers a robust framework for replicable STEM evaluation, making it highly valuable for future educational studies across diverse European contexts (Bahar, 2003; Abd-El-Khalick et al., 2004).

The correlation between increased engagement and motivation was supported by the observed Cohen’s d effect size, indicating a moderate to strong impact. Student reflections further validate the development of critical thinking and identification with scientific roles—echoing the literature on early science identity formation (Dou et al., 2019; Maltese & Tai, 2011).

Furthermore, its alignment with age-appropriate cognitive development and the inclusion of curiosity-driven items underscores its developmental sensitivity, ensuring meaningful interpretation of young learners’ attitudes toward STEM.

4.2. Limitations of This Study

While this study provides valuable insights into early STEM education in Croatian primary schools, several limitations should be acknowledged to contextualize the findings. The sample was restricted to five schools located in urban and semi-urban areas within a single region, which may limit the generalizability of results to broader educational contexts. Additionally, the age range of participants was confined to ten-year-olds, excluding developmental perspectives across different age cohorts.

The sample was drawn from a limited number of schools, which may affect the study’s generalizability. Additionally, the intervention’s duration and the potential variability in facilitator delivery could have influenced outcomes. Future studies should consider larger, more diverse samples and explore the role of teacher training and parental involvement in sustaining STEM motivation.

The intervention lasted three months, which is a relatively short period for assessing long-term impacts. Although a delayed post-intervention measurement was conducted, further longitudinal follow-ups would be necessary to evaluate the sustainability of observed effects over time.

Another limitation concerns the delivery of the intervention. It was implemented by external experts—such as university professors and STEM educators—while the control group was taught by regular classroom teachers. Differences in instructional style and expertise may have influenced the outcomes, introducing a potential “teacher effect” that future studies should control for more systematically.

Moreover, the study did not account for external influences such as parental attitudes toward STEM education, which have been shown to significantly shape children’s motivation and interest (Mercan et al., 2022). Including the family context in future interventions could enrich our understanding of the broader ecosystem that supports early STEM engagement.

To address these limitations, future research could explore the effectiveness of inquiry-based STEM programs in rural or underserved educational settings, where infrastructural and pedagogical conditions may differ. Extending similar designs to younger (e.g., preschool-aged) or older students (e.g., lower secondary level) would allow for a more nuanced understanding of how STEM interest evolves across developmental stages (Lindahl, 2007; National Research Council, 1996). Comparative studies across diverse European regions could also enhance cross-cultural validity and inform inclusive curricular reforms aligned with EU strategic goals for science literacy (EU Publications, 2024; OECD, 2023).

Despite these constraints, the study supports the notion that structured, inquiry-based STEM interventions, especially those grounded in STEAM principles, can positively shape attitudes and interests among young learners (Aktulun et al., 2024). The authors recommend regular inclusion of such activities in formal education, supported by both qualitative and quantitative assessment tools.

While the intervention was designed as a comprehensive STEM program, we acknowledge that the mathematics (“M”) component was less explicitly represented compared to science and technology. Mathematical thinking was embedded in several modules, for example, through logical reasoning and variable manipulation in Micro:bit programming, as well as scale estimation and spatial reasoning in astronomy-related activities.

This imbalance is reflected in the measurement tools as well. Only one survey item (Q19) directly addressed mathematics, which constrains our ability to draw robust conclusions about changes in mathematical interest.

Future research should aim to design and evaluate mathematics-specific interventions within the STEM framework, ensuring that all components are equally represented and assessed.

Another important methodological limitation of this study concerns the assumption of independent observations. Although the data were collected from a limited number of schools and classrooms, the original analysis did not account for potential clustering effects. This may have influenced the precision of estimated effects and increased the risk of Type I or Type II errors. As such, the generalizability of the findings should be interpreted with caution. Future research should consider using hierarchical or multilevel modeling approaches to more accurately account for nested data structures and reduce bias in statistical inference.

In conclusion, the findings of this study confirm that well-designed scientific inquiry-based activities can significantly enhance students’ interest in STEM fields. Such programs not only foster curiosity and a spirit of exploration but also facilitate the development of essential skills such as critical thinking, collaboration, and problem-solving. Integrating these activities into regular classroom instruction, whether in science, mathematics, or technology, can enrich the learning experience and make it more meaningful for students. Moreover, early exposure to hands-on and investigative learning may have a lasting impact on shaping children’s educational and career interests, especially in light of the growing demand for STEM professionals. Therefore, it is recommended that these types of programs be systematically incorporated into school curricula, with support from teachers, educational specialists, and institutions.