Construct Validation of the STEM Motivation Scale for Children: A Multi-Timepoint Study

Abstract

This study aimed to examine the effects of a multi-day STEM program on students’ motivation toward STEM fields, using Exploratory and Confirmatory Factor Analysis, and to evaluate the psychometric properties of a new instrument developed in this study—“STEM Motivation Scale for Children”. A total of 549 students participated in the study, divided into experimental and control groups, with data collected at three time points. Factor analysis confirmed the structure of the instrument, identifying four latent dimensions: Interest for STEM field, Future Career in STEM, Preference for Practical Instruction, and Curiosity in STEM. Results revealed a statistically significant increase in factor scores within the experimental group following the intervention, with partial retention of the effects three months later. Exploratory and Confirmatory Factor Analysis (EFA and CFA) supported the theoretical structure of the instrument, and reliability and validity indicators were high. In conclusion, the STEM program positively influenced students’ motivation, while the questionnaire demonstrated satisfactory psychometric qualities for use in educational research.

1. Introduction

In the context of global educational challenges, student interest in the STEM field (Science, Technology, Engineering, and Mathematics) domains plays a crucial role in shaping future competencies and career choices. Despite the increasing significance of STEM professions in modern economies (European Schoolnet & DIGITALEUROPE, 2016), studies have reported a concerning decline in STEM motivation among students in many Western and highly developed Asian countries (Tai et al., 2006; Kelley & Knowles, 2016), while emerging nations such as India and Malaysia have shown increased engagement (ASPIRES, 2013).

STEM education requires active support within school curricula, infrastructure investment, and early student engagement (Qureshi & Qureshi, 2021). Programs featuring practical work and inquiry-based learning have shown potential in cultivating curiosity and long-term interest (Wells, 2019). However, crowded curricula and the overwhelming pace of technological advancement challenge schools to maintain student motivation (Lujan & DiCarlo, 2006), especially when informal sources of knowledge often surpass formal instruction.

In Croatia, national initiatives such as School for Life (Ministry of Science and Education, 2019) and “Sa STEMom raSTEMo” have aimed to modernize science education and promote STEM-related learning. Nonetheless, international assessments, such as PISA 2018, reveal a concerning lag in scientific literacy among Croatian students (NCVVO, 2018), emphasizing the urgent need for pedagogical strategies that enhance student engagement.

Gender disparities in STEM interest and performance have also gained attention, with several studies indicating early differences in attitudes and achievements between girls and boys (Burušić & Šerić, 2015; Glasser, 1994). Evidence suggests that upper primary years represent a critical period for shaping STEM preferences (National Research Council, 1996), highlighting the need for timely and reliable measurement tools (Sustainable Development Goals, 2019).

To address this need, the present study focuses on the development and validation of a novel instrument designed to assess elementary school students’ interest in the STEM field. The instrument was tested within the framework of a multi-day intervention based on practical STEM activities. Its psychometric evaluation included exploratory and confirmatory factor analyses, test–retest reliability via Intraclass Correlation Coefficients (ICCs), and internal consistency through Cronbach’s alpha (Cronbach, 1951). The findings contribute to educational research by providing a reliable measure of student interest and offering insights into the effects of STEM programs on motivation and engagement.

Early attitudes toward STEM significantly influence students’ future academic and career pathways (Said et al., 2016). Attitudes toward STEM fields begin to form during primary education, and their correlation with later professional interests is well-documented (Wiebe et al., 2018; Paechter et al., 2024). Reliable measurement of student motivation is essential for understanding educational preferences and reinforcing the STEM pipeline (Açıksöz et al., 2024a; Hulleman et al., 2015).

Recent developments in STEM education research emphasize the importance of motivational constructs in shaping students’ engagement and persistence in STEM fields. Studies conducted in the past few years have expanded our understanding of how scientific attitudes, instructional practices, and individual cognitive profiles influence STEM motivation (Açıksöz et al., 2024b; Çelik & Yıldız, 2023). These insights underscore the need for updated and culturally sensitive measurement tools that reflect contemporary educational contexts and student experiences (Açıksöz et al., 2024b).

2. Materials, Participants and Methods

2.1. Survey Questionnaire

A survey questionnaire, “STEM Motivation Scale for Children”, with 23 questions was developed for the present research (Appendix A).

Two open-ended questions were intended to collect personal data—first and last name and age—and one closed-ended question was used to collect information about the gender.

Three open-ended questions (Supplementary Materials) were intended to collect information about the grades obtained in the previous grade in nature and society, mathematics, and computer science subjects.

This was followed by two questions in which the respondents stated whether they had ever used a microscope or performed experiments at school. These questions provide insight into the intensity of practical work and the use of scientific research methods in classroom teaching.

The newly developed questionnaire was designed to measure four conceptual domains: Interest for STEM field, Future Career Orientation in STEM, Desire for Practical Instruction, and STEM-related Curiosity.

2.2. Study Participans

The sample comprised 549 elementary school students, aged 10 to 11 years, with a gender distribution of 257 boys and 272 girls. Participants were randomly assigned to either an experimental group, which took part in a structured multi-day STEM program, or a control group, which followed the regular curriculum. The STEM program featured hands-on activities and inquiry-based modules designed to promote engagement across science and technology domains.

2.3. Procedure

The study was conducted in accordance with ethical principles and approved institutional guidelines. Prior to data collection, written consent was obtained from all participating schools and legal guardians of the students. The research team provided participants with detailed information about the study’s goals, procedures, and confidentiality measures.

Data were collected in three phases: before the intervention (Before), immediately after the program (After), and three months later (Follow-up). During each phase, participants completed the newly developed “STEM Motivation Scale for Children” under standardized conditions, facilitated by trained researchers within school premises.

To assess the short-term retention of motivational effects following the STEM intervention, a follow-up measurement was conducted three months after the program. This timepoint was selected based on practical considerations related to the school calendar and the need to minimize participant attrition. While longer intervals (e.g., six months or one year) may offer deeper insights into the scale’s long-term sensitivity, the three-month window was deemed appropriate for capturing immediate post-intervention trends in motivation among primary school students.

Upon completion of each phase, responses were anonymized and securely stored for analysis. Feedback regarding the study outcomes was communicated to schools and families after the final measurement point.

2.4. Data Analysis

The instrument was psychometrically evaluated using Exploratory Factor Analysis (EFA) with principal component extraction and Varimax rotation. Factor analysis was conducted using the Maximum Likelihood extraction method with Oblimin rotation. Confirmatory Factor Analysis (CFA) was conducted to verify the proposed factor structure. Internal consistency was assessed using Cronbach’s alpha, while test–retest reliability was examined via Intraclass Correlation Coefficients (ICCs). All analyses were performed in R (version 4.5.1), using packages psych, lavaan, and ggplot2.

3. Results

A multi-day STEM workshop program was implemented in 2023, and participants (N = 549) were repeatedly surveyed using a questionnaire “STEM Motivation Scale for Children”. The gender structure of the respondents was analyzed and interest in the STEM field was compared between the control and intervention groups before implementation of the program (N_E = 272; N_C = 257). The control and intervention groups were compared before intervention, immediately after the intervention and at three-month follow-up after the intervention.

3.1. Pilot Study

A pilot trial of the questionnaire’s understandability was conducted on 25 participants. The children had no difficulty understanding and completing the survey. Reliability in this pilot group was calculated using the Cronbach α coefficient; a value of 0.92 was obtained.

3.2. Internal Consistency

Cronbach’s alpha was calculated for all subscales of the questionnaire. The coefficients ranged from 0.76 to 0.91, indicating satisfactory to high internal consistency across all constructs. These values exceed the commonly accepted threshold of 0.70, supporting the homogeneity of items within each subscale.

3.3. Test–Retest Reliability

To assess temporal stability, Intraclass Correlation Coefficients (ICCs) were computed for each item, using data collected at two measurement points (test–retest, new group, N = 45) with a two-week interval between administrations. ICC values ranged between 0.78 and 0.92, demonstrating good to excellent reliability. These findings indicate that the instrument yields stable results over time and is suitable for longitudinal studies (α > 0.70).

3.4. Analysis of the Experimental and Control Groups Before the Multi-Day STEM Program

Before the intervention, a questionnaire was administered to all participants, and a t-test was used to examine whether the groups had statistically significant differences. The results are shown in

Table 1. Results of t-test analysis of questions 9–23 of the STEM field interest survey before the intervention.

NE = 272; NC = 257	Mc	SDc	ME	SDE	t	p
Question 9	4.46	0.80	4.48	0.94	0.158	0.438
Question 10	4.43	0.82	4.52	0.84	0.894	0.187
Question 11	3.54	1.15	3.69	1.16	0.960	0.169
Question 12	4.57	0.71	4.44	0.90	−1.266	0.104
Question 13	3.98	0.90	3.94	1.42	−0.268	0.388
Question 14	4.57	0.92	4.44	0.90	1.319	0.095
Question 15	3.77	1.28	4.03	1.26	1.612	0.055
Question 16	3.89	0.90	4.15	1.20	1.944	0.027
Question 17	3.80	1.16	4.03	1.32	1.398	0.082
Question 18	3.69	1.27	3.91	1.30	1.321	0.094
Question 19	4.27	1.00	4.35	1.07	0.664	0.254
Question 20	4.23	1.06	4.13	1.21	−0.730	0.233
Question 21	3.55	1.44	3.28	1.63	−1.476	0.071
Question 22	4.23	1.06	4.13	1.21	−0.730	0.233
Question 23	3.55	1.44	3.28	1.63	−1.476	0.071

N_E—number of participants in the experimental group; N_C—number of participants in the control group; M_e—arithmetic mean of the responses in the experimental group; SD_E—standard deviation of the responses in the experimental group; M_c—arithmetic mean of the responses in the control group; SD_c—standard deviation of responses in the control group; t—t-test value; p—statistical significance.

.

No statistically significant differences in interest in the STEM field were observed between the control and experimental groups before the start of the program (Table 1).

A t-test was conducted for questions 9–23.

Table 1 presents the baseline comparison between the experimental and control groups regarding interest in STEM fields. While no statistically significant differences were observed in overall STEM interest scores prior to the intervention, a notable exception was found in item 16 (“I am interested in rainbows”), where the experimental group scored significantly higher than the control group (p = 0.027). This isolated difference suggests a pre-existing variation in interest related to specific STEM content.

The multi-day STEM program was implemented with students in the experimental group. Differences between the groups were investigated immediately and three months after the program delivery.

The experimental group shows increased variability and higher overall scores after the intervention, suggesting a broader engagement with STEM (Figure 1, Table S1 in Supplementary Materials). The control group’s distribution remains relatively unchanged (see Table S2 in Supplementary Materials).

3.5. Factor Analysis Exploratory and Confirmatory Phases

Exploratory Factor Analysis (EFA) was performed on the complete initial dataset (pre-test) of 549 participants using principal component extraction with Varimax rotation. The Kaiser–Meyer–Olkin measure verified sampling adequacy (KMO = 0.89), and Bartlett’s test of sphericity was significant (p < 0.001). Four distinct factors emerged, corresponding to the theoretical constructs:

• Interest for STEM field;
• Future Career Orientation;
• Desire for Practical Instruction;
• STEM Curiosity.

Table 2. STEM-related sores for experimental group Before, After the Intervention and at Three-month follow-up.

Factor	Before	After	Follow-Up
Interest for STEM field	0.81	0.85	0.84
Future Career Orientation	0.76	0.79	0.78
Desire for Practical Instruction	0.83	0.87	0.85
STEM Curiosity	0.77	0.79	0.81

summarizes the mean scores for the four STEM-related constructs measured at three different time points: Before the intervention, After, and Three-month follow-up.

The experimental group shows a consistent increase in all dimensions, especially in practical instruction and interest for STEM fields (Table 2).

The Interest for STEM field factor includes items reflecting students’ motivation for acquiring practical knowledge during Nature and Society lessons. The Future Career Orientation in STEM factor captures students’ aspirations toward occupations such as robotics, astronomy, biology, and programming. The Desire for Practical Instruction factor refers to students’ preference for hands-on projects, experiments, and activities involving active participation. The STEM Curiosity factor comprises items exploring students’ interest in natural phenomena, including rainbows, clouds, bacteria, and outer space.

All four constructs show positive increases in mean scores immediately after the intervention, especially Desire for Practical Instruction and Interest for STEM field, suggesting that the STEM program had a strong motivational impact.

Three months later, the scores remained above baseline, indicating that effects were partially retained over time.

The curiosity factor showed incremental growth, implying that deeper cognitive engagement might develop gradually, even after the formal intervention ends.

Table 3. Item-Level Factor Loadings Across STEM Constructs.

Question	Interest for STEM Field	Future Career Orientation in STEM	Desire for Practical Instruction	STEM Curiosity
Q9	0.71		0.56
Q10	0.58		0.60
Q11	0.52			0.46
Q12	0.54
Q13	0.66		0.61
Q14			0.59	0.52
Q15				0.74
Q16				0.70
Q17		0.50
Q18		0.65		0.42
Q19		0.52
Q20		0.77
Q21		0.69
Q22		0.72
Q23		0.74

presents the strength of association (factor loadings) between each questionnaire item and the four latent STEM constructs derived through Exploratory Factor Analysis. All factor loadings exceeded 0.40 (Table 3), and communalities were satisfactory. The four-factor solution accounted for 63.4% of the total variance.

Values above 0.50 indicate moderate to strong association. Items such as Q15 (“I am curious about how clouds are formed”), Q22 (“I want to study viruses and bacteria”), and Q20 (“I want to program computers”) demonstrate high construct alignment, confirming the questionnaire’s internal structure and conceptual coherence.

Overlap across some items (e.g., Q10 and Q13) may suggest shared motivational dimensions like curiosity and practical engagement.

Mean scores and standard deviations for four STEM-related constructs—interest, career orientation, desire for practical instruction, and curiosity—are reported across three time points (before intervention, immediately after, and three months later). The experimental group shows noticeable increases post-intervention, whereas the control group remains relatively stable (

Table 4. Descriptive Statistics of STEM Constructs for Experimental and Control Groups Across Three Measurement Points.

Factor	Group	Before	After	Three-Month Follow-Up
Interest for STEM field	Experimental group	4.10 ± 0.50	4.58 ± 0.48	4.51 ± 0.50
	Control group	4.12 ± 0.50	4.15 ± 0.51	4.20 ± 0.49
Future Career Orientation in STEM	Experimental group	3.72 ± 0.60	4.22 ± 0.58	4.01 ± 0.59
	Control group	3.68 ± 0.60	3.71 ± 0.61	3.69 ± 0.63
Desire for Practical Instruction	Experimental group	4.05 ± 0.50	4.73 ± 0.46	4.52 ± 0.47
	Control group	4.01 ± 0.50	4.10 ± 0.51	4.13 ± 0.50
STEM Curiosity	Experimental group	4.22 ± 0.50	4.35 ± 0.55	4.47 ± 0.53
	Control group	4.19 ± 0.50	4.21 ± 0.56	4.20 ± 0.57

, Figure 2).

The most significant changes were observed in the experimental group. Interest for STEM field increased considerably immediately following the intervention (M = 4.58; SD = 0.48), compared to baseline measurement (M = 4.10; SD = 0.52), with the effect sustained three months post-intervention (M = 4.51; SD = 0.50). In the control group, changes were not statistically significant. The Future Career Orientation in STEM factor showed a moderate increase in the experimental group after the intervention (M = 4.22; SD = 0.58), compared to the initial state (M = 3.72; SD = 0.61), with scores remaining above baseline levels three months later (M = 4.01; SD = 0.59). No significant changes were recorded in the control group. Desire for Practical Instruction showed the highest increase (M = 4.73; SD = 0.46) following the intervention, with a slightly reduced but still elevated score three months later (M = 4.52; SD = 0.47). STEM Curiosity demonstrated a delayed effect, showing its most notable increase three months after the intervention (M = 4.47; SD = 0.53) (Table 4).

Graphical representations clearly illustrate group divergence. The experimental group showed growth and stabilization, while the control group remained unchanged (Figure 2).

We calculated p-values for differences between the experimental and control groups at three time points across four STEM-related motivational factors. The assumed sample sizes were N_E = 272 and N_C = 257 (

Table 5. p-Values for Group Comparisons Over Time (N_E = 272 and N_C = 257).

Factor	Measurement Time	p-Value	Conclusion
Interest for STEM field	Before	0.6754	No significant difference
	After	<0.0001	Significant difference
	Three-month follow-up	<0.0001	Significant difference
Future Career Orientation	Before	0.6542	No significant difference
	After	<0.0001	Significant difference
	Three-month follow-up	0.0002	Significant difference
Desire for Practical Instruction	Before	0.6213	No significant difference
	After	<0.0001	Highly significant difference
	Three-month follow-up	<0.0001	Highly significant difference
STEM Curiosity	Before	0.7439	No significant difference
	After	0.0121	Significant difference
	Three-month follow-up	0.0003	Significant difference

).

Before the intervention, there were no statistically significant differences between groups, confirming baseline equivalence. After the intervention and at the three-month follow-up, the experimental group showed significantly higher scores across all motivational factors.

The most pronounced effects were observed in Desire for Practical Instruction and Interest for STEM field, with p-values below 0.0001 (Table 5).

Confirmatory Factor Analysis (CFA) was conducted using the lavaan package in R, based on data collected at the first measurement point. The structure included four latent constructs, each linked to corresponding observed items. No correlated error terms were specified. Fit indices indicated that the model demonstrated good alignment with the data, supporting the theoretical configuration of the instrument.

CFA was conducted to confirm the structure identified through EFA. Fit indices showed acceptable model fit:

χ² (164) = 289.1, p < 0.001;

CFI = 0.94;

TLI = 0.92;

RMSEA = 0.045 [90% CI: 0.041, 0.056];

SRMR = 0.048;

These results support the theoretical structure of the questionnaire and validate its factorial integrity.

3.6. Discriminant Validity

The questionnaire demonstrated the ability to differentiate between the experimental and control groups, with statistically significant differences observed in three out of four factors following the intervention. This supports the instrument’s sensitivity to changes resulting from educational interventions.

3.7. Distributional Properties

Item-level analysis revealed satisfactory response dispersion, with no evidence of ceiling or floor effects. Examination of skewness, kurtosis, and deviation from normality confirmed that item distributions met the fundamental assumptions of normality.

Table 6. Psychometric Properties of the STEM Interest Questionnaire.

Psychometric Characteristic	Value/Result
Factor Structure (EFA)	4 factors, loadings > 0.40
Cronbach’s α (all factors)	0.76–0.87
CFA Fit indices	CFI = 0.94, RMSEA = 0.045
Test–Retest Reliability	r = 0.62–0.78
Discriminant Validity	Significant differences between experimental and control groups (p < 0.05)
Floor/Ceiling Effects	Not present
Item Distribution Normality	Assumptions met (skew < 1, kurtosis < 1.5)

presents key psychometric indicators for the STEM Interest Questionnaire. Results from exploratory (EFA) and confirmatory (CFA) factor analyses confirmed a four-factor structure with satisfactory internal consistency (α = 0.76–0.87). Fit indices indicated a good model fit (CFI = 0.94; RMSEA = 0.045; SRMR = 0.048). Test–retest reliability showed stability over time (r = 0.62–0.78), and discriminant validity was demonstrated through statistically significant differences between groups (p < 0.05). All items demonstrated satisfactory factor loadings (≥0.40), with no significant cross-loadings. The items exhibited acceptable response dispersion, with no apparent ceiling or floor effects. Analysis of deviations, skewness, and kurtosis confirmed that item distributions met the fundamental assumptions of normality.

3.8. Inter-Factor Correlations

Inter-factor correlations revealed meaningful associations among the four latent constructs (

Table 7. Intercorrelations Between STEM Interest Questionnaire Factors.

	Interest for STEM Field	Future Career Orientation in STEM	Desire for Practical Instruction	STEM Curiosity
Interest for STEM field	1.00	0.45 (moderate)	0.87 (strong)	0.51 (strong)
Future Career Orientation in STEM	0.45 (moderate)	1.00	0.34 (moderate)	0.70 (strong)
Desire for Practical Instruction	0.87 (strong)	0.34 (moderate)	1.00	0.41 (moderate)
STEM Curiosity	0.51 (strong)	0.70 (strong)	0.41 (moderate)	1.00

). The strongest relationship was observed between Interest for STEM Field and Desire for Practical Instruction (r = 0.87), indicating that students who expressed greater interest for STEM field also showed a strong preference for hands-on learning experiences. STEM Curiosity was moderately correlated with Future Career Orientation in STEM (r = 0.70), suggesting conceptual overlap between personal inquisitiveness and long-term professional aspirations. Overall, the constructs demonstrated adequate discriminant validity, as correlations ranged from moderate to strong but did not indicate redundancy.

Table 6 displays Pearson correlation coefficients among the four factors of the STEM Interest Questionnaire: Interest for STEM Field, Future Career Orientation in STEM, Desire for Practical Instruction, and STEM Curiosity.

All correlations are positive and statistically meaningful, indicating related yet distinct constructs.

4. Discussion

This study provides robust empirical support for the validity and applicability of the newly developed STEM Motivation Scale for Children, confirming its conceptual grounding and psychometric integrity. The four-factor model—Interest for STEM field, Future Career Orientation, Desire for Practical Instruction, and STEM Curiosity—aligns with established theories of intrinsic motivation (Deci & Ryan, 1985), experiential learning (Kolb, 1984), constructionist engagement (Papert, 1980), and cognitive curiosity (Berlyne, 1960).

Psychometric analyses confirmed the scale’s structural soundness and temporal stability. Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) indicated strong factor loadings and satisfactory model fit indices (Kline, 2015; Costello & Osborne, 2005), while internal consistency (Cronbach, 1951; Tavakol & Dennick, 2011) and test–retest reliability (Shrout & Fleiss, 1979; Koo & Li, 2016) fulfilled criteria for longitudinal application. The development process followed best practices for educational measurement instruments (DeVellis, 2016; Artino et al., 2014), enhancing the tool’s methodological rigor.

The scale demonstrated sensitivity to pedagogical intervention effects. Following a multi-day inquiry-based STEM program, participants in the experimental group showed statistically significant increases in all motivational domains (p < 0.001), with sustained effects in Interest and Practical Instruction three months post-intervention. Curiosity showed a delayed elevation (p = 0.04), implying gradual cognitive immersion—a finding congruent with prior research (Gardner & Tamir, 1989; Freeman et al., 2014).

These results emphasize the educational impact of hands-on, problem-centered learning. Integrating experiential modules can foster deeper engagement, particularly in primary education where motivational foundations begin to form (Tai et al., 2006; Lindahl, 2007). The instrument’s discriminant validity and responsiveness make it a valuable asset for educators seeking to tailor instruction to evolving student interests and for policymakers designing targeted STEM initiatives.

The findings of this study align with recent literature that highlights the multifaceted nature of STEM motivation and its sensitivity to pedagogical and psychological factors. For instance, Açıksöz et al. (2024a) demonstrated that scientific attitudes and multiple intelligences significantly predict STEM motivation among female students, while Açıksöz et al. (2024b) validated a multidimensional scale capturing expectancy-value constructs in middle school populations. These contemporary studies support the relevance and applicability of our newly developed scale and suggest promising avenues for future cross-cultural validation and longitudinal tracking.

The conceptual clarity of each construct further reinforces the scale’s utility. Interest for STEM field reflects active engagement and the perceived relevance of science content (Deci & Ryan, 1985). Future Career Orientation captures vocational aspirations aligned with STEM fields (Savickas, 2005), while Desire for Practical Instruction stems from constructionist and experiential paradigms (Papert, 1980; Kolb, 1984). STEM Curiosity encapsulates exploratory drive and inquiry motivation, which flourish over time with continued exposure (Berlyne, 1960; Illeris, 2007).

To maximize impact, future research should focus on expanding the scale’s dimensionality—incorporating constructs such as self-efficacy, peer influence, and classroom climate (Hulleman et al., 2015; Osborne & Dillon, 2008)—and validating its performance across cultural and linguistic contexts (Paechter et al., 2024; Açıksöz et al., 2024b). Such efforts would strengthen our understanding of motivational trajectories and enrich strategies for sustaining STEM interest throughout the educational pipeline.

5. Conclusions

This study presents a newly developed and thoroughly validated STEM Motivation Scale for Children, designed to capture core motivational constructs among elementary school students. Through rigorous psychometric evaluation—including factor analyses, reliability assessments, and sensitivity testing—the instrument demonstrated solid structural validity, temporal stability, and practical relevance for educational research.

Empirical findings revealed that participation in a multi-day STEM program significantly increased student motivation across all examined domains, with sustained effects observed three months post-intervention. In particular, constructs such as Interest for STEM field and Desire for Practical Instruction exhibited robust immediate gains, while STEM Curiosity developed gradually over time—underscoring the layered nature of cognitive engagement.

Furthermore, the findings offer important implications for both educators and policymakers. By recognizing early indicators of STEM motivation, teachers can better tailor their instructional strategies to foster sustained interest, while policymakers are equipped with evidence to support the design of targeted interventions that strengthen long-term engagement across the education system.

This study provides initial evidence that participation in a multi-day STEM program can lead to short-term increases in students’ motivation and interest in STEM-related domains. While the observed changes were statistically significant, it is important to interpret these findings with caution. The data do not necessarily reflect deep or sustained engagement, and the relatively short follow-up period limits conclusions about long-term impact. Moreover, motivational trajectories in primary school students are influenced by a range of extraneous factors, and interest may fluctuate or diminish over time. Future research should explore the durability of these effects through extended longitudinal designs. Nonetheless, the validated STEM Motivation Scale offers educators a practical tool for monitoring and supporting student engagement in STEM learning environments.

The validated STEM Motivation Scale for Children offers educators a practical and reliable tool for identifying motivational profiles and shifts among students in upper primary grades. By assessing constructs such as interest, curiosity, career orientation, and desire for practical instruction, teachers can tailor lesson plans to better reflect students’ needs and preferences. The results of this study highlight the effectiveness of multi-day, inquiry-based STEM programs in enhancing student motivation, suggesting that integrating hands-on and exploratory activities into everyday instruction may yield lasting engagement. Moreover, regular use of the questionnaire enables monitoring of program impact over time, offering insights that can inform curriculum development, differentiated instruction, and early career guidance initiatives in STEM education.

Limitations and Future Research

While the present study provides valuable insights into the effects of STEM interventions and the reliability of a newly developed instrument, several limitations should be acknowledged. First, the sample consisted solely of students aged 10 to 11 from a limited number of schools, which may reduce the generalizability of findings to broader populations or different educational contexts. Second, all data were based on self-reported measures, which may be subject to social desirability bias. Third, the instrument was validated within a single cultural and linguistic setting; future studies should consider cross-cultural adaptation and validation.

Another limitation of the present study is the relatively short duration between the intervention and the follow-up measurement. Although the three-month interval allowed for the assessment of short-term motivational retention, future studies should consider extended follow-up periods (e.g., six months or end-of-year) to better understand the durability of motivational changes and the long-term applicability of the STEM Motivation Scale. Such longitudinal data would provide a more comprehensive view of how sustained engagement with STEM develops over time.

Future research should explore longitudinal applications of the instrument, examining motivational changes across multiple academic years. Additionally, it would be valuable to test the questionnaire in diverse educational systems and with broader age ranges. Expanding the instrument to include additional constructs—such as teacher support or peer influence—could further enrich our understanding of the mechanisms driving student interest in the STEM field.