Enhancing Early STEM Engagement: The Impact of Inquiry-Based Robotics Projects on First-Grade Students’ Problem-Solving Self-Efficacy and Collaborative Attitudes

Abstract

This study examines the effects of integrating an inquiry-based final project into an early childhood robotics program, focusing on its influence on children’s problem-solving self-efficacy, attitudes toward collaboration, confidence in applying robotics to real-world challenges, and future interest in STEM. A total of 176 first-grade students (aged 6–7) were randomly assigned to either a research group that completed a culminating inquiry-based robotics project or a control group that followed a traditional structured curriculum. A quasi-experimental post-test-only comparison group design was used, and baseline equivalence was confirmed across groups. Results revealed that children who participated in the inquiry-based final project group demonstrated significantly higher problem-solving self-efficacy and more positive attitudes toward peer collaboration, while also being more likely to see the relevance of robotics to real-world problems and to align with inquiry-based learning approaches. Gender analysis showed that these gains were especially pronounced among girls, who exhibited more statistically significant improvements in problem-solving confidence and self-efficacy in inquiry-based problem-solving. The study’s findings highlight the benefits of incorporating inquiry-based final projects into early robotics curricula, addressing a critical gap in early childhood STEM education by providing evidence-based insights into how to enhance foundational STEM dispositions and engagement through inquiry-based, technology-integrated instruction.

1. Introduction

Across educational theory and practice, differences are often drawn between inquiry-based and more teacher-led approaches, but in reality these categories overlap and exist on a continuum. Many traditional models emphasize direct instruction, where educators transmit knowledge through lectures, demonstrations, and structured assignments; however, such approaches can also include student activities and formative feedback. Inquiry-based learning (IBL), rooted in constructivist theories, places greater emphasis on students generating questions, investigating, experimenting, and reflecting (Chin & Chia, 2004; Pedaste et al., 2015; Spronken-Smith, 2012). Rather than passively receiving information, students are encouraged to explore problems, form hypotheses, and seek evidence-based solutions through hands-on engagement. This process can nurture deeper conceptual understanding, intrinsic motivation, and transferable thinking skills.

While both approaches have merit, a growing body of research suggests that inquiry-based learning can lead to enhanced engagement, autonomy, and long-term retention, especially when learners are supported to take ownership of their learning journey (Kuhlthau et al., 2015; Urdanivia Alarcon et al., 2023). IBL is particularly effective when applied to interdisciplinary and real-world contexts, where solutions are not predefined, and the process of exploration itself becomes central to the learning experience.

Another important pedagogical distinction lies between formative and non-formative (summative) learning practices, in general, and assessment, as a major aspect of feedback driven education. While summative assessment evaluates learning at the end of an instructional unit using fixed measures like exams or grades, it provides limited opportunities for learners to adapt and improve during the process. By contrast, formative learning practices are designed to embed ongoing feedback and reflection throughout the learning process to promote continuous growth (Dixson & Worrell, 2016). Formative practices not only enhance mastery of content but also support the development of self-regulation, resilience, and confidence. When paired with inquiry-based approaches, formative assessment creates a powerful synergy that enhances student competencies by guiding learners to refine their understanding through constructive feedback and ongoing support (Grob et al., 2017).

This dual distinction between inquiry vs. non-inquiry and formative vs. summative approaches becomes especially significant in early childhood education, a stage where foundational attitudes toward learning start to form. Young children are naturally curious, and environment that foster exploration, experimentation, and constructive feedback can have long-lasting impacts on self-confidence, persistence, and enjoyment. Integrating inquiry-based methods alongside formative practices not only supports cognitive development, but also promotes affective growth, nurturing learners who feel capable, curious, and empowered.

In this context, educational robotics (ER) emerges as a compelling tool. ER technology is used to support learning in schools, especially in the subjects of science, technology, engineering, mathematics and informatics (Kim et al., 2015). Robotics activities inherently require learners to ask questions, test solutions, iterate designs, and collaborate—perfectly aligning with inquiry-driven pedagogy. When implemented with formative strategies such as guided feedback and reflective dialogue, robotics projects in early childhood can provide experiences that enhance both confidence and joy, making abstract science, technology, engineering and math (STEM) concepts tangible and engaging through play-based exploration (Çetin & Demircan, 2020; Ioannou & Makridou, 2018; Jung & Won, 2018; Tselegkaridis & Sapounidis, 2022; Zviel-Girshin et al., 2021).

This study investigates the impact of embedding a formative, inquiry-based final project within an early childhood robotics curriculum. The core objective is to compare outcomes between this approach and non-inquiry-based implementations to determine how such a pedagogical design influences both the emotional and cognitive dimensions of young children’s learning.

2. Theoretical Framework

This study situates educational robotics within a robust, integrated theoretical framework that draws upon constructivism, constructionism, inquiry-based learning, project-based learning, Self-Determination Theory, and Social Cognitive Theory. These perspectives collectively provide a comprehensive lens through which to examine how young children engage in meaningful STEM learning through ER activities.

2.1. Foundational Learning Theories: Constructivism and Constructionism

Constructivism and constructionism form the cognitive backbone of the framework. Rooted in history of more than a century, with Dewey’s (1938) pragmatist experimentalism, and Piaget’s (1954) theory of cognitive development, constructivism emphasizes that children actively construct understanding by interacting with their environment. Vygotsky’s (1978) sociocultural theory complements this view by highlighting how learning is enhanced through dialogue and scaffolding from more knowledgeable others.

Papert’s (1980) constructionism extends constructivism by focusing on learning through designing and manipulating tangible artefacts, which he called cognitive artefacts. This perspective is particularly relevant to ER, where children create, program, and test robots, thereby externalizing their thinking and refining their mental models through the process of creation. Studies confirm that children as young as 4–7 years old can successfully engage in these activities, demonstrating the developmental appropriateness of ER in early education (Bers, 2019; Bers et al., 2019; Cejka et al., 2006; Zviel-Girshin et al., 2021, 2024).

2.2. Educational Robotics in International Early Childhood Context

Globally, ER has emerged as a promising means of introducing computational thinking and foundational engineering habits in early childhood and elementary school (Eguchi, 2016; Tzagkaraki et al., 2021). Curricula in Europe, North America, and Asia increasingly integrate robotics to promote 21st-century skills such as problem-solving, algorithmic reasoning, creativity, and collaboration (Castro et al., 2018; Stewart et al., 2021). Programs often combine block-based coding and construction kits to make abstract STEM concepts concrete and developmentally appropriate (Bers, 2019; Çetin & Demircan, 2020).

Research to date shows benefits including improved algorithmic thinking (Cejka et al., 2006), increased motivation and engagement (Jung & Won, 2018; Ioannou & Makridou, 2018), and stronger problem-solving self-efficacy (Canbeldek & Isikoglu, 2023). However, findings remain fragmented; many studies use short interventions, small samples, or teacher-led activities with limited inquiry or project components. There is also a lack of systematic investigation into how inquiry-based methods affect young learners’ confidence and collaboration, and few studies explore gender as a moderating factor (Montuori et al., 2022; Angeli & Valanides, 2020). The present study addresses these gaps by embedding an inquiry-based final project within an established early robotics curriculum.

2.3. Pedagogical Approaches: Integrating Inquiry and Projects

ER projects align closely with two complementary pedagogical approaches: Inquiry-Based Learning and Project-Based Learning.

Inquiry-Based Learning emphasizes exploration, questioning, and problem-solving through authentic, open-ended experiences (Kuhlthau et al., 2015). Through IBL, children formulate hypotheses, test solutions, and revise ideas, mirroring core scientific thinking processes. These hands-on experiences promote deeper understanding and reflective learning.

Project-Based Learning (PBL) provides a structure for extended, collaborative, and interdisciplinary work like final robotics challenges. PBL encourages children to take ownership of their learning, apply knowledge to real-world problems, and enhance critical thinking (Chen & Yang, 2019; Lev et al., 2020). While empirical evidence in early childhood remains limited (Ferrero et al., 2021), PBL and ER are highly compatible, promoting STEM integration and motivation (Castro et al., 2018; Stewart et al., 2021; Tzagkaraki et al., 2021).

Although related learner-centered pedagogies such as problem-based learning, project-based learning, and model-based learning share overlapping features with IBL, the present study adopts IBL as its primary descriptor. The experimental intervention —the culminating final project—was specifically designed to place the greatest emphasis on children formulating their own questions, exploring alternative solutions, testing ideas, and reasoning from evidence—core characteristics of IBL (Pedaste et al., 2015; Chin & Chia, 2004). Elements of project- or problem-based learning were present (e.g., extended final project, real-world challenges), but served as strategies nested within an inquiry-based cycle rather than separate paradigms. This distinction clarifies the study’s theoretical orientation and justifies the use of IBL as the overarching pedagogical framework.

2.4. Motivation and Social Development

Self-determination theory adds a critical motivational dimension. According to Ryan and Deci (2000), satisfying the basic psychological needs for autonomy (making choices), competence (gaining mastery), and relatedness (social connection) fosters intrinsic motivation. ER projects directly support these needs: children make design choices (autonomy), gain mastery through iteration (competence), and collaborate with peers (relatedness), enhancing engagement and perseverance.

Social cognitive theory emphasizes the role of social interaction, peer modeling, and self-efficacy in learning (Bandura, 1986). ER activities naturally involve teamwork and peer feedback, allowing children to learn from one another, share strategies, and build self-efficacy through successful performance and social validation.

2.5. Dimensions of Student Engagement

Within this framework, student engagement is multidimensional:

• Behavioral: Active participation in building, testing, and programming.
• Cognitive: Questioning, problem-solving, and reflecting on outcomes.
• Emotional: Experiencing confidence, enjoyment, and perseverance in the face of challenges.

ER fosters these competencies while also supporting 21st-century skills such as computational thinking, creativity, and interdisciplinary integration beyond STEM alone (Canbeldek & Isikoglu, 2023; Zviel-Girshin et al., 2020, 2021).

2.6. Gender and Inquiry-Based Learning

Gender differences in young children’s engagement with robotics and programming had received limited empirical attention, yet the few studies that existed suggested this was a meaningful area for investigation. Very little research has directly examined how IBL interacted with gender in early STEM; however, related work showed that girls responded positively to collaborative, context-rich robotics activities (Beisser, 2012; Montuori et al., 2022). Research also suggested that IBL effectively trained metacognitive skills, with no significant differences between male and female students’ outcomes (Nunaki et al., 2019). Furthermore, carefully designed scaffolding was found to help counteract later gender gaps in STEM self-efficacy (Angeli & Valanides, 2020). In contrast, research with older students often reported higher robotics self-efficacy and future STEM interest among boys (Atmatzidou & Demetriadis, 2016; Kucuk & Sisman, 2020). Given this limited but suggestive evidence, examining gender as a moderating factor in the present study allowed exploration of whether an inquiry-based robotics project could support more equitable early STEM engagement.

Taken together, these perspectives provide the conceptual rationale for investigating the impact of a culminating inquiry-based robotics project on children’s confidence, problem-solving self-efficacy, collaboration, and future STEM engagement. Crucially, this integrated framework also provides the basis for examining gender as a potential moderating factor, helping to determine the project’s potential for promoting equitable early STEM outcomes. This integrated framework guided the formulation of the research aims and questions presented in the Section 4.

3. Program Description

3.1. The Program

The ECE program was launched a decade ago across several kindergartens and first-grade classrooms and has since expanded steadily, with more educational institutions joining each year. Designed to cultivate essential 21st-century skills in early childhood, the program focuses on developing communication, collaboration, creativity, critical thinking, and problem-solving (Bers et al., 2019; Sharma et al., 2019). In addition, it aims to boost young learners’ confidence in using technology, nurture their self-belief, and foster self-efficacy and resilience in the face of challenges (Zviel-Girshin et al., 2020, 2024). In the current study, only elementary schools were involved.

3.2. Program Description in Elementary Schools

In recent years, a compulsory two-hour weekly robotics lesson was integrated into the first-grade curriculum across participating primary schools. The program was designed to immerse young learners (ages 6–7) in hands-on, collaborative, and inquiry-based learning that nurtured foundational computational thinking, problem-solving, and technological fluency. Lessons were delivered either by trained in-school teachers or external robotics instructors. Each class was divided into small collaborative teams of 3–4 students to maximize participation and peer interaction.

Instruction followed a mediated learning approach, which combined brief direct instruction, followed by hands-on activities and open-ended inquiry tasks. Lessons often begin with short, multimedia-rich presentations introducing key robotics or programming concepts, followed by team-based inquiry activities where children brainstorm, design, build, and program solutions to given challenges. Each session emphasizes student autonomy, curiosity, and teamwork. Communication, prediction, decision-making, and justification are central to class interactions, with teachers prompting deeper thinking through scaffolding questioning (e.g., “What will happen if this sensor is removed?” or “How might this change affect the robot’s behavior?”). Instructional materials included LEGO^® Education kits (primarily WeDo 2.0) and custom-developed tasks. While some challenges were well-defined, others were intentionally left open-ended to foster imagination, creativity, and innovation.

Each lesson culminated in the construction and programming of a tangible artifact, ranging from simple paper models to fully functional robots. Certain projects extended over multiple sessions. Lessons typically concluded with structured peer discussions on proposed solutions, analysis of common errors, and the scaffolding introduction of the Use–Modify–Create learning progression. This pedagogical sequence gradually guided students from initially using existing models, to modifying them based on new goals or challenges, and ultimately to independently designing and implementing original projects.

To promote equitable engagement and active participation, each student was assigned a specific role during both the construction and programming phases. In the construction phase, one student read the instructions, another located the necessary parts, and a third assembled the model. During the programming phase, one student acted as the primary coder, while teammates adopted supportive roles, reviewing code, asking clarifying questions, identifying errors, and contributing to debugging and refinement through collaborative dialogue (Figure 1).

The robotics lab environment included a dedicated table for each team, a computer or laptop, and an age-appropriate programming interface. The icon-based LEGO^® WeDo 2.0 programming environment eliminated the need for text-based coding, making it accessible even to emergent readers. Students assembled logical sequences using drag-and-drop blocks, enabling them to program through a visual interface (Figure 2). The robots’ immediate responses provided tangible feedback, helping students quickly evaluate whether their code worked as intended and make adjustments accordingly. This process bridged the gap between abstract programming concepts and real-world application, fostering ongoing reflection and teamwork.

3.3. Final Project: Integrating Inquiry into Final Robotics Challenges

Over the past decade, the authors, as heads of the program, as integral part of their work, explored and refined several formats for implementing and presenting the final robotics challenge. These approaches evolved alongside the program and included:

Core curriculum only: Early iterations focused solely on delivering foundational content without a culminating project

Classroom-based challenges: Students completed final projects shared only within their immediate classroom or kindergarten setting

School-wide exhibitions: Projects expanded to reach the broader school community, with students presenting to multiple classes

Inter-school “Robotics Day” events: The most comprehensive format, where students presented final models and explanatory posters to peers, teachers, families, and local authorities.

When final challenges were implemented, the program culminated in a final engineering challenge designed to consolidate students’ learning and showcase their developing skills in robotics, programming, and problem-solving. Each year, the narrative framing adapted to maintain relevance and student engagement. Past themes included topics such as space exploration and lunar settlement, autonomous vehicles, and smart transportation systems.

Approximately four to five weeks before the end of the academic year, students formed small teams of 3–5 members and began working on self-directed final projects. Rather than being assigned a fixed task, students were invited to define the problem themselves—they asked questions, investigated real-world needs, proposed creative solutions, and only then proceeded to construct and program their robots.

This enriched model expanded upon the traditional three-phase Build–Program–Test cycle by embedding an initial formative inquiry phase that emphasized exploration, critical thinking, and student ownership of learning. The revised structure followed a seven-phase learning progression. Figure 3 presents the stages of this inquiry-based formative project lifecycle model.

Within this inquiry-driven framework, students engaged with open-ended themes, formulated their own problem statements, and articulated justifications for their proposed robotic solutions. The process began with students posing questions, investigating contextual needs, and defining a specific problem rooted in the theme. Before engaging in hands-on construction, they were encouraged to explain and justify their planned solutions, fostering metacognitive reasoning and deliberate planning.

Although described as “open inquiry”, the culminating project was intentionally adapted to be age-appropriate. Teachers guided children in brainstorming and selecting meaningful real-world problems (e.g., helping a farmer harvest, saving astronauts) and supported planning by prompting them to break challenges into manageable steps. They used scaffolding questions—such as What is the need?, What is the technological problem?, What is the technological solution?, and Which parts and sensors should be included in the motorized model?—to help children structure their ideas while keeping decision-making student-led. Materials and time frames were provided, and teachers ensured safety and basic feasibility.

A dedicated reflection stage promoted iterative thinking, prompting children to refine their ideas as their understanding evolved. This approach shifted the emphasis from simply completing a task to embracing the learning journey itself. Rather than serving as a summative evaluation, the final project was designed as a formative experience, aimed at cultivating habits of inquiry, exploration, creativity, and self-directed learning. Children were encouraged to experience inquiry as an inherently highly enjoyable experience making the process of discovery a meaningful outcome in its own right.

4. Methodology

This section outlines the aims of the study, research design, participant details, data collection instruments, procedures, and methods of data analysis.

4.1. Aims of the Study

The ECE program, initiated at Emek Hefer, Israel in 2016 and continuing to this day, explores the effectiveness of inquiry-based learning in educational robotics in early elementary education. Robotics provides a rich context for developing core STEM outcomes such as computational thinking, problem-solving, teamwork, and an engineering mindset. A central focus is whether incorporating a final inquiry-based project at such a young age provides additional motivational and educational benefits. While our previous research in educational robotics (Zviel-Girshin et al., 2021) demonstrated that young participants were highly creative and valued both their own and their team’s creativity, it remained unclear whether the added time and effort required by a final project in robotics translated into measurable gains for learners.

The motivation for the present study lies in examining how the inclusion of culminating inquiry-based projects in early childhood robotics education influences student outcomes across multiple dimensions. Specifically, we sought to investigate whether children who completed a final open-ended inquiry-based robotics project demonstrate different levels of confidence in building abilities, problem-solving self-efficacy, attitudes toward collaboration, and future interest in STEM compared to children following a traditional structured curriculum. By investigating these dimensions, we aim to contribute to evidence-based curriculum design and instructional practices in STEM education for young learners.

Accordingly, this study addressed the following research questions (RQs):

RQ1. How does participation in an inquiry-based final project influence children’s self-perception of their building abilities?

RQ2. To what extent do inquiry-based final projects enhance children’s problem-solving self-efficacy?

RQ3. How does participation in a collaborative, inquiry-based final project affect children’s attitudes toward peer collaboration?

RQ4. How does participation in an inquiry-based final project influence children’s inquiry-based problem-solving abilities and their belief in applying robotics to real-world contexts?

RQ5. How does an inquiry-based final project influence children’s future interest in pursuing robotics or related STEM fields?

RQ6. Does gender moderate the effect of participating in an inquiry-based final project on: (a) self-perception of building ability, (b) problem-solving self-efficacy, (c) attitudes toward peer collaboration, (d) inquiry-based problem-solving/real-world application, and (e) future interest in STEM?

4.2. Research Design

This study employed a quasi-experimental, post-test-only comparison group design. Random assignment at the individual level was not feasible within the participating schools due to administrative and ethical constraints; therefore, intact classes were assigned to either the research group (traditional robotics instruction with an inquiry-based final project component) or the control group (traditional robotics instruction without the final project). To reduce potential selection bias and strengthen internal validity, we verified baseline equivalence between groups on key demographic and pre-study performance measures (e.g., gender distribution, construction task, mental rotation scores). No statistically significant differences were detected. Both groups received the same core robotics curriculum to minimize instructional variability. The primary focus of the analysis was on post-intervention group differences, and gender-based subgroup analyses were also conducted to examine potential moderation effects. The overall analytical framework, including study stages and key activities, is presented in Figure 4.

4.3. Participants and Setting

The study initially included over 212 first graders; however, only 176 children who completed both pre- and post-study assessments were included in the final analysis. Children with missing data at either time point were excluded to ensure consistency across comparisons. These participants (87 boys and 89 girls), aged 6 to 7, were drawn from multiple classrooms in three public elementary schools.

The children were divided into two instructional groups:

• Research Group (RG) (n = 85): Received traditional robotics instruction with an added inquiry-based final project component (Appendix A).
• Control Group (CG) (n = 91): Received traditional robotics instruction without the final project.

4.4. Instruments and Measures

To comprehensively assess the effects of the inquiry-based final project, a mixed-methods approach was employed, utilizing a combination of self-report questionnaires, objective performance tasks, and post-intervention interviews. For clarity and brevity,

Table 1. Summary of Study Instruments and Measures.

Instrument	Purpose/Construct Assessed	Format & Response Type	Example Item(s)/Scoring
Self-Report Questionnaire	Self-perception of building, problem-solving confidence, collaboration	Emoji-supported yes/no questions	“I feel good about building things.”; yes/no
Construction Task	Hands-on building ability	Build simple model; rated by trained judges	Completed model (Yes/No); Errors (Yes/No)
Mental Rotation Task	Spatial reasoning	5-item paper test; choose correct rotated figure	Score 0–5
Post-Intervention Interview	Deeper exploration of self-efficacy, collaboration, IBL understanding	One-on-one semi-structured interview; emoji yes/no	“When your robot doesn’t work, can you fix it?”

Below we provide only the essential details needed for replication.

summarizes each measure, its construct, response format, and example items.

4.4.1. Self-Report Questionnaire

A child-friendly, emoji-supported survey captured children’s self-perception of building ability, problem-solving confidence, and collaboration (Mellor & Moore, 2014). Each item offered a yes/no choice with smile/sad face emojis to aid comprehension by six- to seven-year-olds.

4.4.2. Construction Task

Children completed a basic construction challenge using a simple building set. Trained judges recorded whether the task was completed successfully (Yes/No) and whether it included any errors (Yes/No).

4.4.3. Mental Rotation Task

Spatial reasoning was assessed with a five-item block rotation worksheet; scores ranged from 0 to 5.

4.4.4. Post-Intervention Interview

At the end of the school year, children participated in individual, semi-structured interviews conducted by a familiar research assistant. All questions used the same child-friendly yes/no format.

The interview included the following items:

• Self-Perception of Building Abilities—“I feel good about building things.”
• Problem-Solving Self-Efficacy—“I feel I can solve problems when things don’t work.” and “When your robot doesn’t work, can you fix it?”
• Attitudes Toward Peer Collaboration—“I enjoy working with my classmates.” and “I would choose to work with a team on my next project.”
• Self-Efficacy in Inquiry-Based Problem-Solving—“I feel I can find a solution to help a farmer with the harvest”, “I can imagine a robot that helps with problems in my town or village” and “When we learn something new, is it better to just get the answer, or is it better to ask lots of questions and figure it out for yourself?”
• Future Interest in Robotics and STEM—“I will be happy to participate in this program next year.”

To assess children’s implicit understanding of inquiry-based learning, we did not request a direct definition. Instead, the final question prompted reflection on their own learning approach (asking questions versus receiving answers). Responses indicated whether children saw themselves as passive recipients of information (“It’s faster to just get the answer”) or active investigators (“I like to find out myself”).

4.4.5. Demographic Data

Children’s gender and age were collected during the interview phase.

4.4.6. Procedure

At the beginning of the school year, all children completed baseline assessments, which included a self-report survey, a hands-on construction task, and a mental rotation task. Throughout the year, all participants received the same core robotics instruction. Both the research and control groups followed the same robotics and programming curriculum throughout most of the program. The intervention period, which took place during the final 4–5 weeks of the program, introduced a differentiation: the research group completed an inquiry-based final project, while the control group continued with standard robotics lessons. At the end of the school year, all children participated in individual interviews using the same set of core items.

4.4.7. Ethical Considerations

The study was approved by the Science Supervisor at the Israel Ministry of Education. Prior to participation, all children and their parents learned about the study’s goals and procedures. Written parental consent was gathered at the beginning of the school year. All data were anonymized using unique participant identifiers to ensure confidentiality and allow pre-post data matching.

4.4.8. Data Analysis

Quantitative data analysis was performed using IBM SPSS Statistics 29. Group equivalence was checked prior to intervention with Chi-square tests for categorical variables and independent-samples t-tests for continuous measures; no significant baseline differences emerged. The same analytical approach was applied to post-intervention data to address the primary research questions. Gender moderation was explored by repeating analyses separately for boys and girls.

5. Results

5.1. Baseline Group Equivalence Analysis

A preliminary analysis assessed the equivalence of the two study groups on key baseline measures. The comparison was based on children’s self-reported perceptions (confidence in building, ability to follow picture instructions, and problem-solving), judged performance on the blueprint task (completion and errors), and performance on the mental rotation test.

Table 2. Baseline comparisons between research and control groups on pre-test measures.

Measure	Project Group (n = 85)	Control Group (n = 91)	Test Statistic	p-Value
I feel good about building things (%)	83.5	82.4	χ2(1) = 0.038	0.845
I can follow picture instructions (%)	82.4	80.2	χ2(1) = 0.131	0.717
I can solve problems when things don’t work (%)	78.8	76.9	χ2(1) = 0.092	0.762
Completed blueprint (judge, %Yes)	80.0	70.3	χ2(1) = 2.192	0.139
Completed with errors (judge, %Yes)	76.5	63.7	χ2(1) = 3.386	0.066
Mental rotation (M, SD)	0.56 (0.53)	0.50 (0.58)	t(11) = −0.17	0.867

summarizes these comparisons.

Chi-square tests of independence were used for the binary questionnaire and judged outcomes, while an independent-samples t-test was applied to the continuous mental rotation scores. As shown in Table 2, no statistically significant differences were found between the Project Group and the Control Group on any baseline measure (all p > 0.05). This finding supports the baseline equivalence of the two groups, strengthening the validity of subsequent post-intervention comparisons.

5.2. Results for RQ1: Self-Perception of Building Abilities

The first research question examined whether participation in an inquiry-based final project influenced children’s self-perception of their building abilities. To explore this, participants were asked whether they agreed with the statement: “I feel good about building things.” Responses were collected in a dichotomous yes/no format. The results are shown in

Table 3. Self-Perception of Building Abilities by Group.

Outcome	Group	No (n, %)	Yes (n, %)	Chi-Square Test
I feel good about building things	RG	13 (15.3)	72 (84.7)	χ2(1) = 0.610 p = 0.435
	CG	18 (19.8)	73 (80.2)

.

The chi-square test for independence assessed the null hypothesis (H₀): There is no relationship between participation in the inquiry-based final project and children’s self-perception of their building abilities, against the alternative hypothesis (H₁): Self-perception of building abilities is related to participation in the inquiry-based final project. The results indicated no statistically significant association between group membership and reported self-perception: χ²(1, N = 176) = 0.610, p = 0.435 Although children in the final project group reported slightly higher self-perception of building abilities, this difference was not statistically significant, suggesting that the inclusion of an inquiry-based final project did not measurably impact children’s confidence in their general building skills.

5.3. Results for RQ2: Problem-Solving Self-Efficacy

The second research question examined the extent to which participation in an inquiry-based final project enhanced children’s problem-solving self-efficacy. To assess this, participants were asked to respond to two items: “I feel I can solve problems when things don’t work” and “When your robot doesn’t work, can you fix it?” Responses were collected in a dichotomous yes/no format. The results are shown in

Table 4. Problem-Solving Self-Efficacy by Group.

Outcome	Group	No (n, %)	Yes (n, %)	Chi-Square Test
I feel I can solve problems when things don’t work	RG	15 (17.6)	70 (82.4)	χ2(1) = 5.42 p = 0.020
	CG	30 (33.0)	61 (67.0)
When your robot doesn’t work, can you fix it?	RG	14 (16.5)	71 (83.5)	χ2(1) = 6.38 p = 0.012
	CG	30 (33.0)	61 (67.0)

.

The chi-square tests revealed statistically significant positive associations for both items. The Research Group was significantly more likely than the Control Group to report confidence in their ability to solve problems when things don’t work (χ²(1, N = 176) = 5.42, p = 0.020) and confidence in their ability to fix the robot when it didn’t work (χ²(1, N = 176) = 6.38, p = 0.012). These findings suggest that the inclusion of an inquiry-based final project contributed meaningfully to strengthening children’s problem-solving self-efficacy.

5.4. Results for RQ3: Attitudes Toward Peer Collaboration

The third research question examined how participation in a collaborative, inquiry-based final project influenced children’s attitudes toward peer collaboration. To explore this, participants were asked to respond to two items: “I enjoy working with my classmates” and “I would choose to work with a team on my next project.” Responses were collected in a dichotomous yes/no format. The results are shown in

Table 5. Attitudes Toward Peer Collaboration by Group.

Outcome	Group	No (n, %)	Yes (n, %)	Chi-Square Test
I enjoy working with my classmates	RG	13 (15.3)	72 (84.7)	χ2(1) = 4.49 p = 0.034
	CG	26 (28.6)	65 (71.4)
I would choose to work with a team on my next project	RG	12 (14.1)	73 (85.9)	χ2(1) = 5.89 p = 0.015
	CG	28 (30.8)	63 (69.2)

.

The chi-square tests revealed statistically significant positive associations for both items. Compared to the Control Group, children in the Research Group were significantly more likely to report that they enjoy working with their classmates (χ²(1, N = 176) = 4.49, p = 0.034) and that they would choose to work with a team on their next project (χ²(1, N = 176) = 5.89, p = 0.015). These findings suggest that participation in the inquiry-based final project contributed positively to children’s enjoyment of collaboration and their willingness to engage in team-based learning experiences in the future.

5.5. Results for RQ4: Self-Efficacy in Inquiry-Based Problem-Solving

The fourth research question examined whether participation in an inquiry-based final project influenced children’s inquiry-based problem-solving abilities and their belief in applying robotics to real-world contexts. To explore this, participants were asked to respond to three items: “I feel I can find a solution to help a farmer with the harvest,” “I can imagine a robot that helps with problems in my town or village,” and “When we learn something new, is it better to just get the answer, or is it better to ask lots of questions and figure it out for ourselves?” Responses were collected in a dichotomous yes/no format. The results are shown in

Table 6. Self-Efficacy in Inquiry-Based Problem-Solving by Group.

Outcome	Group	No (n, %)	Yes (n, %)	Chi-Square Test
I feel I can find a solution to help a farmer with the harvest	RG	18 (21.2)	67 (78.8)	χ2(1) = 6.24 p = 0.012
	CG	35 (38.5)	56 (61.5)
I can imagine a robot that helps with problems in my town or village	RG	14 (16.5)	71 (83.5)	χ2(1) = 5.64 p = 0.018
	CG	29 (31.9)	62 (68.1)
When we learn something new, is it better to just get the answer, or is it better to ask lots of questions and figure it out for ourselves?	RG	13 (15.3)	72 (84.7)	χ2(1) = 6.65 p = 0.010
	CG	29 (31.9)	62 (68.1)

.

Across all three items, chi-square analyses revealed statistically significant differences between groups. Compared to the control group, children in the inquiry-based final project group were more likely to believe that they could devise solutions to real-world challenges such as supporting farmers (χ²(1, N = 176) = 6.24, p = 0.012), imagine robots that could address community problems (χ²(1, N = 176) = 5.64, p = 0.018), and endorse an inquiry-oriented approach to learning over simply receiving answers (χ²(1, N = 176) = 6.65, p = 0.010).

Taken together, these findings suggest that the inquiry-based final project not only enhanced children’s confidence in applying problem-solving skills to practical and societal issues but also encouraged a mindset aligned with inquiry-driven learning. Participation in such projects appears to help children bridge abstract robotics activities with meaningful real-world applications, reinforcing both creativity and problem-oriented thinking.

5.6. Results for RQ5: Future Interest in Robotics and STEM

The fifth research question examined whether participation in an inquiry-based final project influenced children’s future interest in pursuing robotics or related STEM fields. To assess this, participants were asked to respond to the item: “I will be happy to participate in this program next year.” Responses were collected in a dichotomous yes/no format. The results are shown in

Table 7. Future Interest in Robotics and STEM by Group.

Outcome	Group	No (n, %)	Yes (n, %)	Chi-Square Test
I will be happy to participate in this program next year	RG	15 (17.6)	70 (82.4)	χ2(1) = 2.42 p = 0.120
	CG	25 (27.5)	66 (72.5)

.

The chi-square test revealed no statistically significant association between group membership and reported future interest, χ²(1, N = 176) = 2.42 p = 0.120. Although the Project Group reported higher enthusiasm (82.4% vs. 72.5%), this difference suggests that the intervention did not produce a statistically reliable increase in children’s future interest beyond the high baseline level shared by both groups.

5.7. Results for RQ6: Gender Moderation

To address the final research question, whether gender moderates the relationship between participation in an inquiry-based final project and participants outcomes, analyses were conducted separately for boys and girls. The goal was to determine whether the impact of the final project varied by gender across the following domains: (a) self-perception of building ability, (b) problem-solving self-efficacy, (c) attitudes toward peer collaboration, (d) self-efficacy in inquiry-based problem-solving, and (e) future interest in robotics and related STEM fields. The dataset was split by gender, and comparisons were made to assess whether statistically significant differences emerged between boys and girls in response to the intervention.

Table 8. Gender Moderation of Children’s Outcomes in Inquiry-Based Robotics.

Item	Measure	Gender	Chi-Square Values
Self-Perception of Building Abilities	I feel good about building things	boys	χ2 = 0.482, p = 0.488
		girls	χ2 = 0.259, p = 0.611
Problem-Solving Self-Efficacy	I feel I can solve problems when things don’t work	boys	χ2 = 1.931, p = 0.165
		girls	χ2 = 4.147, p = 0.042
	When your robot doesn’t work, can you fix it?	boys	χ2 = 2.612, p = 0.106
		girls	χ2 = 5.572, p = 0.018
Attitudes Toward Peer Collaboration	I enjoy working with my classmates	boys	χ2 = 1.430, p = 0.232
		girls	χ2 = 3.388, p = 0.066
	I would choose to work with a team on my next project	boys	χ2 = 2.909, p = 0.088
		girls	χ2 = 2.977, p = 0.084
Self-Efficacy in Inquiry-Based Problem-Solving	I feel I can find a solution to help a farmer with the harvest	boys	χ2 = 1.864, p = 0.172
		girls	χ2 = 5.282, p = 0.022
	I can imagine a robot that helps with problems in my town or village	boys	χ2 = 0.851, p = 0.356
		girls	χ2 = 6.487, p = 0.011
	When we learn something new, is it better to just get the answer, or is it better to ask lots of questions and figure it out for ourselves?	boys	χ2 = 2.685, p = 0.101
		girls	χ2 = 4.708, p = 0.030
Future Interest in Robotics and STEM	I will be happy to participate in this program next year	boys	χ2 = 0.464, p = 0.496
		girls	χ2 = 2.687, p = 0.101

summarizes all Chi-square test results by gender and outcome domain. The narrative below highlights significant and non-significant findings for each construct.

The analysis revealed a clear pattern of moderation, with girls showing statistically significant gains on key measures where boys did not. Specifically, participation in the inquiry-based final project significantly enhanced girls’ Problem-Solving Self-Efficacy (both items had p < 0.05) and Self-Efficacy in Inquiry-Based Problem-Solving (all three items had p < 0.03), while boys showed no significant change across any of these five measures. No significant gender moderation was found for Self-Perception of Building Abilities, Attitudes Toward Peer Collaboration, or Future Interest in STEM, though marginal trends were visible for collaboration among girls. These findings suggest that the inquiry-based final project had a disproportionately positive impact on girls’ confidence in problem-solving and inquiry-driven thinking.

6. Discussion

This study investigated the impact of incorporating an inquiry-based final project into early elementary robotics education, focusing on its influence on children’s self-perceptions, problem-solving self-efficacy, collaborative attitudes, inquiry-based orientations, and future interest in robotics and STEM. Across the six research questions, a clear pattern emerged: a single, well-structured culminating project was associated with higher reported cognitive and emotional engagement, with some outcomes particularly strengthened among girls. The discussion below connects these findings to the theoretical framework and explores possible alternative interpretations.

Our findings support the idea that thoughtfully implemented inquiry-based projects foster deep learning and engagement even among very young students (Grossman et al., 2019; Vossen et al., 2018). They also help address the empirical gap concerning project-based learning in early childhood identified by Ferrero et al. (2021). By engaging children in systematic cycles of designing, building, and testing, these projects not only strengthen problem-solving and engineering abilities but also nurture positive attitudes and confidence in STEM (Fan et al., 2018; Vongkulluksn et al., 2018).

6.1. Self-Perception of Building Abilities

Both the research and control groups maintained similarly high levels of confidence in their building abilities, indicating that hands-on robotics exposure itself fosters early self-efficacy, a central principle of Constructionism (Papert, 1980). Repeated opportunities to design, assemble, and manipulate robots appear to have built a strong sense of competence early in the program, consistent with research showing that tangible making can rapidly shape mastery beliefs in young children (Cejka et al., 2006). The lack of additional post-intervention gains may reflect a ceiling effect: once children experience success through iterative construction, confidence stabilizes regardless of subsequent instructional variation. Basic assembly skills may also involve less complex inquiry, making them less sensitive to the deeper questioning and exploration emphasized in the culminating project.

6.2. Problem-Solving Self-Efficacy

The gains in problem-solving self-efficacy examined in this study are directly framed by Social Cognitive Theory (Bandura, 1986). A key finding was that children who participated in the inquiry-based final project reported significantly higher confidence in their ability to resolve issues and fix their robots, as reflected in their responses to both problem-solving items.

This suggests that the open-ended, exploratory nature of inquiry-based projects offers authentic opportunities for students to confront challenges, build resilience, and experience successful mastery, thereby providing the direct mastery experiences necessary to enhance their self-belief (Bandura, 1986). This is consistent with research on design-based learning, which links engaging in systematic processes of designing and testing to enhanced problem-solving abilities and positive self-belief in STEM (Fan et al., 2018). Notably, this positive effect was especially pronounced among girls (see Section 6.6), indicating that such approaches may be particularly effective in boosting technical confidence and addressing persistent gender disparities in STEM education.

6.3. Attitudes Toward Peer Collaboration

Participation in the inquiry-based project was associated with more positive attitudes toward teamwork. Children in the research group were significantly more likely to report enjoying collaboration with peers and expressed a stronger preference for team-based projects. These findings align directly with the Relatedness component of Self-Determination Theory (Ryan & Deci, 2000), which posits that social connection fosters intrinsic motivation.

Furthermore, this result is supported by Social Cognitive Theory, as the collaborative inquiry structure facilitates learning through peer modeling and shared strategy development, promoting cooperation and communication. While significant, the effect sizes here were modest, suggesting that while the inquiry format supports collaboration, attitudes toward teamwork are also strongly shaped by the social environment of the entire program.

6.4. Self-Efficacy in Inquiry-Based Problem-Solving and Real-World Application

One of the most robust findings of this study was the significantly greater self-reported self-efficacy among children in the culminating inquiry-based project regarding their inquiry mindset and their perceived ability to apply their robotics learning to real-world contexts. Children who completed the project reported greater confidence in using robotics to address practical challenges and consistently endorsed asking questions and figuring things out over simply receiving answers.

This outcome reflects key principles of Inquiry-Based Learning (IBL) and constructivism: by positioning the project after a year of foundational robotics instruction, children were equipped with the prior knowledge necessary to transfer concepts into novel, open-ended scenarios (de Jong et al., 2023; Pedaste et al., 2015). Our approach also resonates with Baker (2013) and Wang et al. (2024), who emphasize the value of structured scaffolding, hands-on exploration, and computational thinking in preparing young learners for effective inquiry. Meaningful, child-friendly narratives—such as helping farmers with harvest or saving astronauts—likely increased engagement and provided authentic contexts for applying knowledge (Michalopoulou, 2014).

Importantly, this finding contributes to ongoing debates about the impact of project-based learning in early childhood education, where empirical evidence remains limited (Ferrero et al., 2021). Our results suggest that well-designed inquiry-based projects can foster children’s self-reported confidence in applying what they learned to new, real-world situations. Because we did not assess the correctness of solutions or direct knowledge transfer, these findings should be interpreted as changes in perceived ability rather than confirmed performance. Notably, gender analysis revealed particularly strong gains for girls, reinforcing the value of inquiry-based robotics as a strategy to promote more equitable engagement and confidence in early STEM education.

6.5. Future Interest in STEM

The vast majority of participants, regardless of group, expressed a strong desire to continue studying robotics. This reinforces the idea that ER is inherently engaging and motivating for first graders. Consistent with prior research demonstrating high levels of interest in robotics among young children (Bers, 2019; Zviel-Girshin et al., 2020, 2021), both groups showed substantial enthusiasm. Although children in the RG reported a higher level of future participation interest (83.5%) compared to the CG (72.5%), this difference did not reach statistical significance.

This suggests that while educational robotics is an effective vehicle for fostering interest in STEM, the addition of a single inquiry-based final project may not, on its own, be sufficient to generate statistically measurable increases in motivation within the short timeframe of the study. Rather, the inquiry-based approach may have contributed more to the depth of engagement (both cognitive and emotional) than to incremental increases in reported future interest. One possible explanation is a ceiling effect: the intrinsic appeal of robotics may already be so high that it masks additional motivational gains from the project.

Future research should therefore investigate whether a more sustained or programmatic integration of IBL produces a stronger long-term effect on children’s continued interest in STEM. Sustaining and deepening engagement may require not only capitalizing on the intrinsic appeal of robotics but also combining inquiry-based projects with complementary strategies that nurture enduring interest and commitment.

6.6. Gender as a Moderator

Perhaps the most striking finding of this study concerns the pronounced gender moderation effects observed across several outcome measures. Gender-based subgroup analysis revealed that the positive effects of the inquiry-based final project were more consistently significant for girls than for boys, particularly in the domains of problem-solving self-efficacy and inquiry-driven problem-solving.

These results align directly with the theoretical framework’s rationale for including gender as a moderating factor, suggesting that girls may be especially responsive to the collaborative, open-ended, and context-rich nature of Inquiry-Based Learning (Beisser, 2012; Montuori et al., 2022). This supports the idea that the design, emphasizing mastery and social validation (as per Social Cognitive Theory), effectively built self-efficacy in young girls. This is crucial given the limited research on gender differences in young children’s robotics engagement (Montuori et al., 2022). The study demonstrates that instructional design can differentially support learners, which is further supported by findings that girls can excel under certain instructional conditions and benefit differently from targeted scaffolding (Angeli & Valanides, 2020). Conversely, while not a self-efficacy measure, prior work suggests IBL equitably trains metacognitive skills across genders (Nunaki et al., 2019).

The significant gains in girls’ problem-solving confidence are noteworthy as they counteract traditional gendered patterns in STEM engagement, reinforcing the importance of early interventions. In contrast to research with older students that reports higher robotics self-efficacy among boys (Atmatzidou & Demetriadis, 2016; Kucuk & Sisman, 2020), our study found no gender differences in young children’s desire to learn robotics or in self-perceived building abilities. This contrast suggests that gender disparities in confidence are not innate but develop over time, highlighting the potential of early IBL robotics activities to play a protective role by strengthening girls’ technical confidence before later gender gaps emerge.

While no significant effects were found for boys in these areas, indicating they may have benefited similarly from both instructional approaches, the overall findings underscore the potential of inquiry-based robotics curricula to promote more equitable engagement in STEM, specifically by boosting problem-solving confidence among young girls, thus fulfilling a core aim of the study.

7. Conclusions and Future Directions

This study demonstrates that integrating inquiry-based final projects into early robotics education significantly enhances children’s self-reported problem-solving self-efficacy, collaborative attitudes, and their perceived ability to connect robotics to real-world challenges. While confidence in building skills and future interest in STEM remained high across all instructional formats, the inquiry-based approach offered unique advantages—especially for girls—in fostering meaningful engagement and a mindset aligned with STEM practices.

Our findings contribute to several key theoretical frameworks. From a self-efficacy perspective, the results support Bandura’s theory by showing that the authentic, open-ended problems in the project provided students with mastery experiences, thereby strengthening their problem-solving confidence. From a constructivist learning perspective, the positive effects on real-world application and inquiry-oriented thinking support the importance of hands-on, meaningful experiences in which students actively construct their own understanding.

The differential gender effects observed in the study offer valuable insights into how instructional approaches interact with individual learner characteristics. The findings suggest that inquiry-based learning may be especially effective in supporting girls’ development in STEM domains, where confidence gaps have been historically noted.

For educators and curriculum designers, these findings offer several practical considerations. First, the benefits to problem-solving confidence and real-world application justify the time and resources required for such projects. Second, the gender-specific effects emphasize the importance of using diverse instructional approaches to meet the needs of all learners. Finally, the positive impact on collaboration attitudes demonstrates that these projects serve a dual purpose, enhancing both cognitive outcomes and social-emotional skills essential for 21st-century education.

At its core, elementary robotics education offers a unique blend of structure and creativity. Through the thoughtful integration of inquiry-based learning, visual programming, and collaborative project work, young participants are empowered to become curious, capable problem-solvers who think critically, collaborate effectively, and express their ideas through tangible technological artifacts. These findings highlight that inquiry-based learning is a powerful and appropriate tool for developing both technical and social-emotional skills in young STEM students.

Future research should focus on confirming and extending these results. Replicating the inquiry-based final project in a range of schools, grade levels, and learner populations would test its generalizability. Longitudinal follow-ups are also needed to examine whether improvements in problem-solving self-efficacy and inquiry-oriented thinking persist over time and contribute to later STEM interest and achievement. In addition, future studies should explore how different gender-responsive instructional designs can maximize benefits for all children. The field would also gain from comparative investigations of various inquiry-based methods and the use of multiple outcome measures to capture a more complete picture of their impact.

8. Limitations

Several limitations should be considered when interpreting these findings. The quasi-experimental design, while appropriate given the educational context, limits causal inferences compared to randomized controlled trials. However, the baseline equivalence analysis strengthens confidence in the group comparisons.

Measurement constraints are another important consideration. All outcomes relied on self-report, which, while developmentally appropriate, may be subject to social desirability bias and limited self-awareness in young children. In addition, the small number of items used to assess each construct may reduce reliability and depth. Although the scales were adapted from prior early childhood robotics studies and pilot-tested for clarity, future research should validate and expand these instruments and incorporate observational or performance-based measures to triangulate findings.

The short duration of the final project and the use of mostly dichotomous measures may underrepresent nuanced changes. Performance-based measures or confidence scales could provide richer evidence in future work.

The study’s focus on immediate post-intervention effects leaves open questions about the persistence of these benefits unanswered. Longitudinal follow-up would provide valuable insights into whether the positive effects on problem-solving confidence and inquiry-oriented thinking endure over time.