Building an Autonomous Car: Designing, Implementing, and Evaluating an Integrated STEM Teaching–Learning Sequence for Pre-Service Secondary Teachers

Abstract

This paper presents the design of an integrated STEM education teaching–learning sequence (TLS) for secondary education and the adaptation of this design for the training of future science teachers, as well as the implementation and evaluation during the academic years 2022/2023 and 2023/2024 in the master’s degree in secondary teacher training. This is an integrated STEM education project that seeks to design the prototype of an autonomous car using the mBot robot as a base. Thus, it allows for the integration of physics kinematics with robotics programming guided by an engineering design. This study was carried out with 43 pre-service teachers, and the impact on both content and procedural knowledge and attitudes was analyzed. The results show an increase in knowledge; reflect the usefulness of the tools used to work on design, evaluation, and optimization procedures; and, finally, a change in the students’ emotions towards a more positive perception of the disciplines involved and the subject to be dealt with in the project.

1. Introduction

Integrated STEM (science, technology, engineering, and mathematics) education is a reality in our education system and classroom environment, which has created two main challenges: the need for integrated STEM education material to use with students and the need for teachers’ training on this topic.

Starting with the first of the issue, integrated STEM education is proposed as a possible solution to the problems facing the education system today (Millar, 2020), such as the lack of student interest (Bubnick et al., 2016; Thibaut et al., 2018a; Toma, 2019; Vasquez et al., 2013) and the need to enforce STEM vocations (Levrini et al., 2017; Mejias et al., 2021; Zhou et al., 2022) and literacy in society on the STEM sectors (Holmlund et al., 2018; Thibaut et al., 2018a; Zhou et al., 2022). However, implementing integrated STEM education in the classroom faces significant challenges, as the process is hindered by a diverse range of theoretical and material approaches (Martín-Páez et al., 2019; Pérez-Torres et al., 2021). In other words, STEM projects vary widely in their conception, which in turn hinders their creation, implementation, and evaluation (Chu et al., 2019; English, 2017; Ortiz-Revilla et al., 2022). This great variability also appears in the very terms that are used to refer to this educational approach, such as STEM education, integrated STEM education, iSTEM education, and STEAM education. To ensure clarity, we will adopt the term integrated STEM education or iSTEM education to distinguish it from discipline-specific approaches. Likewise, while STEAM education shares some characteristics with iSTEM (Aguilera & Ortiz-Revilla, 2021; Mejias et al., 2021), our focus remains on the latter since Arts as independent discipline does not appear in the teaching–learning sequence (TLS) design presented in this study. Nonetheless, creativity is inherently present in STEM problem-solving, particularly within the engineering design process (Mejias et al., 2021).

Taking all that into account, there is a need to identify criteria and guidelines to enable and guide the selection, design, and adaptation of integrated STEM education projects (Aguilera & Ortiz-Revilla, 2021; Kelley & Knowles, 2016; Moore et al., 2014; Thibaut et al., 2018a). The study presented here follows the theoretical agreements analyzed in the systematic review by Portillo-Blanco et al. (2024) and presents the design of a teaching–learning sequence (TLS) following these criteria. Thus, in the present study, we define iSTEM education as the integration of content, skills, and procedures from at least two disciplines that form the acronym by seeking a solution to a real-world problem (Halawa et al., 2024; Martín-Páez et al., 2019).

Moreover, this new approach presents a challenge for teachers, as they report difficulties in envisioning what iSTEM education should look like, leading to a lack of self-confidence (Arshad & Al, 2021; Shernoff et al., 2017). In addition, the need to implement iSTEM education in classrooms has introduced a new discipline that science teachers are not usually familiar with: engineering (National Research Council, 2012). Several studies highlight the engineering design process as a key driving force in iSTEM TLSs, guiding their structure and development (Dare et al., 2018; Fan et al., 2021; Roehrig et al., 2021; Zhou et al., 2022). However, current teachers face difficulties in understanding the engineering design process itself and in integrating the development of scientific knowledge throughout the process (Cunningham & Carlsen, 2014; Dare et al., 2018).

Professional Development (PD) courses have been shown to have a positive impact on teachers’ attitudes towards iSTEM education as well as their knowledge on the topic (Thibaut et al., 2018b, 2018c). Not only is training in iSTEM education important, but it is also essential to focus on participation in PD programs specifically aimed at the engineering design process to address the existing knowledge gap (Dare et al., 2018; Spikic et al., 2022). Additionally, Lo (2021) defined seven design elements for effective teacher PD based on the work performed by Darling-Hammond et al. (2017). Among these, three are particularly relevant to this study: Content focus, use of models and modeling, and active learning. This refers to the need for teachers to actively participate in instructional models that are focused on the content that they will teach in their classroom and context (Darling-Hammond et al., 2017; Lo, 2021; Spikic et al., 2022). Thus, in this study, we present the design of an integrated STEM education TLS for secondary education students and the implementation and analysis of its impact on the secondary education teacher-training master’s degree students. Given the objectives of this study, the following research questions were formulated to guide the investigation:

How does the iSTEM TLS impact the students’ physics and programming content knowledge and procedures?

How does the presented iSTEM TLS influence students’ attitudes towards the iSTEM project’s topic?

1.1. Building an Autonomous Car: An Integrated STEM Education Project for Secondary Education

This section describes the TLS design implemented with pre-service secondary education teachers. The project, designed for secondary students aged 15–16, is set within the context of a car company aiming to prototype an autonomous vehicle. In this scenario, students need to program an mBot robot as a car prototype, and they have three objectives to fulfill: respect the speed, brake before hitting an obstacle, and maintain the security distance between vehicles. As mentioned, this TLS was designed for secondary education students following the education curricula of two main subjects: Physics for kinematics and Technology for programming and robotics (Basque Government, 2023). In addition, the design criteria for each challenge are presented in meters per second (m/s), whereas the robot is programmed by the engine power of the motor. This means the students need to link those two magnitudes to tackle their specific challenge in the project.

As shown in Figure 1, the approach for the first two challenges follows a consistent process. After presenting the project and identifying their learning goals, students first engage in the content knowledge phase for both kinematics and robotics. Then, they design the program for the specific challenge, and finally, they evaluate and optimize another group’s design. For the third objective of the project about maintaining the security distance between vehicles, there is no new content knowledge, so they only have to do the design and evaluation phases. All the work is performed in teams.

This TLS is grounded in the principles of iSTEM education as outlined by Portillo-Blanco et al. (2024). First, the project fits within the general approach of current education framed in socio-constructivism, where the focus on competence development and active learning are essential. Additionally, the integration principle is present in this TLS in several ways: although the main importance of the project is focused on two disciplines with similar importance (Physics and Technology), engineering creates the axis of the sequence by the engineering design process, and mathematics is present both in the kinematics activities and in some of the challenges for the students. Likewise, the project starts with a real-world context, which is the development of autonomous cars, and the presentation of the challenges or problems they need to answer. This TLS follows an engineering design process and presents a problem that is open-ended, complex, extended, and ill-structured since each challenge has different criteria and the speed of the mBot robots varies from each other. The design principle of the project is best described as a design–built–test engineering category (Purzer et al., 2022), where students engage the design process starting from clearly stated design requirements, and they need to develop, test, and evaluate prototypes.

Another principle identified in the review by Portillo-Blanco et al. (2024) for an iSTEM project is inquiry, which is defined as the “need to know” process to enhance the effectiveness of engineering practices in design and prototype evaluation. Thus, in this TLS, inquiry and scientific practices are mainly present in the content knowledge phase of the project via active learning activities regarding kinematics. Finally, teamwork is integrated throughout the entire TLS, fostering cooperation in each phase of the project.

As explained in the Introduction Section, this project was implemented at the teacher-training master’s degree in the Innovation in Natural Sciences subject. The subject’s objective is to introduce students to iSTEM education and provide examples of projects. At the same time, this creates the opportunity for them to experience the projects as students and analyze them as teachers. This means that the students carried out the activities as they were designed, acting as secondary school students while learning or reviewing kinematics and robotics content. Afterwards, with the guidance of the course instructor, they reflected on the reasoning behind the activities when designing this iSTEM TLS. Moreover, implementing this specific TLS in the training of future secondary school teachers allows them to see how scientific content is integrated into an engineering design process, helping to overcome the challenges highlighted in the literature (Dare et al., 2018).

Thus, due to time management of the subject’s content, it was impossible to carry out the sequence as it had been designed for secondary school students. Moreover, it should be taken into account that the students of this master’s degree come from scientific–technological degrees, and that they all have a prior knowledge of kinematics. So, changes were made to the sequence to reduce the activities to achieve the necessary knowledge to accomplish the first objective, which is the one used with these pre-service teachers.

1.2. Adapting the iSTEM TLS ‘Building an Autonomous Car’ for the Teacher-Training Master’s Degree Students

Based on this premise explained above, the project was presented to the students, and they identified the learning goals necessary to complete it. After this first presentation, as they were only performing the activities for the first objective of the project about respecting speed, the content on uniform motion in one direction was covered. To begin, they started with a general review of the first activities (Figure 2, A3–5) that focus on the concepts of position, displacement, distance traveled, and reference system. Then, individually, they carried out Activity 6 (Figure 2) where the students received a time record of a cyclist passing through different points of a Tour de France stage. Using these data, they had to specify the time and position by placing the origin at the first curve and repeating the process but with the origin at the twelfth curve. They then reviewed the formula for uniform motion in one direction using three simple exercises (Figure 2, A7–9) to see if any questions arose before moving on to graph creation and interpretation. Activities 10 and 11 focused on graphing a path explained in the activity statement and specifying the object’s position at different moments. Activity 12, on the other hand, asked them to represent four different position–time (s/t) graphs based on their movement in the classroom. To accomplish this, groups of four people had to establish a common reference system to draw the four graphs correctly (Orero, 2021).

In order to reinforce the key kinematics content for this first objective of the project, the last exercise was a problem-solving activity following the jigsaw methodology in trios. The students were given three problems to solve following three phases and rotating the problem at the end of the assigned task: (1) approach and modeling of the problem, (2) resolution, and (3) analysis of the result and evaluation. The three problems presented sought to work on different uniform motion in one direction contents and procedures, so there was one focused on the calculation of travel time according to speeds, another on the interpretation of a graph, and the last one on the creation of a graph based on a series of data. This made it possible to review all the necessary knowledge for the first objective of the project.

On the other hand, the content-learning phase for the robotics part (Figure 2) was based on eight simple challenges for the students to learn how to use the mBlock program required for the robot they were using (mBot). These challenges focused mainly on learning about the particularities of the robot and speed control by defining different engine power percentages.

The first objective is presented to the students as a set of six runs composed of two-speed intervals defined in meters per second each. In order to program the robot to respect the two speeds requested in the challenge, they must create a relationship between the power level defined in the mBlock program and the speed of the robot, and it is performed using the Tracker program. In fact, Tracker is a free video-modeling and analysis tool based on the Java Open Source Physics (OSP) framework (Tracker video analysis and modeling tool for physics education, n.d.). The program allows you to upload recorded videos of a moving object, and, defining the reference system and the point to track, it returns the position–time graph and the data table. Using this program, students must identify the real speed of the robot used by their group in relation to six percentages of motor power (30–80%).

Once this task has been completed, students should reflect on and justify their design in the design report, so that the evaluating team can have as much information as possible in the next phase. Therefore, after the design, all the groups proceeded to analyze the work of another group in the classroom, assessing both compliance with the criteria of their specific challenge and the correctness of the program designed for it. Compliance with the speed criteria is measured once again using the Tracker program. The design and evaluation reports are shown in Figure 3.

2. Materials and Methods

2.1. Implementation Context

This integrated STEM education TLS was implemented in the teacher-training master’s degree at the University of the Basque Country (UPV/EHU). More precisely, it was included in a subject called Innovation in Natural Sciences, for students enrolled in the Natural Sciences and Mathematics module. The TLS was carried out over two consecutive courses during the academic years 2022–2023 and 2023–2024 with 22 students in the first year and 21 in the second year. Although the master’s module is designed for students from any scientific discipline, what stands out about these two study groups is that almost all students had studied biology-related subjects, with only four graduates in chemistry and one in physics. Additionally, the TLS was conducted in the classroom by one of the authors of this study, who was highly knowledgeable about the project and its implementation.

The students worked in trios created by themselves for three sessions of 2.5 h each in this project, and after experiencing it as students, they could analyze the material provided to secondary education teachers to understand the theoretical underpinnings of the sequence and why design decisions were made.

2.2. Research Tools

To answer the first research question regarding the impact of the TLS on students’ knowledge of content and procedures, two evaluation tools were used: a pre–post questionnaire and the design and evaluation reports. Meanwhile, the analysis for the second research question on the impact on students’ attitudes was made with a questionnaire at the end of the implementation.

2.2.1. Pre–Post-Questionnaire

By identifying the key contents to be learned in this project, two questionnaires were created with questions aimed at each of them. This resulted in four kinematics questions that were distinct but equivalent between the pre-test and post-test (Table 1). The decision to use different but comparable questions in both assessments was made to prevent students—who are highly motivated and invested in their academic performance—from discussing their answers between tests, which could potentially influence the results. Additionally, given the short time span between the pre-test and post-test, there was a higher likelihood that students might remember their previous responses. However, all questions aligned with the same learning objectives, and after expert review, the more challenging version of each pair was placed in the post-test to better capture the knowledge gained during the sequence.

Similarly, given that the use of the robot and programming was completely new for the students participating in the study, a question on programming was added to the post-test to analyze the point of knowledge reached in this respect (Table 1). They completed the questionnaires individually and by hand.

2.2.2. Design and Evaluation Reports

One of the key phases in the engineering design process includes designing, evaluating, and optimizing the prototype. The design and evaluation reports are intended to assess students’ performance in these processes (Figure 2). When designing the mBot robot program according to the criteria of the selected challenge, the students had to fill in the report that was intended to guide and facilitate the design process (Figure 2). They had to identify the data they knew and the data they needed, specify the calculations performed, put the designed program into mBlock, and explain and argue why the design was performed in this way.

In addition, after the design, the students had to evaluate the work of another group. In the evaluation report (Figure 2), they had to include the graph obtained in the Tracker program by analyzing the video made by the other group’s robot. They then had to explain whether they had met the criteria required in the challenge, showing the calculations made to draw these conclusions and whether there was a way to optimize the program. They were also asked to rate the quality of the design in three items on a Likert scale, and, finally, they were asked to draw the v/t graph complementary to the analysis just carried out.

2.2.3. Attitudes Questionnaire

The aim of this questionnaire is to analyze the change in emotions generated by facing this sequence after the presentation of the project and after its completion. To do this, the questionnaire designed and validated by Nicolás et al. (2021) was used, in which the students, based on a list of emotions, had to select the one that best represented what they felt before and after the project. They were also asked to give reasons for their selection.

2.3. Results Analysis

2.3.1. Pre–Post-Questionnaire

The pre–post-questionnaires consisted of four open-ended questions in the pre-test and five in the post-test that students responded to individually, all aligned with the TLS’s learning objectives for both kinematics and robotics (Table 1). A mixed-methods approach was employed through the analysis of qualitative responses by a phenomenography-based approach, which classify them into distinct categories through an iterative process (Guisasola et al., 2023; Hajar, 2021).

Given that this study combines data from two consecutive school years, the initial analysis and categorization of responses were conducted by a single researcher following the first implementation (2022/2023 academic year). After the second implementation in the following academic year (2023/2024), the same researcher analyzed the student responses using the categories pre-established in the first analysis and confirmed their applicability in another study group.

To assess the study’s reliability, the same researcher re-categorized all the pre–post-questionnaire responses one month after the last analysis. For all questions, a weighted Cohen’s Kappa index above 0.8 was obtained, indicating strong agreement. For the results discussed in this paper, a second researcher independently categorized the responses, and the final categories were determined through consensus, ensuring robustness and accuracy in the findings.

The data are presented using alluvial diagrams to illustrate the evolution of each student and the shifts in category between the pre-test and the post-test (Mauri et al., 2017). Likewise, the results for the last question in the post-test regarding the programming content are presented using the bar graph to summarize the students’ categorization.

2.3.2. Design and Evaluation Reports

As can be noted in Figure 3, the process of designing and evaluating the program for each challenge is divided into different boxes in the tables that the students complete by working in groups. Taking into account the extent to which the procedures were managed, as established by the organization of the design and evaluation report tables, the analysis focused on a qualitative-based analysis of the compliance with the procedures in each of the sections. For program design, the following aspects were evaluated:

The relationship between speed and motor engine power: Students calculate the real speed of the mBot robot at different motor powers in order to design the program.

Specification of calculations: They show the calculations they have made to arrive to the conclusion of the engine power they need for the selected design challenge.

Presentation of the program created in mBlock.

Written explanation of the process: They provide a written explanation of their design decisions based on the calculations made and the criteria of the challenge to which they have to respond, relating the contents of kinematics to the program designed.

For the evaluation report, the analysis considered whether the group presented the following elements:

The graph obtained in Tracker: To arrive at this graph they had to record the video of the robot to be evaluated, set the reference system and the point to follow, obtaining the s/t graph of the robot’s movement.

Necessary calculations for evaluating the mBlock program according to the design criteria: Starting from the s/t graph obtained in Tracker, they calculate the robot’s movement velocities.

Conclusion of the evaluation: They compare the results of their calculations with the design criteria and assess the appropriateness of the program designed for that specific challenge.

Any optimization recommendation (if applicable): In case of errors in the program, they propose and argue how to solve them, as well as possible improvements in the programming for the challenge.

Creation of the v/t graph based on the s/t graph obtained in Tracker.

2.3.3. Attitudes Questionnaire

The students, using a google form, answered several questions related to their opinion about the subject and the examples of projects they had participated in. In this article, we focus on the analysis of the emotions identified during the presentation of the project ‘building an autonomous car’ and after its completion. Four questions were used for this analysis: to examine the emotions felt when the iSTEM TLS was first presented, students were asked to select the emotions they believed best represented their feelings at that moment, choosing as many as they wanted from a predefined list (enthusiasm, tranquility, fear, enjoyment, tension, worry, fun, motivation, frustration, overwhelm, boredom, curiosity, confidence, anxiety, and insecurity). Immediately afterward, they were asked to explain the reasons for their selection. The same structure was repeated in the next two questions, using the same list of emotions and justification process, but this time referring to their emotions upon completing the project. As a result, the analysis in this section followed a mixed-methods approach. The selection of emotions provides quantitative data, allowing for an examination of the occurrence of each emotion at the beginning and end of the TLS. Meanwhile, the justifications offer qualitative insights by revealing the reasons behind each selection. Similarly, by classifying the emotions into positive (enthusiasm, tranquility, enjoyment, fun, motivation, curiosity, and confidence) and negative (fear, tension, worry, frustration, overwhelm, boredom, anxiety, and insecurity) categories, the percentage occurrence of each group was measured.

3. Results

3.1. How Does the iSTEM TLS Impact the Students’ Physics and Programming Content Knowledge and Procedures?

To analyze the impact of content learning, we examined four pre–post-test questions (Figure 4), and an additional post-test question focused on robotics content, which was new to this group of students (Figure 5). The categories created for the first four questions follow a similar pattern. Categories A and B refer to correct answers with the highest category detailing data such as velocity, exact position, and the relationship between different variables. In Q1, Category C refers to correct answers without argumentation, while in other questions, it indicates incorrect answers. Specifically, in Q2, Category C includes incorrect answers based on alternative ideas, whereas Category D is reserved for incorrect answers lacking argumentation.

Overall, 92% of students answered the four pre-test questions correctly, and this percentage remained consistent in the post-test. However, a general analysis shows an increase in category A from before to after implementation, from 38% to 69%. This increase indicates that students improve the level of detail of their answers after completing the sequence of activities of the project presented.

A more detailed analysis of the questionnaire reveals that all students correctly identified the fastest-moving object in Question 1. Notably, in the pre-test, most responses emphasized explaining which object traveled the farthest in the shortest time. However, in the post-test, students enriched their answers by incorporating the definition of speed or referencing the graphical representation of the movement to support their arguments.

For the second open-ended question, eight students stated that two objects at the same position have the same speed. However, these misconceptions disappeared after implementation, and all students correctly explained the relationship between position and velocity based on the initial position, time, or the distance traveled by the vehicles.

The third question focuses on the interpretation of s/t graphs and, apart from one student who misused the graph data to explain the movement, the rest answered correctly. The main difference between the pre-test and the post-test is in the number of students who go on to calculate the speed of each section to describe in greater detail the movement represented, rising from 5 to 18.

The fourth question required students to create a graph, transitioning from a written description in the pre-test to converting a v/t graph into an s/t graph in the post-test. In addition, in the pre-test, they had to extract two position data points at different times, and this is where only two students in category C failed. In category B are the students who have made the graph correctly but have derived the position from the graph itself, without performing calculations. In category A are those students who have calculated the velocity and concluded the position accurately. On the other hand, in the post-test, there is an increase in students in category C (13), who have mistranslated the data from one graph to another. Despite these challenges, the majority of students answered this question correctly in the post-test.

Finally, as explained in Section 2.2.1, a question on programming was added only in the post-test, as the students were not familiar with this topic beforehand. After analysis of the answers, 28 students can identify the errors in the program and propose a correct optimization (Figure 5). The remaining 15 do not see the problem in the script, 10 of which propose some optimizations for the presented proposal. This last group demonstrates a basic understanding of the programming language.

Switching to the analysis of procedural performance through design and evaluation reports, we obtained a total of ten design reports (five in each academic year) and eleven evaluation reports (5 in 2022/2023 and 6 in 2023/2024). All groups correctly identified the relationship between real speed and motor power in their mBot robot, as well as presenting the appropriate robot programming for their challenge. However, the five groups belonging to the year 2022/2023 did not show the calculations to reach these conclusions, nor did they give a verbal description of their design. In contrast, the five groups in the second year completed all phases of the design, and three of the groups incorporated the LED display or sounds to their program to increase the complexity of the program.

As for the evaluation of the design of another group, all groups correctly presented the graph obtained in the Tracker program, the calculations for the speed measurement, the evaluation of the program according to the design criteria for that group, and the optimization options where there was room for them. Nevertheless, only five groups—two in the first year and three in the second—successfully created the v/t graph requested at the end of the report.

3.2. How Does the Presented iSTEM TLS Influence Students’ Attitudes Towards the iSTEM Project’s Topic?

The analysis shows notable changes in the emotions of pre-service teachers before and after the project. It is worth highlighting the increase in tranquility and enjoyment and the decrease in fear, worry, and insecurity, among others (Figure 6). At the beginning of the project, students stated that participating in a new project with new technologies generated positive feelings such as motivation and curiosity, as well as fear and insecurity. Likewise, most of the initial negative feelings are related to three main causes according to the future teachers’ explanations: that they do not like physics, that they do not like technology, and, above all, that they have never used a robot or programmed one. Given that these students have little background in physics and robotics, the presence of negative emotions is expected.

On the contrary, after the implementation, most of the students are grateful for the methodology used, as they state that they have understood the content and have been able to achieve the objectives by themselves. Thus, most positive emotions stem from realizing that the project was easier than they initially expected and that they were able to finish the project successfully. They also explain that they found the use of the robot fun, although, in some cases, the problems generated by the robot lead to emotions such as frustration. Similarly, three students still maintain that these two disciplines are not to their liking and that they were bored.

In summary, by classifying these emotions as positive and negative, negative emotions decreased significantly from 47% to 18%, while positive emotions rose from 53% to 82%.

4. Discussion

This article presents an integrated STEM education TLS aimed at secondary school students and modified for the training of future secondary school teachers, as well as its evaluation and impact on student attitudes. The results indicate that this approach effectively integrates physics and technology through an engineering design process, reinforcing kinematics content via inquiry-based activities.

Following the first research question that drives this study, a general increase in the knowledge of kinematics has been recorded by comparing the pre-test and post-test questionnaires. Overall, students transition to using more specific concepts of kinematics, such as calculating velocity or the graphical expression of movements to support their answers with arguments completing the TLS. Moreover, as Q2 of the questionnaire shows, the misconception about two objects that coincide in time and position having the same velocity is overcome, leading students to take into account the starting point, the path, and the time elapsed to challenge this statement (Sutopo & Waldrip, 2014).

Although the results are mostly positive, Q4 of the quiz reveals that a group of students answered incorrectly in the post-test. This question aimed to create a graph and, while, in the pre-test, the students had to go from written explanation to graph, in the post-test, the objective was to translate a v/t graph into an s/t graph. Analyzing the categories recorded in Q4, we see that students answered correctly in the pre-test, demonstrating their ability to create graphs from a written explanation. However, in the post-test, almost a third of the students make mistakes when translating from one type of graph to another. This means that students continue to encounter difficulties in relating the types of graphs to each other despite the activities carried out in the session (McDermott et al., 1987).

These results show two conclusions to be taken into account: (1) the difficulty of the task is greater in the post-test than in the pre-test; that is, the treatment of the information from one graph to another is more complicated than the conversion from words to graphs. Conversely, (2) the negative results in the post-test mean that the suggested tasks are insufficient to overcome this difficulty. It should be noted that the master’s students participating in this study only completed the activities related to the first objective of the TLS. In the original design, more activities on the graphic expression of movement are included in the second objective. For this reason, if only the first objective is performed with the students, some more activities should be added in this aspect to ensure that this difficulty is overcome.

Regarding their knowledge of robotics, starting with students who are new to this sector, we can confirm that the activities proposed have served to increase their expertise in the programming language. Even so, as explained in the previous case, the completion of the full TLS would likely lead to a deeper understanding since the programming knowledge needed to tackle the other two objectives of the project is more complex, as this requires a deeper understanding of the subject.

The project aims to design the prototype of an autonomous car following an engineering design process that conforms to the design–built–test vision described by Purzer et al. (2022), which entails not only the design of the program but also the evaluation and optimization of the program. For this purpose, in this iSTEM education TLS, the students used a report for each process (Figure 2), detailing each of the steps to be followed. The results at a procedural level have been very positive, where most groups have completed all the steps and have detailed the reasons for their decisions. Two important aspects at the procedural level should be underlined: in the design report, the groups of the 2022/2023 course did not give explanations at the request of the teacher due to lack of time, and, on the other hand, the fact that only half of the groups created the v/t graph in the design report may be due to the same difficulty already described in question Q4 of the post-test questionnaire.

The way these two reports are organized and the results obtained suggest that they can be an effective tool to guide students in engineering design steps that are usually overlooked, such as evaluation reasoning (using graphs or calculations) and optimization (Halawa et al., 2024). Future research could explore how pre-service teachers respond when completing the process without guided reports, helping identify the most challenging steps.

Turning to the second research question regarding the impact on the attitudes of future teachers, it is clear that participation in this TLS has positively influenced emotions. According to Lo (2021), Spikic et al. (2022), and Darling-Hammond et al. (2017), professional development for teachers in iSTEM education requires the use of instructional models that enable active participation from the student perspective. Moreover, such models must be linked to topics and content relevant to the teaching context. Thus, the TLS proposed in this study and its implementation fulfill these three requirements as they relate to future teachers of science and technology subjects who will have to deal with iSTEM projects in the classroom. Taking all this into account, the positive development of the recorded emotions could reflect an increase in confidence in both the disciplines involved and the integrated STEM education methodology itself, addressing the needs identified in the literature (Arshad & Al, 2021; Margot & Kettler, 2019; McLure et al., 2022).

It is also important to note that this study has a small sample of 43 students and that the impact on knowledge of content and procedures is based on the initial academic background of students graduating in science-related fields. Similarly, the fact that the profile of the students is scientific and that they are training voluntarily to become teachers may affect their involvement and motivation towards the project, wanting to understand it and learn from it. Although the future teachers played the role of students in this study, further research will be conducted in secondary schools to see how the TLS works in a real classroom setting.

Given that the students came from biology-related fields, the decrease in negative emotions and their explanations regarding concerns about engaging in a project that integrated physics and technology through an engineering design process stand out. As Thibaut et al. (2018c) explain, participation in iSTEM PD improves students’ attitudes toward the methodology, and based on the results obtained in this study, we could assume that we have helped this group of future teachers approach iSTEM education from a more positive perspective.

Thus, the use of this iSTEM TLS with pre-service science and technology teachers aimed to introduce and train them in a scenario they will encounter in their classrooms. Although iSTEM education has been present for several decades, it was officially introduced into the educational curriculum, where this study was conducted (Basque Autonomous Community, Spain), in 2022–2023 (Organic Law 3/2020, 2020). Therefore, the need for training to meet the new requirements of the education law is crucial. Moreover, the inclusion of the engineering perspective is also novel in schools. To address gaps in understanding the discipline, its procedures, and its application in TLSs, presenting students with an iSTEM TLS model like the one in this study helps them develop this new knowledge as teachers.

Following the explanations of Shernoff et al. (2017) and Martins and Baptista (2024), future teachers demand training both on how to implement iSTEM education in the classroom and on the specific content to be taught. Actively participating in iSTEM TLSs supports both perspectives (Martins & Baptista, 2024; Spikic et al., 2022). Furthermore, the improvement in attitudes toward the proposal could also reflect a more positive attitude toward iSTEM education, making it easier for teachers who have completed the master’s program to implement iSTEM education with greater confidence.

5. Conclusions

The results of this study show the impact and implications of the use of a specific iSTEM TLS on trainee secondary school teachers. First, the implemented activities effectively improved the participants’ knowledge of kinematics and robotics. However, the adaptation of the TLS for trainee teachers did not completely overcome difficulties related to the creation of graphs, which emphasizes the need for further refinement in future implementations.

Secondly, the use of guided design and evaluation reports has proven to be a valuable tool for structuring the engineering design process, helping students to reason, evaluate, and optimize their solutions. These structured debriefs facilitated a step-by-step approach to problem solving, reinforcing essential STEM competencies.

Finally, this study provides evidence of a positive change in students’ emotions and attitudes towards physics, technology, and iSTEM education. The implementation of TLS helped to reduce initial concerns and insecurities, fostering greater confidence in both content and methodology. This change is particularly relevant as trainee teachers prepare to integrate iSTEM approaches into their future classrooms.

In conclusion, the iSTEM project presented in this study could serve as a model for teacher education, offering opportunities for improvement in its design while demonstrating its potential to enhance STEM teaching skills and confidence in trainee teachers. Likewise, it presents the design of an iSTEM TLS that could be used in secondary education, so it also aims to contribute to the creation of iSTEM material. In future research, it would be interesting to further analyze the impact of different iSTEM TLSs in teacher education, with the aim of optimizing the learning process and transfer to the classroom in the future. Likewise, as for the TLS on ‘how to build an autonomous car’, an analysis with secondary school students would allow for analyzing its impact on the group of students for whom it was designed and improving the sequence according to the results.

Author Contributions

TLS design, A.P.-B., K.Z., J.G. and J.G.-B.; TLS application, K.Z.; results analysis, A.P.-B.; writing—original draft preparation, A.P.-B.; writing—review and editing, A.P.-B., K.Z., E.G.-J. and J.G.; funding acquisition, K.Z. and J.G. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

According to the academic committee of the institution, formal ethical approval for this study was exempt for this study which was conducted within the teaching framework of the course, all participants provided their consent and the analysis of the results was conducted anonymously.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Any additional information of interest to the reader will be sent upon request to the corresponding author of this study.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Table

Figure 1 TLS structure for the “Building an autonomous car” integrated STEM education project.

Figure 2 TLS adaptation for the teacher-training master’s degree students and specification of the activities involved.

Figure 3 Design and evaluation reports.

Figure 4 Alluvial diagrams illustrating the results for the Q1–Q4 pre–post-questionnaires. Pre-test results are on the left side of the diagram and post-test results on the right. The number of responses in that category is indicated in brackets, and an explanation of each category is given to the right of the diagram. (n = 43).

Figure 5 Bar graph for the Q5 post-test question. An explanation of each category is given to the right of the diagram. (n = 43).

Figure 6 Results for the attitudes questionnaire regarding their emotions before and after the TLS.

Table 1

Questions of the pre–post-questionnaire and the learning objective behind.

Objective	Pre-Test Questionnaire	Post-Test Questionnaire
Concept of speed (Physics)	1. Here there are the initial and final position of two objects and the time to do it. Select which is the fastest (A or B) and explain why.[Image omitted. Please see PDF.]	1. Here there are the initial and final position of two objects and the time to do it. Select which is the fastest (A or B) and explain why. [Image omitted. Please see PDF.]
Concept of acceleration and speed(Physics)	2. A police officer on a motorcycle observes a car committing an offense as it passes her. She then starts pursuing the car until it is caught. Do the car and the motorcycle have the same speed at any point? Justify your answer and, if they do, specify when.	2. Indicate whether the next question is true or false and why: Do two objects have the same speed if they reach the same position at the same time?
Graph Interpretation(Physics)	3. Here is the graph s/t of a person running along the beach. Describe the movement based on the graph.[Image omitted. Please see PDF.]	3. Here is the s/t graph of a person’s bike ride. Describe the movement based on the graph. [Image omitted. Please see PDF.]
Graph Creation(Physics)	4. A family travels to see the coast. They left San Sebastian at three o’clock in the afternoon and made a three-hour journey to Bilbao 120 km later. They have been in a bar for an hour and have taken the car back to Santander (another 120 km) at nine o’clock in the evening. Draw the position-time graph. Where is the family at 5 p.m.? and at 6:30 p.m.?	4. We have the following v/t graph (Speed in m/s and time in s). Create the corresponding graph s/t. Note that when t = 0 it is 2 m from the origin. [Image omitted. Please see PDF.]
Programming(Technology)		5. The teacher has asked you to program the mBOT so that at first you have to move fast for 3 s and at the same time carry the red lights on to indicate that you are going fast. Then for another three seconds it will move slowly and as it moves it will have green lights on. One of your teammates has proposed the following program. Would you give it up like that? If something needs to be changed, what would you change and why? [Image omitted. Please see PDF.]