Do High School Students Learn More or Shift Their Beliefs and Attitudes Toward Learning Physics with the Social Constructivism of Problem-Based Learning?

Abstract

Rooted in social constructivist learning theory, problem-based learning (PBL) is a tool that deepens students’ learning of complex subjects and improves students’ attitudes and beliefs towards learning. Physics is a subject that students themselves view as challenging. When taking physics, students develop negative beliefs about their own learning of the subject. There is a call for more innovation in the subject area of physics. This study addresses the following: (1) What is the effect of PBL on the achievement of 10th-grade students in mechanics when compared to traditional instruction? (2) How do students’ beliefs and attitudes towards physics change before and after the intervention, and how might these beliefs and attitudes relate to their competency outcomes? The sample of this study comprised 63 students in the 10th grade in a public (non-elite) school distributed across four classes, where the teacher used PBL with the experimental group and traditional teaching with the control group. The physics teacher who implemented PBL has 13 years of teaching experience. The two instruments used were the Force Concept Inventory and the Colorado Learning Attitudes About Science Survey. The result of this study revealed that, although students’ knowledge of physics increased when their teachers adopted the PBL approach, there were no significant changes in their attitudes and beliefs towards learning physics. The implications suggest that there is potential for PBL to be taken up by high school science teachers to improve their students’ physics knowledge, but may this not impact their attitudes and beliefs, which presents questions to investigate further.

1. Introduction

Problem-based learning (PBL) is ideal for engaging students in complex disciplinary subject areas and can introduce new teaching methods that even new teachers can take up. PBL contrasts with traditional lecture and textbook instruction because it employs the student-centered, constructivist approach to learning, paired with social constructivism [1,2,3,4]. It centers meaningful problems in highly social contexts to motivate students in science and math [5,6]. In science disciplines, learning as inquiry through open-ended cases or problems can be taken up by teachers who were taught traditional methods to shift their practice [7].

Given the potential benefits of the PBL approach, which can achieve learning results even in classrooms accustomed to traditional pedagogy [8,9,10] there is a need to expand the research on PBL. The field of science education needs more investigation about the benefits of the approach for learning challenging theoretical subjects and mandatory courses in high school, such as physics, where negative attitudes impact student learning [11]. Much of the research on PBL, especially in the area of science, is conducted with undergraduate and graduate college-aged students specializing in a specific science discipline (e.g., [12]). Further, PBL studies in high school science classes, such as chemistry and engineering, have shown some promise (e.g., [11]; however, these studies show mixed results. There is more to be known about whether and how PBL is an effective tool for learning gains, and about attitudes and beliefs toward difficult subject areas.

The following sections describe PBL and how it approaches learning through social constructivism and sensemaking. Then, we present research on PBL in science education and physics education. Next, we describe research about students’ attitudes and beliefs about physics, and the instrument the Colorado Learning Attitudes about Science Survey (CLASS) [13] is used to measure change in attitudes and beliefs. Finally, the force concept inventory [14], which is used to assess students’ achievement, is described.

2. Social Constructivism

PBL promotes learning that contrasts with traditional teaching because it is rooted in two mutually reinforcing theories of learning [15,16]. It builds on the constructivist theory of learning, where learners draw connections among their own previously held ideas to make sense of new experiences [17,18] through social constructivism [19]. The social constructivist theory views learning as situated participation in activity [20], which positions social others as scaffolds critical to learning [18], and thus places a primacy on the role of interaction as the mediator for knowledge and practice construction [1]. In particular, collaborative interaction for problem solving not only enables the perspective from the individual’s empirical construction of the problem, but is also a conduit for the individual’s unique “intellectual construction arising from their activity in the world” [21] (p. 278).

Community members, such as students and teachers, are immersed in practices focused on collaborative problem solving. Over time, they develop as individuals as they transform their classroom community practices [22]. PBL moves the meaning of expertise from being solely vested in the teacher to the students of the classroom community, as they mediate artifacts, language, roles, and norms for the purpose of participating and transforming practices for mutual endeavors [23]. In this framework, the development and application of the core scientific ideas essential to the problem is promoted through the social construction of meaning through mutual endeavors.

3. Problem-Based Learning

PBL originated in medical schools, but has since been adopted by various educational institutions, including middle schools and professional education. PBL has undergone extensive research and analysis by experts such as Barrows and colleagues in 1980 and Hmelo-Silver in 2004 [24,25,26]. Since the 1990s, PBL has gained immense popularity outside of the medical field and has become increasingly prevalent in both higher education and K-12 institutions.

Hmelo-Silver [25] defines PBL as an instructional method wherein students acquire knowledge through facilitated problem-solving. According to this scholar, PBL addresses complex problems that lack a singular correct answer. Additionally, Hmelo-Silver highlights the collaborative nature of PBL, emphasizing that students work together in groups to identify the required learning objectives, engage in self-directed learning, apply acquired knowledge to solve problems, and subsequently reflect on the learning outcomes and the efficiency of the strategies employed. Due to the emphasis on student-centered problem solving, PBL is promising for science education because it affords focused, experiential learning organized around investigating, explaining, and resolving meaningful problems [24,25].

PBL aims to help students develop comprehensive cognitive models of the problems presented to them, and so it is essential for teachers to have the competency to support this approach. Teacher guidance plays a significant role in developing critical reasoning skills in students. The success of PBL is heavily reliant on the teacher’s ability to provide this support [24,26]. Some of the responsibilities of teachers in PBL include facilitating the students’ development of thinking or reasoning skills that promote problem-solving, metacognition, and critical thinking, as well as helping them to become independent and self-directed learners. Likewise, teachers who employ PBL approaches often have the implicit goal of developing criticality in their students [27]. Additionally, the teacher guides the enculturation into the community of practice as the core aim of PBL [9].

Problems are a key element of PBL, and through teacher guidance, they are presented through a series of phases [28]. Problems in PBL refer to the educational materials provided to students to stimulate their learning processes. These problems are typically presented textually, occasionally supplemented with images and computer simulations [26]. Occasionally, they depict real-life situations or phenomena, compelling students to elucidate or address the problems presented [29].

PBL requires students to engage with one another toward the construction of new meanings, i.e., meanings related to practices for problem-solving and disciplinary meanings. Students encounter a real-life problem that frames their learning trajectory for a learning cycle whereby learners engage with an authentic problem that requires investigative and problem-solving skills. The problem presented should capture students’ attention and stimulate their curiosity [28]. Students collaborate to analyze the problem collectively, seeking out resources and critically assessing information to enhance their comprehension and generate potential solutions [26]. In PBL, students actively participate in collaborative dialogue, suggesting potential solutions and creating strategies to tackle the problem at hand. This prioritizes critical thinking, creativity, and the utilization of pertinent knowledge and skills to devise effective solutions [30]. Students are urged to examine various viewpoints and contemplate diverse problem-solving approaches. Last, in PBL, students reflect on their learning processes, evaluate their progress, and pinpoint areas necessitating enhancement. Reflection encourages metacognition and self-directed learning, empowering students to recognize their cognitive processes, augment their comprehension, and apply their learning to future scenarios [24,31].

While PBL is widely recognized for its potential to enhance student engagement and critical thinking skills, implementing this approach presents significant challenges for both teachers and students. Teachers may encounter difficulties managing diverse students’ needs and maintaining a balance between providing support and promoting independent problem-solving [29,32]. For students, PBL can initially feel overwhelming due to its open-ended, student-centered nature, which contrasts with the more structured, directive methods to which they are accustomed. Catz and colleagues (2018) discuss how these challenges can impact students’ motivation, as seen in their study of an electronics laboratory course where students struggled with the ambiguity and self-directed requirements of PBL [32]. Such challenges underscore the need for ongoing support and gradual adaptation when introducing PBL, especially in subjects like physics, where students’ existing attitudes and beliefs may already be influenced by perceived subject difficulty.

4. PBL and Students’ Learning

PBL is broadly accepted as an approach that benefits learning by presenting learners with real-world problems and motivating them to work together to develop solutions to them. As students grapple with these shared problems, they acquire subject-specific knowledge and develop critical thinking and problem-solving skills. Similarly, when students see the value in what they are learning and their educational activities are personally meaningful, they are more motivated to engage in the learning process [33].

Additionally, within the context of PBL, students engage in social interaction to participate in collaboration processes using reflection and interdependence to promote group learning [33]. Studies have examined the performance of students engaged in PBL. Notable studies by Hmelo and colleagues [34] and Patel and colleagues [35,36] contribute valuable insights to this line of inquiry. Patel et al. undertook a comparative analysis involving both traditional and PBL students, tasking them with providing diagnostic explanations for a clinical problem. Interestingly, the findings revealed distinctions in the quality of explanation between the two groups. While the PBL students exhibited a higher degree of elaboration in their explanations, surpassing their counterparts in traditional curricula, their responses also displayed a marginally higher error rate. This dichotomy in the performance of PBL students sheds light on the nuanced nature of their problem-solving abilities, emphasizing both strengths and areas for improvement [37]. In conclusion, the collective findings from studies by Hmelo and Patel [34,35,36,37] suggest a positive impact of PBL on students’ problem-solving skills.

In 2017, Hussain and Anwar described the high school students’ responses to PBL in chemistry class: “[they] eagerly attended the problem-solving classes and wanted to participate in a discussion about the topic in a productive manner” [38] (p. 37). However, when Derry and colleagues [39] examined students’ reactions to a PBL course in statistical reasoning, they found that their reactions were mixed. While some students enjoyed the class, others did not like the idea of working collaboratively. This variability emphasizes the importance of considering individual preferences with respect to attitudes in the implementation of PBL.

5. PBL as a Needed Approach for Complex Subject Areas

Improvement in learning has been found to be connected to students’ motivation to learn physics [40], problem-solving ability [41], and critical thinking ability [42], all of which are components of PBL. Thus, PBL could be an ideal approach for subjects that are viewed negatively by students, and attitudes impact learning.

There is a paucity of research that shows that high school students are losing interest and motivation in learning physics, which is reflected by their declining exam performance [43,44]. Therefore, physics teachers need to develop better teaching methods. Despite the common perception that physics is a challenging subject, preliminary research demonstrates that PBL can increase student achievement in physics [45,46,47].

A study carried out by Argaw and colleagues in 2016, investigated the impact of PBL strategy on students’ problem-solving skills and motivation in physics education [47]. Using a quasi-experimental method, data were collected from 81 grade 12 students through problem-solving inventory tests and motivation scales. The results indicated a statistically significant improvement in problem-solving skills in the experimental group compared to the control group, with an above-average effect size. However, there was no significant increase in students’ motivation to learn physics.

Polanco conducted an assessment of an innovative PBL curriculum designed for second-year engineering students [12]. This integrated curriculum merged the subjects of physics, mathematics, and computer science into a unified course structure. The findings revealed that students enrolled in the PBL curriculum exhibited significantly greater improvements in scores on the Mechanics Baseline Test (Hestenes and Wells compared to their counterparts in the control group [14]. However, both groups demonstrated similar levels of improvement on the Force Concept Inventory (FCI). This suggests that PBL may offer advantages in specific areas of understanding and calls for investigations about similar results for high school students.

Yeo and colleagues (2012) conducted research to examine the challenges teachers encounter when incorporating PBL in their classrooms [48]. They detailed the experiences of a high school physics teachers as they introduced PBL into their teaching practice. The study highlighted the challenges teachers encountered, mainly stemming from discrepancies between their teaching goals and the expectations set by formal education systems, highlighting the divide between academic ideals and practical realities. This observation aligns with the findings of Pease and Kuhn (2011), who emphasized the significance of problem engagement over the social aspects typically associated with PBL [49].

To improve students’ performance in physics, teachers are advised to adopt innovative approaches, such as the PBL method, in their teaching to engage students and motivate personal meaning and perseverance. Education in physics is intricately linked to beliefs and attitudes, constituting a domain of research that will be our next focal point.

6. Students’ Beliefs and Attitudes Toward Learning Physics

Encouraging a positive mindset towards science can contribute significantly to academic achievement in science-related subjects, as per research conducted by Hammer and Morse and Morse [50,51]. Compared with other science disciplines [52,53], physics education demonstrates this as particularly critical, where negative student attitudes prevail, as highlighted by Godwin and Okoronka [54]. One commonly used measure of attitudes towards science education is the CLASS [13].

CLASS measures students’ attitudes and beliefs about physics compared to expert career physicists. It uses a 5-point Likert scale to compare expert-like and novice-like responses. CLASS is an instrument widely used in the United States, and studies have shown that interpretations are transferrable to populations in countries outside of the USA [55,56,57,58]. CLASS is designed to be accessible to students at all levels of physics, even students who have not taken a physics class [13].

Essentially, CLASS is grounded on students’ epistemological understandings of physics as a science, specifically about the practices related to physics, which may be more or less similar to those of experts in the field. The CLASS survey is useful to gauge the effectiveness of courses with factors that transcend cognition. As students acquire expertise in physics, attitudes and beliefs change accordingly [55,58]. The 42 statements in CLASS are divided into three categories: (1) personal application and relation to the real world, (2) problem-solving self-efficacy, and (3) effort and sensemaking. For example, a question that probes students’ effort and sensemaking asks, “In doing a physics problem, if my calculation gives a result very different from what I’d expect, I’d trust the calculation rather than going back through the problem”. An expert physicist would respond that they would return to the problem and test their calculation.

Students’ attitudes towards school and academic success significantly impact their academic performance [59]. Rosenberg and Hovland defined attitude as the intermediary for all kinds of reactions [60]. Attitude has three main components: emotion, cognition, and behavior. Attitude encompasses a person’s instincts, feelings, prejudices, biases, preconceived notions, fears, threats, and convictions about a particular subject. The impact of attitude on students’ academic achievement, internal motivation, and participation in school is significant [61].

While PBL can effectively enhance students’ problem-solving skills in physics, it found no significant effect on students’ motivations to learn the subject [45]. This finding raises questions about the relationship between PBL and affective outcomes, such as motivation and attitudes, in physics education. Research highlights the importance of students’ attitudes, noting that positive attitudes can significantly impact achievement, motivation, and participation. Building on these findings, our study seeks to examine whether the PBL approach not only influences cognitive outcomes, like problem-solving skills, but also affects students’ attitudes and beliefs toward physics [61].

There is a strong link between a student’s attitude and their academic performance in physics [54]. Furthermore, a student’s attitude towards physics is crucial to how they approach problem-solving methods, not only in physics but also in other science classes [62].

Physics students’ dispositions have a significant impact on their accomplishments in the subject. A majority of learners find physics challenging, primarily due to the learning methodologies involved in comprehending the subject, which requires students to work with different types of representations like formulas, calculations, graphics, and even an abstract level of conceptual understanding [63,64]. Similarly, the lack of understanding of the problem and poor mathematical skills also constitute the major obstacles in the circle of difficulties students experience in solving physics problems [65]. Students who exhibit negative attitudes toward physics achieve lower-than-desired examination scores [66,67]. Guido’s (2011) study echoed these findings, revealing that students with negative attitudes toward science similarly expressed discontent with physics courses and instructors [59]. Likewise, teachers attributed students’ achievements in physics to their negative attitudes and lack of interest in the subject [11].

Overall, physics is an ideal candidate for the innovative approach of PBL for impacting students’ attitudes. Studies about students’ attitudes toward learning physics are contrary to general attitudes toward learning and present the need to rethink how physics is taught. According to many researchers, when students are given the opportunity to showcase their problem-solving abilities, either through a teacher-led or a student-led approach, it can have a positive impact on their attitudes towards physics [68]. Studies comparing traditional teaching methods with more modern ones have shown that incorporating new teaching techniques and technology is essential to improve students’ success rates and attitudes towards physics education [69]. The examination of students’ attitudes revealed that a majority held positive views towards learning [70]. Lindstrom and Sharma (2011) emphasize the significant impact of students’ attitudes and beliefs on their academic success in physics. According to these researchers, students who maintain a positive attitude toward the subject and hold optimistic beliefs about their abilities in physics are more likely to succeed. Having a positive attitude and mindset stimulates students to put more effort and leads to high achievement in that subject, while a negative attitude towards a certain subject makes learning more difficult [11].

Many studies have demonstrated the effectiveness of PBL in enhancing critical thinking skills in high school students across various subjects, including science and mathematics (e.g., [12,24,26,30,68,71,72,73,74]. These studies indicate that PBL fosters higher-order thinking by engaging students in complex, real-world problem-solving tasks. However, fewer studies have explored whether PBL also influences students’ attitudes and beliefs about the subject matter, particularly in challenging fields like physics. Our study seeks to address this gap by examining not only the effects of PBL on achievement, but also its impact on students’ attitudes and beliefs toward physics. Building on the context outlined above, this study seeks to address the following research questions:

What is the effect of PBL on the achievement of 10th-grade students in mechanics when compared to traditional instruction?

How do students’ attitudes and beliefs toward physics change before and after the PBL intervention, and how might these attitudes relate to their competency outcomes in physics?

7. Methods

In this study, an experimental approach was employed to investigate the effects of PBL on high school students’ achievement, attitudes, and beliefs toward physics.

Participants and Context

In our study, we employed a convenience sampling approach due to the specific characteristics and accessibility of the population we intended to study. The sample of this study comprised 63 students in a public school in the 10th grade, randomly distributed across four classes: 10A, 10B, 10C, and 10D, with 39 females and 24 males. The experimental group included 32 students from the 10A and 10B, with 23 females and 9 males, while the control group comprised 31 students from the 10C and 10D, with 16 females and 15 males. All students were either 15 or 16 years old and of Kazakh ethnicity.

No formal training in PBL or teamwork was provided to the students in the experimental group prior to the intervention. The PBL approach was introduced directly within the context of their regular physics classes, and students were gradually guided through the process by the teacher. The teacher incorporated scaffolded activities that increased collaboration and problem-solving skills, allowing students to adapt to the PBL structure as they engaged with real-world physics problems.

In this study, the physics teacher who implemented PBL was a female teacher of 35 years old with 13 years of teaching experience. She held a degree in physics education and was knowledgeable in both traditional and constructivist teaching methods, although PBL was a novel instructional approach for her. Her teaching style aligned with the principles of active learning, focusing on student-centered instruction and fostering critical thinking. The teacher’s educational background and emerging expertise facilitated her adoption of PBL strategies. She built on prior constructivist teaching strategies to effectively support students in collaborative problem-solving and critical reasoning.

The school, located in a rural area of Almaty-Kazakhstan, accommodates 1337 students from grades 4 to 11 across 57 classes. Physics instruction begins in the 7th grade. After 7th grade, students diverge into two tracks: physics–mathematics and chemistry–biology. The academic year starts in September, with students spending 38 weeks in school annually. In grades 7–9, students have three 45 min physics lessons per week. From the 10th grade onward, students study physics four times a week. By grades 10–11, students in the study school typically show interest in courses related to their future professions, such as mathematics, chemistry, biology, English, and computer science.

The high school employs 151 teachers, with seven dedicated to physics. All lessons are conducted in the Kazakh language, and only six classrooms are equipped with interactive boards. The teaching approach primarily follows a traditional didactic lecture format. Students in the school generally come from families of average or low social status. Due to the proximity of the village to the city, students with higher academic performance or economically affluent families often choose to pursue education in the city.

8. Implementation of PBL and Traditional Teaching

The academic calendar in Kazakhstan commences on September 1st and concludes at the end of May, spanning a total of 34 weeks divided into 4 terms. The introduction of PBL occurred over a seven-week period within the first term, with post-tests administered at the end of this period. Students had prior knowledge of mechanics, as they had studied the mechanics section of physics in grades 7 and 9.

During the course of the study in the experimental group, lessons on 11 topics of physics for 10th-grade students in schools were conducted. These topics were taught using PBL techniques in experimental groups, while the control groups were taught in a traditional format. Through PBL, the teacher planned lessons, organized students, posed problems, monitored student actions, assisted them (guided inquiry), checked student responses, and facilitated discussions. Some real-world problem-solving tasks were also given to students as homework assignments. Students worked first individually and then in groups, engaging in problem-solving tasks. For example, the following problem (Figure 1) was given to help students understand Galilean relativity:

During the theoretical portion of the class, students in the experimental groups undertook the above task collaboratively, working through solutions to the problem according to the steps of PBL. As the lesson concluded, the teacher designated it as homework for students to illustrate this phenomenon via a video recording. An example of a student-produced video can be found in the following YouTube link: https://www.youtube.com/shorts/3hJdV_-qZEA, accessed on 18 September 2023.

During the regular teaching format (control group), the teacher planned the lesson, explained new topics, clarified the algorithm for solving problems, and evaluated students who completed the assigned tasks according to the guidelines provided by the teacher. Homework was assigned at the beginning of the lesson. After that, new topics were introduced. Then, the teacher demonstrated the process of solving problems related to the topic. During the wrap-up phase of the lesson, students completed the assignments provided by the teacher. At the end of the lesson, the teacher evaluated the students. Students played the role of attentive listeners to the teacher and performers of the given tasks. Detailed explanations and formulas were provided for homework assignments.

Two examples of tasks for traditional teaching are as follows:

• Car accelerates from rest to 100 km/h in 6 s. Calculate acceleration of car.
• A centrifuge accelerates from rest to 1000 rpm in 3 min. What is the tangential acceleration, centripetal acceleration, and total acceleration of a point 10 cm from the axis of rotation?

Students studied 11 topics on mechanics during this research.

Table 1. Topics studied.

No	Topics	Objectives	Date
			10A	10B	10C	10D
	Equations of uniformly accelerated motion, graphs of accelerated motion	- Derive formula of displacement in accelerated motion from velocity–time graph; - Apply equations of motion for solving analytical and graphical problems.	04.09.2023	04.09.2023	04.09.2023	04.09.2023
	Invariant and relative physical quantities, Galilean relativity	- Tell the difference between invariant and relative physical quantities; - Apply equations of Galilean relativity for problem-solving.	08.09.2023	11.09.2023	08.09.2023	11.09.2023
	Curvilinear motion	Determine radius of curvature of trajectory, tangential acceleration, centripetal acceleration, and total acceleration during curvilinear motion.	11.09.2023	13.09.2023	01.09.2023	12.09.2023
	Projectile motion	Determine kinematic parameters of an object performing projectile motion.	13.09.2023	18.09.2023	13.09.2023	18.09.2023
	Force. Newton’s laws of motion. Addition of forces	Make possible problem-solving algorithms for motion of an object under effect of several forces.	18.09.2023	25.09.2023	18.09.2023	25.09.2023
	Law of universal gravitation	Apply Newton’s law of universal gravitation in problem solving.	22.09.2023	27.09.2023	22.09.2023	26.09.2023
	Moment of inertia of rigid bod	Use parallel axis theorem (Huygens–Steiner theorem) for calculation of moments of inertia of bodies.	26.09.2023	02.10.2023	26.09.2023	02.10.2023
	Conservation of angular momentum	Apply equation of rotational motion for problem solving.	03.10.2023	04.10.2023	03.10.2023	03.10.2023
	Center of mass	Determine the center of mass of the rigid body, determine the center of mass of a system of bodies.	09.10.2023	09.10.2023	09.10.2023	09.10.2023
	Types of equilibrium	Determine cause–effect relationships for different types of equilibrium.	11.10.2023	11.10.2023	11.10.2023	10.10.2023
	Conservation of momentum and energy	Apply law of conservation of momentum and energy for solving analytical and experimental problems.	16.10.2023	16.10.2023	16.10.2023	16.10.2023

indicates both the topics and objectives over eight weeks. The PBL activities carried out for each topic are presented in the Appendix A.

The use of traditional teaching methods in the control group was a deliberate and ethically considered choice. While PBL is recognized for its effectiveness, the study aimed to compare its impact against the traditional methods still widely used in the educational system of Kazakhstan. The ethical justification for using traditional methods lies in the need to establish a baseline for understanding the relative effectiveness of PBL. Moreover, students in the control group continued to receive the standard curriculum delivered by the same teacher, ensuring that their educational experience was not compromised.

In the context of this study, traditional teaching methods involve a didactic, teacher-centered approach that emphasizes direct instruction and individual student work. The teacher delivers content primarily through lectures, focusing on explaining concepts, theories, and facts while students listen and take notes. The teacher provides specific problems or tasks related to the lesson content, typically algorithmic and requiring students to apply formulas and procedures demonstrated by the teacher.

In the control group classrooms, the degree of social engagement was minimal compared to the experimental group implementing PBL. The problems given in the control group were typically straightforward and focused on applying specific physics formulas and principles that had been directly taught by the teacher. Students were generally expected to work on these problems alone, reinforcing the traditional focus on individual accountability and mastery of the content. There was little to no discussion of problems among students during the lesson.

9. Adapting the Study to the Country Context

Kazakhstan’s educational system has traditionally emphasized teacher-centered, didactic instruction. Introducing PBL in this context represents a significant shift towards student-centered learning. Our study aimed to evaluate the feasibility and impact of such a shift in a system where traditional methods are deeply entrenched [75]. The study was conducted in Kazakh, the native language of the students, ensuring that the instructional materials and assessments were culturally and linguistically appropriate. The study was conducted in a rural school with limited resources, reflecting the socioeconomic conditions of many schools in Kazakhstan. This setting contrasts with the often resource-rich environments where PBL is commonly studied in Western countries. Our study reflects the implementation of PBL in less affluent settings, demonstrating both challenges and opportunities unique to such contexts [76,77]. The teacher involved in the study received specific training on PBL within the Kazakhstani educational context. This included addressing local pedagogical norms and adapting PBL strategies to fit within the constraints and opportunities of the local school system [78]. The study also considered the local students’ prior experiences with education, which were predominantly based on traditional teaching methods. The transition to a PBL approach required careful consideration of students’ readiness and receptiveness to this new method. Understanding students’ perceptions and engagement in a Kazakhstani context indicates the cultural adaptability of PBL [79].

10. Instruments

Two instruments were used in this study. The first instrument was used to measure students’ beliefs about physics and about learning physics. The second instrument was used to investigate the effect of PBL instruction on 10th-grade students’ physics achievements.

To measure changes in student physics achievements, the FCI test was used [14]. FCI consists of 30 items related to force and motion concepts. The instrument can be used pre- and post-test in a physics course to determine student improvement in physics understanding [14]. The FCI was previously translated into Kazakh by the authors [80]. The validity and reliability study of the translated version of FCI was conducted by Authors et al.

To measure changes in student beliefs toward physics, the CLASS questionnaire containing 36 statements was used. The CLASS originally consisted of 42 items; however, developers [13] suggested the removal of six items that needed further development. Responses to the items on the survey ranged from a 1 (Strongly Disagree) to a 5 (Strongly Agree), with a midpoint defined as “No Opinion”. The first author translated the survey. To ensure the appropriate translation, an English language teacher and a physics teacher who knows both Kazakh and English at an advanced level checked the translation. Moreover, before the administration of the translated version to our sample, two students read aloud and filled in the survey while the first author was observing [81]. During this process, minor corrections were made on the translated version.

11. Data Collection

The first author, serving as the physics teacher implementing PBL, gathered student questionnaires during the lesson. The FCI test spanned 80 min, with some sessions extending over two periods. Completing the CLASS survey typically required 10–15 min. All data were gathered in the classroom using paper formats to ensure meticulous completion by students. Additionally, considering that some students lacked phone or internet access, the paper format proved convenient for collection.

12. Data Analyses

Descriptive statistics were conducted to unveil the groups’ means, medians, standard deviations, and normality. Initial differences between the control and experimental groups were examined using independent samples t-Tests. The effect of PBL instruction was analyzed using MANOVA, a robust test against Type 1 error in hypothesis testing. Additionally, we employed both Pillai’s Trace and Wilks’ Lambda values for the main effect in MANOVA, which are resilient against violations of the MANOVA assumption. In cases where students’ initial achievements and beliefs were unequal, MANCOVA was employed. Finally, for practical significance, we calculated the gain scores developed by Hake (1998).

13. Ethical Approval and Ethical Issues

This study adhered to strict ethical guidelines to ensure the protection and respect of all participants involved. Ethical approval was obtained from the Al-Farabi Kazakh National University’s ethics committee, ensuring that the study met all institutional and national ethical standards for research involving human subjects. The dual role of one of the researchers, who also served as a teacher, was carefully managed to avoid any potential conflicts of interest or undue influence on the students. This researcher maintained a professional boundary and ensured that her dual role did not affect the students’ participation or responses. Furthermore, informed consent was obtained from all participants, ensuring that they were fully aware of the study’s purpose, procedures, and their right to withdraw at any time without any consequences.

14. Results

The results section begins with the introduction of the descriptive statistics. There were two experimental and two control groups in this study. In the following analyses, they will be called Group One (G1-10A and 10C) and Group Two (G2-10B and 10C). In the case where the groups are combined, it is called “Overall”.

15. Descriptive Statistics

All group’s pre- and post-FCI and CLASS score means, medians, standard deviations, and normality, measured with Shapiro’s Wilk, are presented in

Table 2. Group descriptives.

	Group	N	Mean	Median	SD	Shapiro Wilk (p)
G1 Pre-FCI	Control (10C)	12	9.42	9.5	3.06	0.739
	Experimental (10A)	12	9.00	8.5	3.84	0.985
G1 Post-FCI	Control (10C)	12	13.42	12.5	3.4	0.086
	Experimental (10A)	12	16.83	16.5	3.86	0.834
G1 Pre- CLASS	Control (10C)	12	108.83	109.5	7.94	0.159
	Experimental (10A)	12	125.92	124	8.87	0.753
G1 Post-CLASS	Control (10C)	12	116.25	115	8.88	0.757
	Experimental (10A)	12	120.58	118.5	8.97	0.503
G2 Pre-FCI	Control (10D)	19	7.63	7	2.06	0.204
	Experimental (10B)	20	6.63	6	1.89	0.115
G2 Post-FCI	Control (10D)	19	10.11	11	2.56	0.138
	Experimental (10B)	20	13.89	13	4.37	0.281
G2 Pre-CLASS	Control (10D)	19	107.11	108	7.22	0.888
	Experimental (10B)	20	103.53	105	8.42	0.475
G2 Post-CLASS	Control (10D)	19	113.16	113	7.13	0.770
	Experimental (10B)	20	111.68	114	8.23	0.703
Pre-FCI Overall	Control	31	8.32	8	2.6	0.045
	Experimental	32	7.50	7	2.95	0.068
Post-FCI Overall	Control	31	11.39	11	3.29	0.106
	Experimental	32	15.09	15	4.31	0.689
Pre-CLASS Overall	Control	31	107.77	109	7.42	0.224
	Experimental	32	112.16	111	13.71	0.800
Post-CLASS Overall	Control	31	114.35	114	7.86	0.782
	Experimental	32	115.34	115	9.39	0.893

. The maximum score that can be attained from FCI is 30 (30 questions each one point), and that from CLASS is 180 (36 items rated on a 5-point scale). Higher scores on FCI indicate higher academic performance, and higher scores on CLASS indicate expert-like beliefs. Items indicating novice beliefs were reverse-coded.

As seen in Table 2, the data are normally distributed for all variables. The mean score of FCI varies between 6.63 (G2 pre-FCI-experimental) and 16.83 (G1 post-FCI-experimental). Similarly, the mean score of CLASS varies between 103.53 (G2 pre-CLASS-experimental) and 125.92 (G1 pre-CLASS-experimental).

To determine the effect of PBL on both the students’ academic performance and their expert-like beliefs, students’ post-tests scores should be compared. However, initial equality of the groups is required. For this, the control and experimental groups’ initial differences (see Table 2) were tested with the independent samples t-test (

Table 3. Independent samples t-test.

	Statistic	df	p
G1 Pre-FCI	0.29	22	0.771
G1 Pre-CLASS	−4.97	22	<0.001
G2 Pre-FCI	1.56	36	0.128
G2 Pre-CLASS	1.28	37	0.209
Pre-FCI Overall	1.17	61	0.246
Pre-CLASS Overall	−1.57	61	0.122

).

The only significant differences (bold in Table 3) between the initial inequalities were observed for G1 pre-CLASS (p < 0.001, effect size = −2.03). The beliefs of the control and experimental students in this group were not equal at the initial stage. Thus, G1 pre-CLASS was used as a covariate in the related MANCOVA analysis that was carried out for G1.

Inferential statistics. One one-way MANCOVA (for Group One) and two one-way MANOVAs (for Group Two and the Overall) were performed to unveil disparities among groups, where the independent variable was the teaching method, encompassing two levels (traditional and PBL), and the dependent variables were the scores obtained by students in FCI and CLASS (See

Table 4. Multivariate tests.

Group	Test	Value	F	df1	df2	p
Group One (MANCOVA)	Pillai’s Trace	0.27	3.76	2	20	0.041
	Wilks’ Lambda	0.73	3.76	2	20	0.041
Group Two (MANOVA)	Pillai’s Trace	0.25	5.88	2	36	0.006
	Wilks’ Lambda	0.75	5.88	2	36	0.006
Overall (MANOVA)	Pillai’s Trace	0.20	7.46	2	60	0.001
	Wilks’ Lambda	0.80	7.46	2	60	0.001

). To ensure the validity of the MANOVA analyses, several key assumptions were tested prior to conducting the statistical procedures. The assumption of homogeneity of variances was tested using Levene’s Test for Equality of Variances, which indicated that the variances across groups were approximately equal, meeting the criterion for MANOVA. Additionally, Box’s M test was performed to assess the homogeneity of covariances. The test yielded a χ² value of 0.56 with three degrees of freedom (df) and a p-value of 0.905. Since the p-value was greater than 0.05, the assumption of homogeneity of covariance matrices was met, indicating no significant differences in covariance matrices across groups. Multivariate normality was assessed through examining the Shapiro–Wilk Multivariate Normality Test. The test produced a W value of 0.96 with a p-value of 0.421. With a p-value above 0.05, the assumption of multivariate normality was satisfied. To check for multicollinearity, Pearson correlation coefficients were calculated among the dependent variables, revealing that correlations did not exceed the recommended threshold.

As indicated in Table 4, there exists a statistically significant impact of the instructional method for Group One (F(2, 20) = 3.76, p = 0.041; Wilk’s Λ = 0.73), Group Two F(2, 36) = 5.88, p = 0.006; Wilk’s Λ = 0.75), and Overall (F(2, 60) = 7.46, p = 0.001; Wilk’s Λ = 0.80). To discern the variations in FCI and CLASS scores across the control and experimental groups, attention should be directed to the univariate tests (see

Table 5. Univariate tests.

	Dependent Variable	Sum of Squares	Df	Mean Square	F	p
Group One	G1 Post-FCI	70.04	1	70.04	5.07	0.035
	G1 Post-CLASS	112.67	1	112.67	1.35	0.258
Residuals	G1 Post-FCI	290.1	21	13.81
	G1 Post-CLASS	1753.02	21	83.48
Group Two	G2 Post-FCI	151.62	1	151.62	11.92	0.001
	G2 Post-CLASS	8.94	1	8.94	0.15	0.703
Residuals	G2 Post-FCI	470.74	37	12.72
	G2 Post-CLASS	2235.73	37	60.43
Overall	Post-FCI Overall	216.34	1	216.34	14.66	<0.001
	Post-CLASS Overall	15.4	1	15.4	0.2	0.652
Residuals	Post-FCI Overall	900.07	61	14.76
	Post-CLASS Overall	4586.32	61	75.19

).

Table 5 reveals that the instructional method significantly influences students’ academic performance, as indicated by the FCI scores for G2 post-FCI (F (1, 37) = 11.97; p= 0.001), and post-FCI Overall (F (1, 61) = 14.66; p < 0.001). Notably, we implemented an alpha correction to address the multiple ANOVAs conducted, employing a Bonferroni correction. Therefore, in this instance, we considered statistical significance at p < 0.025, and relatedly, p = 0.035 for G1 post-FCI was not considered significant.

Two distinct instructional methods were employed: traditional teaching and teaching with PBL. To determine the effectiveness of each method in enhancing students’ academic performance, descriptive statistics in Table 2 were examined for each group and the Overall. After Bonferroni correction for the alpha level, the effect of PBL was significant for Group Two and for the Overall; however, it was not significant for Group One.

For the G2 post-FCI, the control group demonstrated an average score of 10.11 on the 30 FCI questions, whereas the experimental group, instructed with PBL, achieved a higher average score of 13.89. Similarly, for the post-FCI Overall, the control group demonstrated an average score of 11.39 on the 30 FCI questions, whereas the experimental group, instructed with PBL, achieved a higher average score of 15.09. Consequently, we infer that PBL significantly contributed to the improvement of students’ performance, although it did not yield a statistically significant effect on students’ attitudes and beliefs.

Hake’s gain scores: For practical significance, we calculated the gain scores developed by Hake in 1998 [82]. A Hake’s gain close to 0 indicates little to no improvement from the pre-test to the post-test relative to the maximum possible gain, indicating that participants, on average, did not gain significant knowledge or skills. On the other hand, a Hake’s gain close to 100 indicates that participants have achieved the maximum possible gain. This suggests a substantial improvement in understanding or performance.

Hake’s Gain = (Posttest − Pretest)/(Maximum Possible − Pretest) × 100

The mean scores of all groups, as well as Overall, are presented in Figure 2 and Figure 3. To see the changes from the pre-test to the post-test, the pre-test box plots are colored gray, while the post-test box plots are colored white.

As seen in Figure 2, all post-test scores were higher than the pre-test scores.

As seen in Figure 3, even though the post-test scores were higher, the difference was not significant. The practical effect of PBL is presented in

Table 6. Hake’s gain scores.

Group	Gain Score	Group	Gain Score
CG1-FCI	19	CG1-CLASS	3
CG2-FCI	11	CG2-CLASS	3
EG1-FCI	37	EG1-CLASS	0
EG2-FCI	32	EG2-CLASS	2
CG-Overall	14	CG-Overall	3
EG-Overall	34	EG- Overall	1

with Hake’s gain scores.

As observed in Table 6, the gain scores for CLASS exhibited minimal changes, while those for FCI varied between 11 and 37. Specifically, in Group One, the control and experimental gain scores are 19 and 37, respectively. However, the 18-point gain in this group did not reach statistical significance according to MANOVA analyses. In Group Two, the control and experimental gain scores were 11 and 33, with the 22-point gain in this group being deemed statistically significant by MANOVA analyses. Likewise, the overall gains for the control and experimental groups in FCI were 14 and 34, respectively. This 20-point increase was found to be statistically significant through MANOVA analyses. However, for CLASS, the overall gain was not considered significant.

16. Discussion

The results of the study add to the small but growing evidence that PBL is a viable and effective method to employ in mandatory high school physics classes. We compared the control group taught with the traditional teaching method—a didactic teacher-centered approach—with an experimental group, which was PBL–inquiry through problem-posing with open-ended solutions over the course of one academic year.

MANCOVA (one-way) was used for comparison and to accommodate groups’ unequal results on the pre-test. PBL, as an instructional method, demonstrated a statistically significant impact on the achievement of both of the experiment groups (p = 0.041; p = 0.006) and the Overall (p = 0.001). However, using MANCOVA, no significant difference was found in students’ attitudes and beliefs related to physics. In brief, the results indicate that the introduction of PBL as a teaching method increased student learning of physics, but did not affect the students’ attitudes or beliefs.

The teacher and students were able to make a transition to the social constructivist approach, where students act as scaffolds for each other for both problem-solving practices and disciplinary learning. The large-scale study by Bando and colleagues in 2019 highlights the potential for PBL [83]. We examine the findings of disciplinary learning and beliefs, which may be related to how students need time to understand themselves as agents in their own learning community when they are accustomed to teachers dispensing information [17,19].

The first finding that PBL as an instructional method helped students learn physics corresponds with previous studies. PBL is associated with higher achievement in complex content than traditional teaching approaches, such as medical education, according to a 1993 meta-analysis [84]. Hmelo-Silver and colleagues also found similar results in students’ problem-solving skills for mathematics due to exposure to PBL [34].

The second finding that PBL did not significantly affect the students’ attitudes and beliefs also corresponds to some literature. Some research shows that, compared to achievement, attitudes and beliefs do not show significant changes [85]. There are many other studies, however, that contradict this finding, showing a correlation between PBL and positive attitudes [38]. Barrows and Tamblyn showed that students’ beliefs and attitudes do positively shift when learning through PBL (e.g., [24]). Other studies demonstrate that the higher achievement level realized through PBL impacts student attitudes (e.g., [68]). Further, attitudes toward learning are impacted by academic success in a difficult subject area of English learning [70].

Given the research related to efforts to innovate an educational system, we find the results both promising and curious; they inspire new questions. First of all, we provide initial evidence that teachers used to traditional methods can take up PBL and implement social constructivist methods. The fact that students gained physics knowledge, even in the first trial, demonstrates that there is potential for students to benefit from the approach. However, the results are also curious. The lack of significant change in the students’ attitudes and beliefs is surprising because it contradicts the bulk of the research that reveals associations between PBL and students’ responses.

That there was no significance in the results initially calls into question how the students responded to the new method and, given the association of positive beliefs following learning gains, whether or not they experienced themselves gaining knowledge. Further, we wonder if there are unknown factors in the unique setting that could be explored. These questions notwithstanding, our study points out the potential benefits PBL can offer to students.

We wonder if more needs to be known about how PBL would be designed to better enable students’ ability to reflect on their experiences, which are developed through the process of solving problems [24]. This aspect of PBL should be studied further in a similar context. Professional learning is necessary to support teachers in developing collaborative skills through PBL [48].

This study further supports the call for professional learning with teachers and in classroom community cultures unused to student-centered approaches to science learning. Furthermore, even if PBL is adopted to support students’ learning, it is likely that beliefs and attitudes may also be impacted [68]; this would benefit developing educational systems in growing economies, such as in central Asia [78].

Studies overwhelmingly show that PBL positively impacts problem-solving skills [38]. Adopting PBL at the high school level could stimulate the creative efforts needed to solve challenges related to shifting the economy to greener technologies, expanding opportunities for poor regions (e.g., [76]), and solving the conservation of scarce habitats [86,87,88]. PBL and the educational and economic benefits created could be one avenue to further educational innovation for success.

After seven weeks of PBL intervention, our study found no significant change in students’ attitudes and beliefs about physics. It is important to note that changing attitudes and beliefs often requires longer interventions than those needed to affect cognitive outcomes. Research suggests that significant changes in students’ attitudes and beliefs about learning are typically observed over extended periods of sustained instructional change [89,90]. Black and Wiliam emphasize the need for sustained and continuous assessment practices to impact students’ attitudes and beliefs about learning [89]. Richardson’s work highlights the complexity of changing teachers’ attitudes and beliefs, emphasizing that such changes require extended periods of reflection, practice, and feedback [90].

In physics education, female students often experience a more negative affective relationship with the subject than male students, which could have influenced their receptivity to PBL and affected the measured shifts in attitudes and beliefs [91,92]. This gender imbalance should be considered when interpreting the results, as it may have contributed to the observed patterns. It is worth considering that PBL itself is not primarily designed to shift attitudes and beliefs, but rather to promote deeper understanding and application of content knowledge. While some studies suggest that PBL can positively impact affective outcomes, this effect may not be universal or consistent, especially in challenging subjects like physics [39,85]. The limitations of PBL as an approach for altering students’ attitudes and beliefs should be acknowledged, as this intervention may have been more successful in targeting cognitive outcomes rather than affective ones.

The PBL in our study is designed to engage students in solving both theoretical and practical problems to ensure a comprehensive understanding of fundamental physics principles and their applications. While some problems may appear traditional, they serve as a foundation for more complex real-world scenarios [26]. Barrows discusses various PBL methods and highlights the importance of combining theoretical knowledge with practical applications to enhance problem-solving skills [24]. Hmelo-Silver emphasizes that PBL involves students working on complex problems that require both theoretical understanding and practical application, fostering a deeper understanding of the subject matter [25].

17. Conclusions

This study emphasizes the importance of examining both cognitive and affective outcomes in PBL research, especially within high school physics, where students’ attitudes and beliefs can strongly impact their engagement and achievement. While PBL has demonstrated benefits in promoting problem-solving skills and deepening subject understanding, its potential to shape students’ attitudes and beliefs about challenging subjects like physics remains less consistent. This study contributes uniquely by investigating these affective dimensions, adding to the broader understanding of how instructional approaches influence students’ perceptions and motivation in STEM fields. Future research is essential to explore the nuanced effects of PBL on affective outcomes, considering factors such as intervention duration, classroom dynamics, and subject-specific challenges. Expanding on this research can help educators design PBL interventions that not only enhance cognitive gains, but also positively shape students’ attitudes, supporting a more holistic approach to learning in high school science education.

This study engaged a small number of students and one teacher who used the PBL approach. There is a need to gather a wider sample and to attempt this study across more diverse areas of the enormous country. Further, because of the surprising result that did not show an association between learning gains and attitudes, there is a need for closer ethnographic research to determine how students and teachers understand the problem-posing context and how the new approach to teaching fits into the organizational context such as that which exists in this study’s context of Kazakhstan. Studies show that PBL requires enabling conditions for it to flourish, such as support from the systems that surround the schools, including assessment and other stakeholders such as the community and parents [93]. This research offers exciting possibilities for accommodating the PBL approach, such as understanding the wider system’s support; novel approaches to problem-solving in physics; and redesigning teachers’ professional learning, problem types, and assessments.

The following limitations should be considered when interpreting the findings of this research. First, the study was conducted with a relatively small sample size of 63 students from a single public school in Almaty, Kazakhstan. This limits the generalizability of the findings to other schools and regions. Second, the use of convenience sampling may have introduced selection bias. The selected school and participants may not fully represent the broader population of high school students in Kazakhstan or other contexts. Third, one of the researchers also served as the teacher for the experimental and control groups. This dual role may have introduced bias, despite efforts to maintain professional boundaries. Fourth, the PBL intervention was implemented over a seven-week period. While this duration allowed for initial insights into the effects of PBL, it may not have been sufficient to capture longer-term impacts on student attitudes, beliefs, and academic performance. Fifth, there was a notable gender imbalance between the experimental and control groups, with approximately 72% of the experimental group comprising female students compared to 52% in the control group. Given that the teacher implementing PBL was female, this gender disproportion may have influenced the dynamics within the experimental group and the overall outcomes. Sixth, the timing of the intervention, which took place later in the school year, presents another limitation. By this stage, students may have already had established attitudes and beliefs about physics, which could have been challenging to alter through a relatively short intervention.