Challenge-based learning implementation in engineering education: A systematic literature review

1.1.1 Existing definitions and conceptualizations of CBL

Another popular definition of CBL specific to engineering education can be found in the work of Malmqvist et al. (2015). They defined CBL in engineering education as “the identification, analysis, and design of a solution to a sociotechnical problem. The learning experience is typically multidisciplinary, involves different stakeholder perspectives, and aims to find a collaboratively developed environmentally, socially and economically sustainable solution.”

Finally, van den Beemt, van de Watering, and Bots (2023) suggested that existing conceptualizations of CBL vary and highlighted the need for a comprehensive framework to capture this variety within CBL implementation in higher education. Based on a literature search and reflecting on current practices in their institutions, they suggested a model in which CBL is analyzed based on its vision, teaching and learning practices, and support systems. They suggested 11 relevant indicators of CBL, at the level of vision, teaching and learning, and support practices. They also reflected on the importance of having minimum requirements for an educational intervention to be called CBL. They defined three minimum indicators for a CBL educational intervention. Those included (i) real-life and authentic challenges that (ii) stimulate students' combination of deep content understanding and a general view of engineering issues and (iii) the use of learning activities that promote a rigorous treatment of fundamental engineering knowledge and skills.

All the above definitions emphasize the importance of real-world and authentic challenges as the starting point for students' learning. The importance of students' engagement in active, collaborative, (multi)disciplinary learning to tackle those challenges is also stressed in terms of the learning process. The definition by Malmqvist et al. (2015) focused on the characteristics of the solution for the challenge, and its wider scope as including not only technical but also societal considerations. Finally, in addition to the characteristics of the challenge, van den Beemt, van de Watering, and Bots (2023) focused on the students' expected learning, with an emphasis on developing disciplinary and transversal skills.

1.1.2 Comparison of CBL and other active learning pedagogies

Studies of CBL have suggested that a deeper theoretical grounding in educational science and learning theories could enrich the CBL literature (Leijon et al., 2022; van den Beemt, Vázquez-Villegas, et al., 2023). Given that CBL promotes real-world, active, self-directed, and collaborative learning (Sukackė et al., 2022; van den Beemt, van de Watering, & Bots, 2023; van den Beemt, Vázquez-Villegas, et al., 2023), it resonates with learning theories and active learning pedagogies, such as problem-based learning (PBL), project-based learning (PjBL), and design-based learning (DBL) in terms of learning principles (Leijon et al., 2022; Sukackė et al., 2022; van den Beemt, van de Watering, & Bots, 2023; van den Beemt, Vázquez-Villegas, et al., 2023). These pedagogies all aim to link students' academic knowledge with professional practice by introducing them to problems, tasks, or challenges relevant to their future practice (Sukackė et al., 2022). However, they also differ across critical dimensions, including the focus of learning, the type of problems/challenges used to trigger students' learning, the learning activities, the type of solution students are required to develop, and the teacher's role in the learning process.

Although an exhaustive comparison of CBL with other pedagogies is outside this review's scope, in Table 1 we briefly summarize some critical differences across CBL, PBL, PjBL, and DBL as have been reported so far in the literature (e.g., Binder et al., 2017; Guo et al., 2020; Holgaard et al., 2017; Kohn Rådberg et al., 2020; Kolmos & de Graaff, 2014; Krajcik & Shin, 2014; Membrillo-Hernández, Ramírez-Cadena, et al., 2019; Sukackė et al., 2022; van den Beemt, Vázquez-Villegas, et al., 2023).

Table 1 presents some prototypical examples of the four pedagogies as a way to highlight their fundamental differences regarding the type of challenges/problems students are presented with, the learning process, the role of the teacher, and the expected outcomes. However, it is noteworthy that in practice, similarities and boundaries among these paradigms are often nuanced and less distinct. An example is Aalborg University, where elements of PBL and PjBL are combined in the engineering curriculum (Kolmos, 2012; Kolmos & de Graaff, 2014).

In terms of learning, CBL drives students' knowledge integration and application through engagement in real-world challenges that lack a predefined solution. Challenges in CBL are sociotechnical, which means that they require addressing pressing current issues (e.g., climate change and energy transitions, data privacy) which require a combination of technological innovations, policy development, and societal engagement to find effective solutions (Malmqvist et al., 2015; McGowan & Bell, 2020). Students should work toward a solution using technical knowledge from various disciplines, taking into account the complex interplay between societal, technological, and environmental factors, and must present their (proposed) solution to a panel of experts and stakeholders (Gallagher & Savage, 2020; Kohn Rådberg et al., 2020; Membrillo-Hernández, Ramírez-Cadena, et al., 2019). The challenges presented to students in CBL differ from those of PBL where problems are often predefined and developed by the teachers as vehicles for knowledge acquisition (Gijselaers, 1996; Kolmos et al., 2019; Zin et al., 2017), and PjBL where problems are often predefined and developed by the teachers for students to learn and apply new knowledge and skills. CBL also differs from DBL in this respect, as design challenges tend to be more specific and focus on creating an artifact or product, emphasizing the function of the particular production of artifacts (Gómez Puente et al., 2011, 2013). In terms of product, in CBL, students are expected both to focus on their learning process, as in PBL, where the main focus is on reasoning and the learning process rather than the product, and to develop a concrete solution, similar to PjBL and DBL (Gallagher & Savage, 2020; Gómez Puente et al., 2011, 2013; Membrillo-Hernández, Ramírez-Cadena, et al., 2019). Finally, teachers in CBL act as coaches and co-creators of a solution, in addition to being a tutor and facilitator of knowledge acquisition, which are the usual teacher roles in PBL, PjBL, and DBL (Gallagher & Savage, 2020; Gómez Puente et al., 2013).

1.1.3 A working definition of CBL

CBL builds on the strengths of other active learning pedagogies and encourages students to learn autonomously and collaboratively in a contextualized way. CBL aims to develop students' disciplinary and transversal competence by having them work with challenges. What is new is the type of challenges students are invited to tackle. In CBL, challenges to be tackled by the students are societally relevant and connected with sustainable development goals (SDGs), often presented by external partners (from industry partners to local communities). Regarding the learning process, emphasis is given to the collaboration between students, teachers, and external stakeholders as the point of departure for learning and to the co-creation of a solution for the challenge. The outcome of such a co-creation process is development of a tangible solution that addresses both technical and societal considerations.

3.1.1 Theoretical frameworks used in the studies

To be able to interpret the findings of each study, we first analyzed the theoretical background used for the study, the research aims, and the methodology selected to attain those aims. Although mentions of the theoretical grounding of the research were sparse in most of the studies analyzed, we identified three frequently mentioned theoretical frameworks guiding the implementation of CBL. Those were experiential learning, active learning, and self-determination theory. Several studies mentioned experiential learning as a theoretical framework (Chanin et al., 2018; Charosky & Bragós, 2021; Jensen et al., 2018; Palma-Mendoza et al., 2019). These studies recognized the value of hands-on experiences, projects, and real-world application of knowledge in enhancing student learning, developing innovation competencies, and preparing students for future challenges. In addition, a number of studies shared the common use of active learning as a theoretical framework (Arrambide-Leal et al., 2019; Chanin et al., 2018; Cheung et al., 2011; Gaskins et al., 2015; Membrillo-Hernández et al., 2018; Membrillo-Hernández & García-García, 2020; Membrillo-Hernández, Munoz-Soto, et al., 2019; Membrillo-Hernández, Ramírez-Cadena, et al., 2019; Torres-Barreto et al., 2020). Active learning was used to encourage students to engage in the learning process, promoting deeper understanding and skills development. Finally, a few studies drew on theories such as self-determination theory to highlight the importance of engaging students in their learning process and motivating them to participate actively (Bombaerts, 2020; Bombaerts et al., 2021).

3.1.2 Research aims and methodologies of the included studies

The common focus of the studies was the implementation of CBL as an educational approach in various engineering and technology-related courses and projects (e.g., Binder et al., 2017; Charosky et al., 2018; Cheung et al., 2011; Gama, Alencar, et al., 2018; Gama, Castor, et al., 2018; Portuguez Castro & Gómez Zermeño, 2020). However, the research aims of the studies differed. Some of the studies focused on comparing the effectiveness of CBL with other teaching approaches, such as case-based learning or traditional methods (Bombaerts et al., 2021; Charosky et al., 2018; Charosky & Bragós, 2021; Gama, Castor, et al., 2018). Some studies examined the impact of CBL on students' learning outcomes and competencies (Charosky & Bragós, 2021; Gudonienė et al., 2021; Martínez & Crusat, 2017; Membrillo-Hernández, Ramírez-Cadena, et al., 2019) or student engagement and motivation (Bombaerts et al., 2021; Gutiérrez-Martínez et al., 2021; Huettel et al., 2015). Many studies aimed to evaluate the effectiveness of CBL in solving real-world challenges or addressing societal issues through engineering education in multidisciplinary contexts or collaborative projects (e.g., Fidalgo-Blanco et al., 2016; Gama, Alencar, et al., 2018; Gama, Castor, et al., 2018; Hassi et al., 2016; Maya et al., 2017).

In terms of the methodologies, the majority of the studies analyzed cases of implementation of CBL in a specific course or setting. They employed mixed methods, combining quantitative and qualitative data collection and analysis techniques (e.g., Binder et al., 2017; Bombaerts et al., 2021; Chanin et al., 2018; Cirenza & Diller, 2015; Clegg & Diller, 2019; Gama, Alencar, et al., 2018; Gama, Castor, et al., 2018; Lasso-Lopez et al., 2020; Membrillo-Hernández & García-García, 2020; Membrillo-Hernández, Munoz-Soto, et al., 2019; Membrillo-Hernández, Ramírez-Cadena, et al., 2019; Membrillo-Hernández et al., 2018, 2021). A mixed-method approach allowed for developing a comprehensive understanding of the impact of CBL on various aspects of engineering education, as well as collecting information on students' and teachers' perceptions of the implementation. Studies such as those by Arrambide-Leal et al. (2019), Charosky and Bragós (2021), Cheung et al. (2011), Gaskins et al. (2015), Gudonienė et al. (2021), Huettel et al. (2015), and Rodríguez-Chueca et al. (2020) adopted quantitative methods, involving surveys and assessments to measure the effectiveness and outcomes of CBL in engineering education. Qualitative methods were employed in studies that focused on gathering in-depth insights and perceptions of students through interviews, reflections, and document analysis (e.g., Charosky & Bragós, 2021; da Costa et al., 2018; Pepin & Kock, 2021; Quweider & Khan, 2016; Santos et al., 2015; Torres-Barreto et al., 2020; Valencia et al., 2020). Regarding the representativeness of the student population in the included studies, most of the studies did not report whether the included sample of students was representative of the student population in their university; thus, claims about the generalizability of findings should be carefully examined.

3.2.1 CBL implementation at the course level

At the course level, the rationale for using CBL as a pedagogy was to overcome the shortcomings of traditional teacher-centered approaches by focusing on active learning and engagement of students with disciplinary content, using a real-world challenge as a starting point (Clegg & Diller, 2019; Gaskins et al., 2015; Huettel et al., 2015; Lovell et al., 2013). CBL courses aimed at various learning aims and objectives, including students' in-depth understanding of content knowledge (Clegg & Diller, 2019; Detoni et al., 2019; Gaskins et al., 2015), knowledge transfer (Clegg & Diller, 2019; Gaskins et al., 2015; Rodríguez-Chueca et al., 2020), application of knowledge (Gutiérrez-Martínez et al., 2021; Santos et al., 2015), and development of disciplinary and transferable skills (Félix-Herrán et al., 2019; Mora-Salinas et al., 2019; Membrillo-Hernández, Munoz-Soto, et al., 2019; Membrillo-Hernández, Ramírez-Cadena, et al., 2019; Membrillo-Hernández et al., 2018; Ramirez-Mendoza et al., 2018).

Seven CBL courses focused on content and disciplinary knowledge acquisition (Cirenza & Diller, 2015; Clegg & Diller, 2019; Fidalgo-Blanco et al., 2016; Gaskins et al., 2015; Huettel et al., 2015; Lovell et al., 2013; Quweider & Khan, 2016). The teacher developed and tailored the challenges in those courses to support students' disciplinary content learning (Clegg & Diller, 2019; Gaskins et al., 2015). On the other hand, the remaining 22 articles described the implementation of CBL courses that set as learning objectives students' acquisition and application of disciplinary knowledge, as well as the development of transversal skills such as ethics, problem-solving skills, leadership, communication, and collaboration skills (Cuevas-Ortuño & Huegel, 2020; Gonzalez-Hernandez et al., 2020; Ramirez-Mendoza et al., 2018). Those studies represented the largest body of articles in this review.

In terms of challenge characteristics, challenges were open-ended and real-world; in 14 cases, external industry partners and societal stakeholders were involved, increasing the complexity of the challenges (Bombaerts et al., 2021; Membrillo-Hernández & García-García, 2020). Students had to use prior knowledge and acquire new knowledge for developing a solution for the challenge by following a process of identifying and analyzing the challenge, designing a solution, and presenting it (Gonzalez-Hernandez et al., 2020; Palma-Mendoza et al., 2019; Ramirez-Mendoza et al., 2018). To guide this process, six studies (Chanin et al., 2018; da Costa et al., 2018; Detoni et al., 2019; Palma-Mendoza et al., 2019; Portuguez Castro & Gómez Zermeño, 2020; Rodríguez-Chueca et al., 2020) used the framework developed by Apple (Nichols & Cator, 2008). According to the Apple framework, students had to define a big idea, a broad concept that they could explore in several ways. Then, following the big idea, students had to develop essential questions, define their challenge, develop guiding activities, questions, and resources, and develop, implement, and evaluate their solution. Two studies (Clegg & Diller, 2019; Cuevas-Ortuño & Huegel, 2020) adopted the STAR (software technology action reflection) Legacy Cycle methodology, which consisted of the following steps: Students had to design the challenge, generate the ideas, approach the challenges from different viewpoints, conduct research, revise, assess the skills needed for the development of the solution, and publish the solution.

CBL courses described various learning activities and resources to support students' learning. As knowledge acquisition was a key objective for CBL courses, lectures, modules, textbooks on foundational subjects, and disciplinary content were provided to students (Gaskins et al., 2015; Huettel et al., 2015; Lovell et al., 2013; Quweider & Khan, 2016). Furthermore, to support students in the CBL process of solution development, workshops on transversal skills such as problem solving and creative thinking, resources to facilitate collaboration and design processes such as group discussions, feedback by peers and external stakeholders, debates, and technology tools were also used (Clegg & Diller, 2019; Membrillo-Hernández, Ramírez-Cadena, et al., 2019).

In CBL courses, teachers have various roles, including being the designers of the challenge (Clegg & Diller, 2019), developing learning modules and lectures to provide students with theoretical input to accompany students' work (Detoni et al., 2019), coaching students (Bombaerts et al., 2021; Mora-Salinas et al., 2019), providing feedback on students' deliverables such as presentations and reports, and assessing students' achievement in terms of competency development and project outcomes (Binder et al., 2017; Gaskins et al., 2015). Several studies mentioned that a multidisciplinary group of teachers collaborated in the development of the challenges and the implementation of the CBL course (da Costa et al., 2018; Félix-Herrán et al., 2019; Membrillo-Hernández, Ramírez-Cadena, et al., 2019). Regarding grouping, all studies required students to work together as a team. The teams comprised peers from the same discipline (Félix-Herrán et al., 2019) or different disciplines (da Costa et al., 2018; Membrillo-Hernández, Ramírez-Cadena, et al., 2019). Often, the teams were even broader, involving collaboration among students, teachers, and stakeholders (Ramirez-Mendoza et al., 2018).

The articles mentioned several assessment methods for the students depending on the learning objectives of each course. Below, we provide only some examples of assessment instruments used for different learning objectives. First, assessment of students' knowledge acquisition was conducted via individual exams, final individual or group reports, and presentations (Clegg & Diller, 2019; Gaskins et al., 2015; Lovell et al., 2013). Second, knowledge application was assessed via prototypes, final reports, and presentations (Membrillo-Hernández et al., 2018; Rodríguez-Chueca et al., 2020). Third, students' learning process was assessed via students' reflections and portfolios (Chanin et al., 2018; da Costa et al., 2018). Finally, transversal competencies (e.g., collaboration) were assessed using self-reported questionnaires and competence rubrics (Gutiérrez-Martínez et al., 2021; Membrillo-Hernández et al., 2019).

In terms of time, in CBL courses, students had both synchronous activities (e.g., lectures or workshops) and asynchronous activities (self-study of modules) encouraging self-paced learning (Gutiérrez-Martínez et al., 2021; Membrillo-Hernández & García-García, 2020). Regarding CBL courses' location, CBL courses that involved stakeholders mentioned the importance of students' learning outside lecture rooms and classrooms. For example, students considered visits to the stakeholders' premises and libraries, as well as working in open, collaborative spaces, very inspiring (Membrillo-Hernández & García-García, 2020).

3.2.2 CBL implementation at the project level

CBL projects presented great variability, and three main types were identified: (i) CBL bachelor's and master's thesis projects (Kohn Rådberg et al., 2020; Maya et al., 2017; Pepin & Kock, 2021; Valencia et al., 2020); (ii) CBL projects that were part of a collaboration between different departments, faculties, or institutes and usually were intensive and lasted a couple of weeks (Charosky & Bragós, 2021; Charosky et al., 2018; Cheung et al., 2011; Hassi et al., 2016; Jensen et al., 2018; Lasso-Lopez et al., 2020; Martínez & Crusat, 2017; Torres-Barreto et al., 2020); and finally (iii) extracurricular CBL projects such as participation in competitions and hackathons (Arrambide-Leal et al., 2019; Gama, Alencar, et al., 2018; Lara-Prieto et al., 2019; López-Fernández et al., 2020).

CBL projects aimed to support students in tackling open-ended and real-world engineering challenges (Charosky & Bragós, 2021; Gama, Castor, et al., 2018; Jensen et al., 2018; Lasso-Lopez et al., 2020) that had a broader focus on sociotechnical challenges connected with global issues, often termed grand challenges (Kohn Rådberg et al., 2020; Maya et al., 2017; Pepin & Kock, 2021; Torres-Barreto et al., 2020). These challenges aimed to benefit an individual or community by promoting a collaborative and multidisciplinary approach and fostering students' awareness about the responsibility of engineers in society (Kohn Rådberg et al., 2020; Maya et al., 2017).

CBL projects that were not tied to specific courses or disciplinary content placed more emphasis on developing solutions for the challenges identified by the students. In terms of the learning process, students needed to identify and address gaps in their knowledge and skills (Gama, Alencar, et al., 2018; Hassi et al., 2016), integrate knowledge from multiple disciplines and independently master this content to develop a solution, and tackle challenges (Jensen et al., 2018; Kohn Rådberg et al., 2020; López-Fernández et al., 2020). The learning process was characterized by openness and iteration. As students worked on open-ended challenges, they learned to switch between divergent and convergent phases (Charosky et al., 2018; Hassi et al., 2016; Jensen et al., 2018). During divergent phases, students were expected to expand and explore their understanding of challenges, redefine them, generate ideas, and engage in explorations with users. Students also learned to select the best solution from various options, defend their choice, and communicate the results to relevant stakeholders (Charosky et al., 2018; Hassi et al., 2016). Furthermore, hands-on learning played a central role in CBL projects. Students were required to apply theoretical knowledge to real-world challenges, conduct experiments, design solutions, and reflect on their experiences (Chanin et al., 2018; Cheung et al., 2011; Lara-Prieto et al., 2019; Rodríguez-Chueca et al., 2020). They were expected to act in a self-directed and collaborative manner (López-Fernández et al., 2020).

In terms of grouping, the teams in CBL projects included not only students but also teachers and stakeholders (Charosky et al., 2018; Charosky et al., 2022; Gama, Alencar, et al., 2018; Gudonienė et al., 2021; Martínez & Crusat, 2017; Jensen et al., 2018; Torres-Barreto et al., 2020). Students had the freedom to decide how to assign different roles in their groups and were responsible for distributing their tasks. Multidisciplinary groups of students were mentioned in 12 CBL projects (Charosky & Bragós, 2021; Charosky et al., 2018, 2021; Gama, Alencar, et al., 2018; Gudonienė et al., 2021; Jensen et al., 2018; Kohn Rådberg et al., 2020; López-Fernández et al., 2020; Martínez & Crusat, 2017; Maya et al., 2017; Pepin & Kock, 2021; Torres-Barreto et al., 2020; Valencia et al., 2020).

Teachers in CBL projects were expected to fulfill the role of expert, coach, and co-learner (Charosky et al., 2018, 2021; Gama, Alencar, et al., 2018; Gudonienė et al., 2021; Jensen et al., 2018; Martínez & Crusat, 2017; Torres-Barreto et al., 2020).

Learning activities and resources involved group discussions, online modules, workshops on collaboration, and feedback sessions (López-Fernández et al., 2020; Martínez & Crusat, 2017; Torres-Barreto et al., 2020). Information and communications technology (ICT) tools and technology platforms were used to support students' collaboration and project management (Jensen et al., 2018; López-Fernández et al., 2020).

Assessment of CBL projects focused on the quality of the solutions that were developed. Students usually produced a report and a presentation to assess the solution for the challenge, where they provided argumentations for their suggested solution. This product was usually evaluated with rubrics and checklists of criteria (Charosky et al., 2018; Hassi et al., 2016; Jensen et al., 2018; Martínez & Crusat, 2017; Torres-Barreto et al., 2020).

The duration of CBL projects differed significantly. Some CBL projects had a very brief duration of a few days (López-Fernández et al., 2020; Gama, Alencar, et al., 2018) or a few weeks (Charosky & Bragós, 2021), and others lasted a semester, in the case of bachelor's and master's thesis projects (Kohn Rådberg et al., 2020; Valencia et al., 2020).

Regarding location, some studies stressed the importance of the physical learning environment as an aspect that facilitated learning within CBL and the use of maker spaces and innovation labs as creative and interactive locations ideal for students' collaboration (Jensen et al., 2018; Kohn Rådberg et al., 2020; Valencia et al., 2020).

3.3.1 Development of complex sociotechnical challenges with the collaboration of external stakeholders

Teachers reported difficulties related to the development of challenges. For example, in the case of CBL courses, developing an open-ended and complex challenge was reported to conflict with the limitations of traditional learning environments, such as time constraints and the need to align the challenge with intended learning objectives (Bombaerts et al., 2021; Fidalgo-Blanco et al., 2016; Gaskins et al., 2015; Quweider & Khan, 2016). To achieve this, teachers had to scope the developed challenge to be manageable for the students within the time allowed for a course (Membrillo-Hernández et al., 2021).

The involvement of real-world stakeholders made the challenges relevant for students (Bombaerts et al., 2021; Membrillo-Hernández et al., 2021; Pepin & Kock, 2021). However, co-designing a challenge in collaboration with industry partners presented some difficulties for teachers, who reported struggling to align industrial partners' interests with a challenge that provides a good learning experience for students (Bombaerts et al., 2021; Membrillo-Hernández et al., 2021; Mora-Salinas et al., 2019; Ramirez-Mendoza et al., 2018; Valencia et al., 2020). In some reported cases, when stakeholders were involved, the challenges were open-ended and ill-defined to the extent that, in several cases, the teachers did not know in advance what the exact challenge or the solution was, which means that the levels of uncertainty were high not only for students but also for the teachers (Hassi et al., 2016).

Finally, time and effort spent were two important aspects influencing teachers' experiences with CBL (Bombaerts et al., 2021; Cuevas-Ortuño & Huegel, 2020; Membrillo-Hernández et al., 2021; Mora-Salinas et al., 2019; Palma-Mendoza et al., 2019). Teachers faced a heavier workload in CBL courses. For teachers, developing a challenge demands time, resources, and networking. The study by Bombaerts et al. (2021) mentioned that preparing and conducting a CBL course required approximately 60% more time than a traditional case-based approach. However, whether the time and effort remained the same throughout all future iterations of the same CBL course or project was not reported. In agreement with Bombaerts et al. (2021), Papageorgiou et al. (2021) reported the need for educators to exhibit flexibility and improvisation when designing and implementing CBL, because the necessary hours invested cannot be exactly predicted at the beginning of a CBL course or project.

3.3.2 Development of effective CBL assessment methods

When designing the CBL courses, choosing or developing suitable assessment methods was another challenge for teachers (Membrillo-Hernández et al., 2021; Pepin & Kock, 2021; Valencia et al., 2020). Furthermore, aligning the assessment methods with the various learning objectives was difficult in cases where challenges were open-ended, as teachers did not have an overview from the beginning as to what students were expected to learn (Hassi et al., 2016; Membrillo-Hernández et al., 2021; Valencia et al., 2020). Another difficulty related to assessment was capturing individual students' discipline-specific learning gains when students mostly worked in collaborative and often multidisciplinary groups (Valencia et al., 2020). In addition, assessment of transferable skills such as communication, teamwork, problem-solving, and self-directed learning also presented difficulties because students' development of those skills often requires a longer exposure to CBL than a single course or project, and students' development is difficult to capture after only one CBL experience (Membrillo-Hernández et al., 2021).

3.3.3 Need for CBL training for teachers

As reported in many articles, teachers were challenged to adjust their roles from lecturers to facilitators and coaches in the CBL environment (Membrillo-Hernández et al., 2021; Pepin & Kock, 2021). Without theoretical knowledge and skills regarding CBL methods, teachers often encountered difficulties designing the course activities, facilitating students' teamwork, and balancing between helping and influencing the students' work (Pepin & Kock, 2021). Membrillo-Hernández et al. (2021) and Charosky et al. (2018) stressed the importance of training teachers in CBL before engaging in CBL-based education, while Valencia et al. (2020) mentioned the importance of teachers' assessment literacy for designing and implementing effective assessment methods that are aligned with the CBL's intended learning objectives.

3.3.4 Need for support from departments and institutions for CBL implementation beyond the course or project level

As most studies described CBL implementation in single courses and projects at the system level, an important difficulty was the lack of support from departments or institutions to support CBL implementation beyond the course or project level (Membrillo-Hernández et al., 2021). Teachers reflected that they need more supportive materials, resources, and policies from the faculty or university to improve the effectiveness of CBL (Charosky et al., 2021; Cuevas-Ortuño & Huegel, 2020; Membrillo-Hernández et al., 2021).

3.3.5 Students' difficulties in balancing knowledge acquisition and application

In the early years of their bachelor's studies when they had yet not built solid disciplinary knowledge, working on complex and open-ended challenges proved difficult for students (Binder et al., 2017; Membrillo-Hernández et al., 2021). That was especially true for CBL courses involving a real-world challenge with an external stakeholder (Membrillo-Hernández, Munoz-Soto, et al., 2019; Membrillo-Hernández, Ramírez-Cadena, et al., 2019; Membrillo-Hernández et al., 2021; Mora-Salinas et al., 2019; Palma-Mendoza et al., 2019; Ramirez-Mendoza et al., 2018). In those studies, it was commonly reported that students struggled to combine knowledge acquisition and implementation simultaneously within the timeframe of one semester.

3.3.6 Students' difficulties in dealing with the ambiguity and uncertainty of open-ended challenges

One crucial aspect that influenced students' experience with CBL was the uncertainty that was integral to CBL (Arrambide-Leal et al., 2019; Kohn Rådberg et al., 2020). The openness and ill-defined nature of challenges combined with the student's need for self-directed learning and initiative created uncertainty and sometimes resistance (López-Fernández et al., 2020; Membrillo-Hernández, Ramírez-Cadena, et al., 2019). Several researchers pointed out that it was challenging for students to narrow down broad challenges and translate knowledge into practical solutions (Rodríguez-Chueca et al., 2020). Within this context, students felt insecure about achieving the desired outcomes. As a result, some studies suggest introducing students first to a smaller-scale challenge as an exercise, so that they could practice the process before moving to bigger and more complex challenges (Fidalgo-Blanco et al., 2016).

As the expected outcome and the process were not known in advance, students experienced increased levels of uncertainty when navigating an open-ended challenge (Gama, Castor, et al., 2018; Membrillo-Hernández, Ramírez-Cadena, et al., 2019). In the study by Jensen et al. (2018), students mentioned difficulties in the analysis phase of the CBL project, leading to difficulty transitioning from the divergent to convergent stages of CBL. Similar findings were reported in the studies by da Costa et al. (2018) and Detoni et al. (2019). In the study by da Costa et al. (2018), students experienced difficulties narrowing down the broad challenge and developing guiding questions, and in the study by Detoni et al. (2019), students reported difficulties in the investigation phase, mainly when conducting research necessary to analyze the challenge. Other student-reported difficulties were associated with a lack of technical knowledge (Binder et al., 2017), lack of familiarity with the CBL process (López-Fernández et al., 2020), and difficulties with self-study (Cheung et al., 2011). Students also struggled with time constraints in CBL. They reported needing more time than for traditional courses or projects (Cuevas-Ortuño & Huegel, 2020; Valencia et al., 2020). Many studies suggested that new pedagogies and CBL processes were unfamiliar to students, and more time was needed to adapt (López-Fernández et al., 2020; Membrillo-Hernández, Ramírez-Cadena, et al., 2019). That was especially true for CBL courses, where within one semester students had to master knowledge and prepare for an exam, as well as to work on a real-world challenge and communicate their solution to the clients or external stakeholders (Detoni et al., 2019; Membrillo-Hernández, Ramírez-Cadena, et al., 2019; Membrillo-Hernández et al., 2021).

3.3.7 Need to provide training in transversal skills for students

In addition, students are expected to be at the center of learning in the CBL context, which requires higher self-regulated learning skills for students and working as a member of a group which was often multidisciplinary (Pepin & Kock, 2021; Valencia et al., 2020). Students lacking intrinsic motivation or self-reflection skills often found their experience with CBL and teamwork negatively affected (Rodríguez-Chueca et al., 2020). In addition, conflicts arising from teamwork were frequently mentioned by researchers (López-Fernández et al., 2020), highlighting the importance of training students in transversal skills and group processes. López-Fernández et al. (2020) reported that a combination of students' low motivation, low technical competencies, and aversion to working autonomously could negatively influence their experience with CBL. The study by Jensen et al. (2018) focused on multidisciplinary collaboration and stressed the importance of teaching students teamwork and communication skills in order to decrease the occurrence of disciplinary egocentrism.

4.7.1 Development/design of challenges

Our results suggest that CBL courses and projects use a real-world challenge as the central vehicle for instruction in CBL classrooms (Mora-Salinas et al., 2019; Rodríguez-Chueca et al., 2020; Valencia et al., 2020). The difficulty of designing open-ended and complex, yet manageable, challenges within course constraints showcases the intricate balance required in CBL design. Educators need to strike a balance that stimulates students' interest, creativity, and problem-solving without overwhelming them. Although collaboration with industry partners and other stakeholders can enhance the real-world relevance of challenges, it also introduces complexities. Aligning educational objectives with industrial interests can be challenging but essential for ensuring that students derive valuable learning experiences. It is recommended that educators allocate enough time for the development of challenges and collaborate with various experts, such as educators from different disciplines and educational researchers (Félix-Herrán et al., 2019; Kohn Rådberg et al., 2020; Membrillo-Hernández et al., 2021). When developing a CBL course or project, educators should define learning goals at various levels, including disciplinary and transversal competencies, individual or group objectives, and expected learning outcomes (Membrillo-Hernández et al., 2021; Valencia et al., 2020). In addition, educators should agree beforehand on the relevant knowledge, skills, and attitudes engineering students should develop by participating in a specific CBL course or project (Valencia et al., 2020). Finally, it is important to involve external stakeholders early in the discussion and make sure that their expectations about intended objectives and outcomes align with the course or project objectives to ensure meaningful, real-world challenges for students (Bombaerts et al., 2021; Membrillo-Hernández et al., 2021; Mora-Salinas et al., 2019; Ramirez-Mendoza et al., 2018; Valencia et al., 2020).

4.7.2 Development of appropriate assessment methods

Assessment of students' learning was highlighted as a CBL aspect that needs further attention from educators (Membrillo-Hernández et al., 2021; Pepin & Kock, 2021; Valencia et al., 2020). Achieving alignment between learning goals and assessment methods presents significant challenges, especially when students from different disciplines collaborate (Valencia et al., 2020). In addition, the learning process in CBL is considered as important as the final product or solution development, emphasizing individual and group learning (da Costa et al., 2018; Membrillo-Hernández et al., 2021). Thus, a combination of assessments focusing on the individual and the group, process, and products should be adopted to enhance the validity of the assessment of learning outcomes (Gallagher & Savage, 2020; van den Beemt, van de Watering, & Bots, 2023).

4.7.3 Scaffolding students' self-directed learning

Students reported difficulties in acting autonomously during CBL, dealing with uncertainty and the open-ended aspect of challenges, and self-regulating their learning (Bombaerts et al., 2021; Lara-Prieto et al., 2019; Membrillo-Hernández, Ramírez-Cadena, et al., 2019). Despite the emphasis on self-directedness, our study suggests that students did not always welcome the autonomy provided in CBL (López-Fernández et al., 2020). Thus, we recommend that before immersing students in full-fledged CBL projects, institutions or educators could consider introducing them to smaller, more manageable challenges. Introducing smaller challenges first can serve as a warm-up, familiarizing students with the CBL approach (Fidalgo-Blanco et al., 2016). Given the autonomy expected from students, the role of teachers was also different compared to traditional teaching approaches. The teacher was viewed as a coach, rather than an expert, who guided students through the learning process (Membrillo-Hernández, Munoz-Soto, et al., 2019; Membrillo-Hernández, Ramírez-Cadena, et al., 2019). Thus, we suggest that self-regulated learning should be supported by coaching and scaffolding from teachers and adjusted to the students' current level of skill and understanding (Doulougeri, van den Beemt, et al., 2022; Doulougeri, Vermunt, et al., 2022).

4.7.4 Supporting students in the collaborative co-creation process

All studies in our review suggested that students in CBL learn while being part of a group, indicating that collaborative learning is a central feature of CBL. However, when some students had not had prior experiences with collaborative and open-ended courses, this adversely affected the CBL experience and teamwork (López-Fernández et al., 2020; Pepin & Kock, 2021; Rodríguez-Chueca et al., 2020). Thus, the review highlights the importance of fostering collaborative skills (Clegg & Diller, 2019; Martínez & Crusat, 2017). CBL studies described multidisciplinary or interdisciplinary collaboration that extended beyond the group of students and involved collaboration with teachers and often with external stakeholders (Bombaerts et al., 2021; Hassi et al., 2016; Jensen et al., 2018). Through collaboration, students learned to explore challenges from different perspectives via continuing discussion and integration of multiple perspectives and expertise (Jensen et al., 2018). However, our findings suggested that the collaborative process is easier said than done in CBL, with difficulties reported by students in collaborating and communicating with peers and stakeholders (Jensen et al., 2018; López-Fernández et al., 2020). It is necessary to establish processes by which students can check up on their collaboration throughout CBL. To facilitate collaboration during CBL, educators should consider carefully how to organize the course/project to facilitate formative interactions between all parties involved. In addition, to accompany CBL courses, institutions should offer workshops or modules focused on transversal skills such as collaboration, self-regulated learning, and conflict resolution to equip students with the requisite skills to excel in CBL contexts (López-Fernández et al., 2020).

4.7.5 Professional development of teachers

Teachers faced challenges in adjusting their roles from being traditional lecturers to acting as facilitators and coaches in the CBL environment. Without adequate knowledge and skills regarding CBL methods, teachers often found it difficult to design course activities, facilitate student teamwork, and balance assisting and influencing students' work. The review highlights that not only do students need to shift toward active and self-directed learning, but teachers need to redefine their role, as coaches, co-learners, and co-creators of solutions (Membrillo-Hernández et al., 2021; Pepin & Kock, 2021). Therefore, to improve CBL implementation and effectiveness, it is important to provide CBL pedagogical training to enable engineering staff to develop CBL pedagogical knowledge and coaching skills and invest additional time in designing a CBL course or project (Pepin & Kock, 2021). Institutions should offer comprehensive training programs for educators, equipping them with the skills, methodologies, and best practices specific to CBL (Membrillo-Hernández et al., 2021).

4.8.1 Developing rigorous CBL educational research

The current landscape of research on CBL in engineering education, as observed in this review, brings to light a pressing concern. Many investigations of CBL practices, while promising, remain confined to small-scale, bottom-up case studies that often need more robust theoretical grounding (Leijon et al., 2022; van den Beemt, van de Watering, & Bots, 2023). To truly harness the potential of CBL, educational practices must be informed by rigorous research (Leijon et al., 2022). We can ensure that CBL experiences are effective and theoretically sound by bridging the gap between research and practice.

Researchers and practitioners should develop CBL experiences and articulate the principles underpinning the effects of those interventions in terms of theories of students' learning processes and gains (Doulougeri, van Beemt, et al., 2022). We recommend collaborating with an interdisciplinary team of policymakers, teachers, and researchers to develop sound educational research that accompanies the implementation of CBL educational innovations. Sound research fulfills a double objective: on one hand, it will lead to good educational practices, and on the other, it will lead to the advancement of our knowledge and theory about which aspects of CBL work and why (Vermunt, 2021). Hence, intensifying the educational research on CBL is not just helpful but is crucial for the evolution and success of CBL as an innovative pedagogy.

Considering this, CBL implementation should be accompanied by rigorous educational research and documentation of the research methods used (Leijon et al., 2022). For example, many studies did not explicitly clarify whether their student samples were truly representative of their respective university populations. Accurate documentation of the study population is paramount for ensuring the representativeness and validity of research findings. For transparency and rigorous scientific inquiry, future studies must provide precise details about their study populations (Pawley, 2017).

4.8.2 Research on the process/mechanisms of CBL

There needs to be more evidence about how CBL processes influence student learning. Most of these studies measured student learning in a course in general, rather than focusing on the effects of various educational components and learning activities on learning. This is especially problematic for studies combining two or more active learning activities, because it is difficult to assess each activity's unique contribution and effectiveness if they are not studied separately (Prince & Felder, 2006). Thus, the CBL research community needs to assess which CBL processes and mechanisms are effective and supportive of students' learning outcomes. To achieve this, practitioners should emphasize systematic documentation and analysis, and reflection on the design, development, evaluation, and implementation of CBL. Researchers should report clearly the research methods and analyses and use validated instruments to capture the effects of CBL and compare it with other approaches. In this way, educational policymakers can make evidence-informed decisions based on research about CBL practices. Experimental studies may be needed as well to achieve this.

4.8.3 Research on students' learning gains

In terms of learning outcomes or learning gains, despite the overall positive outcomes associated with adopting CBL on student learning and competence development, it is important to look critically at the methodologies reported by the authors for answering their research objectives and questions. In most studies, the measurement methods relied on students' perceptions of their learning through self-reports or analysis of student deliverables and grades. This can be problematic, because students' opinions of the course or any aspect of it can impact their estimates of their learning (Prince & Felder, 2006). The development of a number of competencies was investigated by using self-perceptions. The results showed that few CBL interventions applied control group designs. However, when no control groups are used, researchers may reach varying conclusions regarding whether the influence of an educational aspect is significant. In addition, many studies only measured the learning and competence of students after adopting CBL, without establishing a baseline. Pre and post tests could provide interesting insight into the development of or changes in student learning, but according to the findings of this review they were rarely used. Currently, a limited set of studies have compared CBL with other methodologies such lecture-based or project-based instruction (Berland et al., 2013). There is no supporting evidence as to whether CBL leads to better knowledge acquisition and application. Longitudinal studies are also essential to explore the cumulative benefit of CBL education rather than the short-term learning gains of a specific CBL course or project (Berland et al., 2013; Herzog et al., 2022).

4.8.4 Research on teachers' professional development

The role of teachers is central to CBL (Membrillo-Hernández et al., 2021; Pepin & Kock, 2021). Assisting educators in developing suitable learning materials for CBL offers only one necessary element in promoting reform. Professional development at a broader level is a critical component in scaling up CBL across the curriculum (Doulougeri et al., 2021). Future research should explore what competencies and skills teachers should have to meet the demands of their roles as tutors, coaches, and facilitators. This research will provide more information about the support teachers need to develop a CBL curriculum (cf. Stevens et al., 2024).