International Open Access Journal Platform

1 Introduction

Project-Based Learning (PBL) is a learner-centered dynamic teaching model that emphasizes integrating multidisciplinary knowledge and skills through solving realistic and open-ended problems to develop critical thinking, collaborative awareness, problem-solving skills, interpersonal communication, media literacy and other interdisciplinary practical competencies (Maros et al., 2021). As a flexible methodology integrating multiple practical skills, PBL has been investigated in a great number of studies on a global scale over the last decade. In language instruction, PBL plays an essential role in education in general and L2 teaching in particular, especially in the era of education reform calling for cultivating 21st-century skills. It not only stimulates students’ curiosity, creativity, and agency but also equips them to adapt to global economic shifts in their ensuing professional lives (Hadakova & Papcunova, 2019). Meanwhile, with the digital transformation in the field of education, the rapid iteration of media technologies, including artificial intelligence, big data and the metaverse, poses unprecedented challenges to talent training, which underscores the necessity of innovative teaching revolutions. Although many studies have proven its medium to large positive effects on students’ L2 achievements compared with teacher-centered traditional skill-based instruction, research findings on its effectiveness and influencing factors are still inconclusive due to the complexity and diversity of PBL, internal factors (e.g. individual differences such as affective attitudes and varying L2 initial proficiency), and external factors (e.g. time duration, class monitoring level, group and class size) (Zhang & Ma, 2023). Based on this, this quasi-experimental comparison aims to generate rigorous empirical evidence regarding the instructional impacts of PBL. The results can offer clear instructional references for teachers to implement PBL in daily college English classrooms.

2 Literature Review

2.1 Project-Based Learning

Project-based learning has been widely implemented in L2 instruction for over two decades. In general, PBL involves active learning, systematic teaching, product-oriented teaching and skill-based assessments based on Piaget’s constructivism (Greeno, 2006). Constructivist theory advocates that meaning is constructed in the social context rather than passive knowledge acceptance and teacher-centered knowledge transmission. Learner agency, autonomy and self-efficacy are highlighted under the framework of constructive theory. Another important aspect of the effect of PBL on the learning process is on affective attitudes. According to the results of Zhang Ying’s intrinsic motivation scale, which was administered to 21 private university students before and after receiving project-based learning, there were significant differences in students’ interest, boredom, autonomy, and competence before and after, which positively influenced students’ intrinsic motivation to learn (Zhang, 2022).

However, contradictory results are shown in other studies, maintaining that project-based learning has the same effect or even some negative effects compared to traditional instruction. Xu and Li (2019) selected three classes constituted by non-English major undergraduate students for the role performance of one excerpt of the New Vision College English textbook, and found that the overall effect of PBL was not outstanding in the classes with a weaker foundation. Gratchev and Jeng (2018) explored whether project-based learning activities could improve students’ learning experiences. Their three-year longitudinal dataset indicated no substantial gaps in achievement between the two groups, and the findings indicated that students were less motivated to accept new learning methods such as PBL. Karaçalli and Korur (2014) used a quasi-experimental design to teach the experimental group by using a project-based learning approach, and the results showed no statistically significant effects on students’ attitudes toward learning across groups.

Possible reasons for the inconsistent findings on the effectiveness of PBL can be attributed as follows: 1) The initial proficiency of participants varies. Current empirical studies on project-based learning mainly pay attention to primary and secondary schools, with less research on the impact on higher education. Those studies that support the significantly positive effects of PBL in higher education mostly focus on relatively high-level L2 students or professional English major students, while neglecting the moderate and low proficiency ones, making conclusions lack generalizability; 2) The effects of project-based learning are easily impacted by various moderating variables, including disciplinary background, academic period, group size, class size, experiment period, and complexity of tasks. Zhang and Ma (2023) used meta-analysis to conclude that PBL has a more obvious effect in engineering and technology disciplines and is more suitable for small-class teaching with a group size of 4-5 students in 9-18 weeks; 3) PBL, as a complex task-driven method, brings huge challenges to students beyond their capabilities. Nakamura et al. (2021) affirmed that activity mismatch, task difficulty and insufficient L2 proficiency can result in the emergence of boredom. Prior studies clearly authenticated that foreign language learning boredom (quantified by the Foreign Language Learning Boredom, FLLBS) exerts debilitating impacts on learners’ motivation, engagement, cognition, strategy use, behavior, and flow experience in class, which further hinder L2 achievement and proficiency (Dewaele et al., 2023).

2.2 The Present Study

Although PBL has been widely approved in second language teaching, research findings on its effectiveness are still inconclusive, especially in higher education. Furthermore, the impact on FLLBS and L2 achievements among learners with varying L2 initial proficiency still lacks empirical support. If we cannot develop a comprehensive view of appropriate conditions and suitable scenarios for applying PBL, this laborious and time-consuming teaching model will be an enormous consumption of teaching resources. Based on this, this study aims to provide more accurate and reliable evidence in higher education by assessing the effects of project-based learning under the guidance of a comparative empirical research, fill the existing research gap and deliver actionable pedagogical suggestions for PBL implementation so that educators can effectively apply PBL to real class teaching.

Based on the research objective, we raised the following two research questions (RQs) in order to fill the gap:

RQ1: Can PBL tasks significantly reduce EFL learners’ FLLBS and improve their L2 achievements compared to traditional skill-specific training?

RQ2: Does the effectiveness of PBL on reducing FLLBS and improving L2 achievements differ among EFL learners with varying initial language proficiency?

As regards the above research questions, we proposed the following hypotheses (Hs):

H1: PBL tasks would significantly reduce learners’ FLLBS and improve their L2 achievement compared with traditional skill-specific training.

H2: Different initial levels of learners’ L2 proficiency would affect the effectiveness of PBL during the learning process. High-proficiency group learners would exhibit lower FLLBS than low-proficiency group learners after the PBL programs are completed.

3 Methodology

3.1 Participants

The study was conducted at a university in Zhejiang, where 80 non-English freshmen who had just passed the College Entrance Examination were selected. All participants came from non-English majors to ensure the maximum effects of PBL tasks compared with traditional teacher-led training. 80 first-year college students were divided into two groups based on their National College Entrance Examination (NCEE) English scores.

The high proficiency group comprises 40 students whose NCEE scores all exceed 135. Half of them are randomly assigned to take PBL task training, and the remaining receive traditional skill-specific training.

The low proficiency group comprises 40 students whose NCEE range from 90 to 100. Similar to the high proficiency group, half of them receive PBL task training, while the remaining participants receive traditional skill-specific training.

3.2 Measure Instruments and Research Design

The two RQs were answered in a comparison study, which lasted 10 weeks from mid-September to late November 2025, before College English Test Band 4 (CET4). The post-test CET4 examination was arranged in mid-December 2025, which allowed a complete 10-week teaching intervention. The whole research cycle strictly followed the university’s autumn semester teaching schedule without cutting the designed intervention duration. After the comparative training was completed, all 80 participants were required to complete an FLLBS to measure their levels of boredom, and later entered the CET4 to test L2 achievements. The measure instrument we chose is a 3-item shortened version of the FLLBS adapted from the Achievement Emotions Questionnaire, which can effectively assess students’ academic boredom in the English subject, a criterion variable of FLLBS (Li et al., 2023). As a nationally standardized language proficiency test, CET4 boasts well-established reliability and validity. Thus, these instruments were taken into account in this comparison study to ensure the reliability and validity of experiments.

Teaching contents and requirements for the PBL group and the traditional skill-specific group were different. The PBL group comprises 20 freshmen with high initial L2 proficiency and 20 low initial proficiency freshmen (high proficiency PBL groups and low proficiency PBL groups). Group members were required to collaboratively produce news reports on social issues as a biweekly group production (video ≤ 20 minutes, text ≤ 2,000 words) involving roles such as narrators, interviewers, writers and one representative who showed the final product via PPT in class next week. The teacher’s role shifted from a dominant knowledge transmitter to a facilitator and class monitor where PBL students received tutor assessment and peer assessment to cultivate their cooperative awareness, social interactions and comprehensive skills with role rotations every two weeks. The topics of news reports included, but were not limited to, ecology, education, employment and technology, based on the syllabus of the New Vision College Public English textbook. By the end of PBL training, each PBL group was required to complete and submit five news reports by the end of the intervention and submit them to their teacher. All experimental and traditional classes were taught by the same certified college English instructor with 6 years of teaching experience, eliminating teacher style as a confounding variable. Weekly teaching hours, class schedule and textbook syllabus were fully synchronized across all four subgroups. The time span covered 10 weeks from mid-September to late November before CET4.

Compared with the PBL group, 40 traditional group members received single-skill teacher-oriented training following textbook order (listening-reading-speaking-writing) with no cross-skill integration tasks, synchronizing with the PBL group teaching. After 10-week comparison studies, PBL group members and traditional skill-specific group members were asked to participate together in the completion of the 3-item shortened version of the FLLBS, a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). All scale data were calculated as item average scores ranging from 1 to 5, in which higher scores indicate a higher level of FLLBS. The 3-item shortened version of the FLLBS yielded satisfactory internal consistency in this sample, Cronbach’s α = 0.826. For CET4, the national standardized test has well-established reliability and validity. All participants took part in the CET4 test (2025.12) to collect their CET4 scores.

3.3 Data Analysis

The experimental data were collected by SPSS 27 to explore whether there are significant differences between different groups. To address RQ1, an independent samples t-test was conducted to compare CET4 scores and FLLBS total scores between the PBL group and the traditional skill-specific group to find whether PBL tasks can significantly reduce EFL learners’ FLLBS and improve their L2 achievements. For RQ2, independent samples t-tests were used to compare FLLBS scores between different proficiency PBL groups to test whether participants’ initial proficiency could affect anti-boredom effects and effectiveness on improving L2 achievements of PBL. To further explore how initial proficiency affects their L2 achievements in PBL pedagogy, path analysis was supplemented to establish relationships between L2 initial proficiency, FLLBS and L2 achievements in order to reveal the direct and indirect effects between three variables in a constructed model.

4 Results

4.1 Differences in FLLBS and L2 Achievements Across PBL and Traditional Skill-Specific Groups

The result of the independent samples t-test (Table 1) shows that students in the PBL group were significantly different from those in the traditional group on FLLBS scores (t (df = 78) = 4.32, p < 0.05) and CET4 scores (t (df = 78) =
2.23, p < 0.05). A comparison of group means revealed that the average FLLBS score of students in the PBL group (M =
3.15) is significantly lower than that of the traditional group (M = 3.65). The difference between the means is -0.50 points on a 5-point test. Meanwhile, the average CET-score of students in the PBL group (M = 502) is significantly higher than that of the traditional group (M = 463). The difference between the means is 39 points. The results above confirm the hypothesis that PBL tasks can significantly reduce EFL learners’ FLLBS and improve their L2 achievements compared to traditional skill-specific training.

Table 1 Comparison of the Effectiveness of PBL on Students’ FLLBS and L2 Achievements

*p < 0.05

4.2 Effectiveness of PBL on FLLBS and L2 Achievements Among Students With Varying Initial Levels

The result of the independent samples t-test (Table 2) shows that students with high proficiency in the PBL group were significantly different from those in the traditional skill-specific group on FLLBS scores (t (df = 38) = -2.21, p < 0.05). A comparison of group means revealed that the average FLLBS score of students with high proficiency in the PBL group (M = 2.73) is significantly lower than that of students in the traditional skill-specific group (M = 3.51). The difference between the means is -0.78 points on a 5-point test. The results support the anti-boredom effects of PBL on students with high proficiency.

Table 2 Comparison of Students’ FLLBS Scores Between the High Proficiency PBL Group and the High Proficiency Traditional Skill-Specific Group

*p < 0.05

The result of the independent samples t-test (Table 3) shows that students with low proficiency in the PBL group were similar to those in the traditional skill-specific group on FLLBS scores (t (df = 38) = -1.34, p > 0.05), which indicates that the anti-boredom effects of PBL are not conspicuous on low proficiency students.

Table 3 Comparison of Students’ FLLBS Scores Between the Low Proficiency PBL Group and the Low Proficiency Traditional Skill-Specific Group

p > 0.05

The result of the independent samples t-test (Table 4) shows that students with high proficiency in the PBL group were significantly different from those in the traditional skill-specific group on CET scores (t (df = 38) = 3.17, p < 0.05). A comparison of group means revealed that the average CET4 score of students in the PBL group (M = 551) is significantly higher than that of students in the traditional skill-specific group (M = 492). The difference between the means is 59 points. The results support the effectiveness of PBL to improve L2 achievements in high-proficiency students.

Table 4 Comparison of Students’ CET4 Scores Between the High Proficiency PBL Group and the High Proficiency Traditional Skill-Specific Group

*p < 0.05

The result of the independent samples t-test (Table 5) shows that students with low proficiency in the PBL group were similar to those in the traditional skill-specific group on CET scores (t (df = 38) = 1.67, p > 0.05), which indicates that the effectiveness of PBL to improve L2 achievements is not obvious for low-proficiency students.

Table 5 Comparison of Students’ CET4 Scores Between the Low Proficiency PBL Group and the Low Proficiency Traditional Skill-Specific Group

p > 0.05

4.3 Path Analysis Between L2 Initial Proficiency (Before PBL), FLLBS (During PBL) and L2 Achievements (After PBL)

Through previous statistical data, we found that students’ L2 initial proficiency can affect the effectiveness of PBL on students’ L2 achievements. However, we still cannot conclude that this influence is direct, because the role of FLLBS is unclear among them. To further explore the relationships between students’ L2 initial proficiency, FLLBS and L2 Achievements, we adopted the path analysis to test if FLLBS has a mediation effect between students’ L2 initial proficiency and L2 Achievements. The results of path analysis are presented in Figure 1.

Figure 1 Path Analysis Between L2 Initial Proficiency, FLLBS and L2 Achievements in the Predictive Model

According to Figure 1, the standardized path coefficient (β) quantifies how initial L2 proficiency and FLLBS predict learners’ L2 achievement. Among them, FLLBS has a significantly negative predictive effect on students’ L2 achievements (β = -0.52, p < 0.05), while students’ L2 initial proficiency has a non-significant direct effect on their L2 achievements (β= 0.12, p > 0.05). Model fit indices met acceptable standards: χ²/df = 1.42, RMSEA = 0.041, CFI = 0.963, TLI = 0.958. Bootstrap sampling (5000 samples) was adopted to verify the significance of the indirect mediating effect. Thus, we confirm the mediation effect of the FLLBS and conclude that through the mediation effect of FLLBS, students’ L2 initial proficiency has a significant indirect effect on their L2 achievements.

Higher proficiency students often display higher motivation, confidence, agency and less FLLBS, while lower proficiency students often adopt avoidance learning strategies (e.g., in-class disengagement behaviors including dozing off, engaging in irrelevant small talk, browsing irrelevant online content, and staying silent in class), leading to disengagement in complex PBL tasks. The complexity of PBL tasks beyond their capability turns into the antecedent of learning boredom, and thus, the results of the limited effectiveness of PBL pedagogy in this context are consistent with theoretical expectations.

5 Discussion

This study explores the effectiveness of PBL on regulating FLLBS and improving L2 achievements, particularly among students with varying initial L2 proficiency, through a comparative experiment. The results support Hypothesis 1
(PBL significantly reduces FLLBS and improves L2 achievements overall) and Hypothesis 2 (effects of PBL vary by initial proficiency), though further analysis is needed in light of existing research and experimental design.

First, the PBL group showed significantly lower FLLBS scores (M=3.15) and higher CET4 scores (M=502) compared to the traditional teacher-centered group (M=3.65 for FLLBS; M=463 for CET4). These findings align with prior studies highlighting the role of PBL in enhancing motivation and integrating skills (Zhang, 2022). By driving collaborative tasks, role rotation, and output-oriented projects, PBL likely increases students’ interests, engagement and agency, thereby reducing boredom and improving L2 performances.

Second, initial proficiency moderated the effects of PBL. High-proficiency PBL students had significantly lower FLLBS (M=2.73) and higher CET4 scores (M=551), while low-proficiency PBL students showed no significant differences in FLLBS (M=3.56) or CET4 scores (M=452). This divergence may stem from high-proficiency students’ stronger language foundations and learning strategies, enabling them to navigate PBL tasks with greater confidence and satisfaction. In contrast, low-proficiency students, constrained by weaker language abilities, may have felt overwhelmed by task complexity, leading to persistent boredom (Nakamura et al., 2021; Dewaele et al., 2023).

Path analysis confirmed the role of FLLBS as a mediator. Initial proficiency indirectly influences L2 achievement by mitigating boredom, rather than exerting a direct effect on performance. High-proficiency students’ lower FLLBS thus indirectly boosted their achievements, whereas low-proficiency students’ unchanged FLLBS limited further gains. This study also clarified discrepancies in prior conflicting research (e.g., Xu & Li, 2019; Gratchev & Jeng, 2018), which may have averaged results across proficiency levels, causing subgroup differences. By stratifying participants, this study revealed PBL’s heterogeneous effects under different teaching circumstances, offering new insights into its applicability.

6 Conclusion

The results show that: 1) Compared to traditional skill-specific training, PBL significantly reduces FLLBS and improves L2 achievements overall, validating PBL’s effectiveness in language teaching; 2) Effectiveness of PBL varies by initial L2 proficiency. High-proficiency students benefit more from reducing boredom and improving achievements, while low-proficiency learners yield no statistically significant gains; 3) FLLBS mediates the relationship between initial proficiency and L2 achievements. Initial L2 proficiency indirectly affects achievements by reducing boredom, rather than directly influencing their performance and achievements. The major findings suggest that the learning effects of PBL are not invariably positive, but are more or less influenced by various moderating variables. Combining the effects of different variables is crucial for PBL in order to achieve the optimal effect size.

The study sheds light on PBL pedagogy and provides implications for practical L2 teaching in higher education. Educators should employ appropriate teaching methods that allow for students’ initial proficiency. For high-proficiency students, complex and interdisciplinary tasks, such as in-depth social issue reports, are effective in maximizing skill development. For low-proficiency students, simplified tasks aligned with their current levels are supposed to be provided in order to avoid boredom from task overload based on Vygotsky’s zone of proximal development (ZPD) and Krashen’s “i+1” hypothesis. Institutions should promote PBL as a complement to traditional teaching by assessing initial proficiency and designing differentiated tasks.

Several limitations still exist in this study. First, moderate L2 initial proficiency participants are not included in the research. The effectiveness of PBL tasks on students with moderate levels has not been validated. The results may vary when high, moderate and low proficiency students are compared together. Second, the sample size can be enlarged to include diverse grades, majors, and language backgrounds to validate the generalizability of the conclusion. Third, other measurement instruments, such as students’ reflection journals and focus group interviews, can be supplemented to determine the benefits of PBL on students’ diverse learning outcomes in future directions.

References

[1] Dewaele, J. M., Botes, E., & Greiff, S. (2023). Sources and effects of foreign language enjoyment, anxiety, and boredom: A structural equation modeling approach. Studies in Second Language Acquisition, 45(2), 461–479.

[2] Gratchev, I., & Jeng, D. S. (2018). Introducing a project-based assignment in a traditionally taught engineering course. European Journal of Engineering Education, 43(5), 788–799.

[3] Greeno, J. G. (2006). Learning in activity. In R. K. Sawyer (Ed.), The Cambridge handbook of the learning sciences (p. 79–96). Cambridge University Press.

[4] Hudáková, J., & Papcunová, V. (2019). Project training company in the Slovak Republic. 4th International Conference on Education Science and Development, Shenzhen, China, 21–34.

[5] Karaçalli, S., & Korur, F. (2014). The effects of project-based learning on students’ academic achievement, attitude, and retention of knowledge: The subject of “Electricity in our Lives”. School Science and Mathematics, 114(5), 224–235.

[6] Li, C., Dewaele, J. M., & Hu, Y. (2023). Foreign language learning boredom: Conceptualization and measurement. Applied Linguistics Review, 14(2), 223–249.

[7] Maros, M., Korenkova, M., Fila, M., Levicky, M., & Schoberova, M. (2021). Project-based learning and its effectiveness: Evidence from Slovakia. Interactive Learning Environments, 43(6), 1–9.

[8] Nakamura, S., Darasawang, P., & Reinders, H. (2021). The antecedents of boredom in L2 classroom learning. System, 98, 1–15.

[9] Xu, Y., & Li, X. Y. (2019). A brief analysis of the application of project-based learning in college English teaching from the constructivist perspective. Science Education Article Collects, 10, 168–169.

[10] Zhang, L., & Ma, Y. (2023). A study of the impact of project-based learning on student learning effects: A meta-analysis study. Frontiers in Psychology, 45(7), 1–14.

[11] Zhang, Y. (2022). Research on the influence of project-based teaching on private college students’ intrinsic motivation in English learning. Science and Education Review, 13, 29–31.