Introduction

Traditional didactic lecture is the most common method for medical education. It has the inherent advantage of being able to address and provide knowledge to a large group of students in a short period of time. The main drawback of the lecture method is that the audiences receive information passively. Therefore, it may impair the students’ opportunity of critical thinking and long-term knowledge retention.[1] To overcome this drawback, the medical colleges are in the midst of a paradigm shift–moving from a passive, teacher-centred and individual-based way of education, towards a more active, student-centred and group-based way of education.[2]

Team-based learning (TBL) is a teaching method that focuses on the needs of students. The program is directed by a subject specialist. It gives the students opportunities to apply their knowledge through a series of activities. These include individual work, team work, immediate feedback, and task-based problem-solving assignments.[3] TBL has been studied extensively by a great number of medical educators on different audiences (second-year medical students,[4] clerkships[2,5–7] and residents[7,8]), in both short-term and long-term,[6,9,10] and in relations to various educational components (such as attitude,[11] leadership and professionalism[12]). Improved academic outcomes and examination scores have been reported in histology,[13] anatomy[14] and psychiatry.[15] However, there are only limited researches that tried to integrate TBL into ophthalmology clerkship curricula.[16] In this study, we aimed to assess the academic impacts of TBL on ophthalmology clerkships and sought to investigate whether any other effects of TBL would emerge in that context.

Materials and Methods

Results

Class Performance

In the TBL classes, the GRAT scores were significantly higher than the IRAT scores, both overall and in the groups stratified by BOLs (Fig 1). These results confirmed that group problem-solving was more effective than individual problem-solving, regardless of BOLs. The ANOVA was used to compare examination scores across ophthalmology levels (Table 1). The IRAT, GRAT and application exercise scores did not differ significantly (P = 0.121, 0.175 and 0.485, respectively), whereas the final examination score revealed significant differences (P = 0.01). When the groups stratified by BOLs are compared with one another, the IRAT scores differed significantly between the first and fourth groups, but not between the second and third groups. By contrast, the third and second groups scored significantly higher than the fourth group with regard to FESs, which showed that TBL improved the performance of academically weaker–but not the weakest–students.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. The effect of group learning on students’ performance: comparison of the IRAT and GRAT scores according to BOLs of the most recent ophthalmology exam before the clerkship.

By the paired t-test. *The mean difference is significant at the 0.05 level. Individual Readiness Assurance Test (IRAT), Group Readiness Assurance Test (GRAT), Group Application Problem (GAP).

https://doi.org/10.1371/journal.pone.0154250.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. The impacts of team based learning (TBL) on students’ performance according to BOLs of the most recent ophthalmology exam before the clerkship.

https://doi.org/10.1371/journal.pone.0154250.t001

There were sex-specific differences in students’ test performances (Fig 2). Female students scored the same as male students on the IRAT; however, on the final exam, female students scored significantly higher than their male peers (P = 0.039; t-test).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Sex differences in test performance of team-based learning (TBL) students.

T-test for differences of means.*The mean difference is significant at the 0.05 level.

https://doi.org/10.1371/journal.pone.0154250.g002

Student Evaluation of TBL

Students were asked to rate their TBL experience at the end of the module (Table 2); 57.65% of students agreed that TBL helped their learning experience. The dissection lab, traditional lectures and textbooks were rated higher. TBL sessions were deemed more helpful than on-line material and educational computer programs by a considerable margin. In a focused evaluation of TBL, students believed TBL helped to assess their present knowledge, provided them with a higher level of knowledge and had a positive impact on their learning attitudes. In addition, 80% of students agreed that the TBL module was well organized and 56.47% thought TBL should be offered more frequently in the curriculum. However, some students felt that the amount of time given to complete the TBL assignment was inadequate in light of other required and on-going learning activities.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Students’ questionnaire responses to team-based learning (TBL) courses: comparisons between gender and quartile students (1 –strongly disagree, 6 –strongly agree)(85 students).

https://doi.org/10.1371/journal.pone.0154250.t002

Notably, there was a sex-specific difference in the evaluation profile. Male students tended to rate most items higher than female students (Table 2). The item “TBL challenged me to give my best” and “TBL had a positive impact on my learning attitudes” received significantly higher mean scores (P = 0.001 and P = 0.033, respectively) from male students than from female students. Male students also rated other general statements higher.

Discussion

This study assessed the educational effectiveness of TBL in the ophthalmology clerkship context at the Zhongshan Ophthalmic Centre of Sun Yat-sen University. Our results suggested that applying TBL had a positive impact on acquiring ophthalmology knowledge. By working together, students’ GRAT scores were significantly higher than their IRAT scores. Students in the middle academic quartiles (groups II and III) have improved their academic performance. Additionally, the students were satisfied with our TBL approach.

It has been generally accepted that with peer teaching and group learning methods such as TBL, the groups should outperform the individuals. The analysis of readiness assurance tests (Fig 1) indicated that the average group scores were 11% higher than the average individual scores. This result suggested that the interactions occurred in small groups are effective to some extents. During the discussion process, the students could deepen their understanding through communication and debate. The knowledge acquired from exploration is more stable and durable compared to that acquired from listening to teachers. However, we also found that the overall team score did not always surpass the score of the team’s best member. We thought this phenomenon is because that the students in different academic quartiles would share and discuss their ideas with each other, sometimes they might arrive at an incorrect answer which was supported by the majority. However, it could also be due to the fact that one week is too short to develop team accountability.

Previous studies [4,14,15,21] have shown that one of the major benefits of TBL was to improve the middle academic quartile. We evaluated the effect of TBL on students’ performance by dividing students into different groups based on their BOLs to examine the mechanisms underlying this benefit (Fig 1 and Table 1). The result showed that the FESs of the middle academic quartile (groups II, III) had been greatly improved from the IRATs. This might result from the enhancement of personal knowledge through interaction with team members and faculty and the advance assignment. However, there was no evidence showing that helping other students would influence the performance of the students in the highest quartile. In line with the previous study,[22] the final examination score of the lowest academic quartile was significantly lower than that of the other quartiles. We believe it is due to the limited implementation period being too short for the academically weakest students to develop effective team communication and team learning skills. Moreover, TBL is superior to the lecture-based learning only when students have reached a comparably high level of related ophthalmology theories; otherwise, it will be difficult for students to engage in effective discussion without understanding of the abstract ophthalmic concepts.[5]

Our study indicated that the female and male students responded differently to both educational format and knowledge tests (Fig 2). Females outscored males by only 0.7 on the IRAT, but scored 4.3 points higher on the final exam. The TBL seems to be more suitable to female students. Generally speaking, the students are highly satisfied with TBL and their improvements during the experimental period. However, males were significantly more satisfied with TBL than females. These results indicate that the male medical students have higher probability of over-estimating their learning outcomes compared to female students. A previous study indicated that despite performing equally as well as their male peers, female medical students reported decreased self-confidence and increased anxiety.[23] The female students may consistently under estimate themselves because they consider the negative effect of overestimation outweighing that of underestimation. Social norms regarding modesty may weigh more heavily the female students.[24]

The second goal of this investigation was to document students’ evaluations of TBL. These data are somewhat compromised in that only 85 out of 99 (86%) potential participants have completed the survey. The students’ evaluations (Table 2) indicated that the TBL is an effective learning process for ophthalmology. More than half of the students reported that the TBL helped them assess their present knowledge, achieve a higher level of knowledge, and reduce the amount of time required for self-study, in addition to having a positive impact on their learning attitudes. Overall, students were satisfied with the TBL module. They also commended that it should be used more frequently in medical education.

On the other hand, most students still considered traditional didactic instruction and textbooks to be more appropriate than the TBL module, because the majority of students typically used to study alone. The questionnaire revealed that only one-third of the students studied with colleagues frequently. Many students felt uncomfortable with team learning and had never discovered its value. These students might rate TBL based on how much information acquired from the course can actually be applied in further examinations. The scores in exams have been considered as the most important assessed outcome.[25] On the basis of these observations, we propose that students with who achieve higher performance are more prepared to overcome any negative implications of TBL and can use interactive learning successfully in mastering their ophthalmology studies.

This study has several limitations. First, we have modified TBL by replacing peer evaluation with faculty evaluation which was based on the student’s participation and performance during the module. While faculty evaluation may be more objective, it is likely to reduce accountability to the team not using peer evaluation. Second, the relatively small sample size limited the study’s statistical power and prevented statistically significant results. The response rate for the evaluations of the TBL was low (85 out of 99), which may have biased our results. Third, the one-week ophthalmology clerkship does not provide enough time for the students to develop high-functioning learning teams, which may reduce the effects of TBL. Because the final examination was conducted right after the TBL module, our results only revealed a relatively short-term impact on learning. Fourth, our study design did not allow us to compare the students’ performance among the different teaching techniques (i.e., PBL, TBL and classical case-based small group discussions). Our results are encouraging, but more outcome-centred studies of TBL are required to provide objective evidence of the effectiveness of this active learning strategy in medical education. Further study can explore the long-term effects of TBL strategies in the ophthalmology clerkship curriculum. Additionally, a prospective research design that compares the learning outcomes of academically similar student cohorts exposed to the TBL strategy versus another active learning method might produce meaningful data. We expect that satisfaction with team experience would increase year after year as the faculty and students become familiar with TBL and ultimately benefit from it.

In conclusion, the application of TBL to the ophthalmology clerkship curriculum improved students’ performance, particularly that of academically weaker students–and increased students’ engagement and satisfaction. We believe that with time and experience, TBL will prove to be an effective and highly rated innovative learning method in ophthalmology, leading to more faculty members to adopt it in their teaching and providing more active learning and deeper understanding for medical students.