Attendance of Near-Peer Tutoring Sessions Improves Academic Performance of First-Year Medical Students

Abstract

Introduction

Many medical schools have implemented near-peer tutoring (NPT) programs based on the substantial benefits of these programs to both tutors and tutees. This study correlated the frequency of attending weekly NPT sessions to students’ academic performance, and also compared performance of high and low attending students.

Methods

Twenty-one weekly NPT sessions were delivered by fourth-year (M4) students to first-year (M1) students. Attendance was recorded, and accordingly students were divided into three groups, high (15–21 sessions), moderate (7–14 sessions), and low (0–6 sessions) attendance groups. Pearson product-moment correlation coefficient was computed to assess the relationship between students’ frequency of attendance and their overall performance on overall basic medical sciences at the end of M1 year and M2 mid-year NBME® CBSE. Students’ performance was also analyzed using ANCOVA and Bonferroni post hoc test to compare differences between the three attendance groups.

Results

An increase in number of attended NPT weekly sessions corresponds with an increase in students’ performance on overall M1 basic sciences weighted average (r(196) = .240, p ≤ 0.001) and the M2 mid-year NBME® CBSE (r(196) = .325, p ≤ 0.001). High attending students performed significantly higher than low attending students on overall M1 weighted average of basic science courses (F(2, 192) = 8.518, p < .001) and M2 mid-year NBME® CBSE (F(2, 192) = 15.494, p < .001).

Conclusion

Attending NPT weekly sessions is a valuable experience that is associated with an improvement in M1 medical students’ academic performance. However, low performing students tend to miss attending NPT weekly sessions.

Introduction

Near-peer tutoring (NPT) is described as more advanced tutors facilitate and support the learning of tutees less advanced by at least 1 year [1, 2]. Near-peer tutoring/mentoring is a way of promoting professional and personal development, and could aid first-year medical students in the transition from undergraduate to medical school, and in maintaining their well-being [3].

Many medical schools implement NPT programs that are perceived to be beneficial for tutors and tutees [4, 5]. In the preclinical years, the NPT programs have been implemented to support the teaching of basic science courses [6–11]. Although some studies have reported that participation in NPT improves tutees’ academic performance [5], there are limited studies that assessed the impact of a weekly NPT program on first-year medical students’ overall academic performance on basic sciences.

Positive evaluation of NPT was observed in many studies [12–16]. Peer mentoring is perceived by first-year medical students to provide psychosocial and academic benefits, and regular involvement with a mentor was observed to be a consistent predictor of perceived academic and nonacademic benefits [15].

Tutoring programs were found to be valuable in learning and revision [12, 14], provide psychological and academic benefits [15], and help in teaching communication skills [17]. Participation in NPT was reported to improve tutees’ performance on basic sciences [11, 18, 19], biochemistry [10], anatomy [20, 21], and physiology [7, 12].

The University of South Carolina School of Medicine Greenville (USCSOMG) developed a NPT program with the overarching goal of improving students’ performance in the preclinical years and subsequently United States Medical Licensing Examination^® (USMLE) Step 1 examinations. Weekly NPT sessions were developed and implemented by fourth-year (M4) students to assess previous weeks instructional materials. The sessions employed practice retrieval strategies [22, 23] to engage tutees to actively participate in answering a number of multiple-choice questions.

The main goal of this study is to assess the impact of attending NPT weekly session on first-year (M1) medical students’ academic performance. The study correlated the frequency of attending weekly NPT sessions to students’ performance, and also compared academic performance of high attending and low attending M1 medical students.

Methods

Analysis of covariance and correlation methods were used to investigate the impact of weekly NPT sessions on improving M1 students’ performance on M1 overall basic sciences weighted average, and M2 mid-year National Board of Medical Examiners^® Comprehensive Basic Science Examination (NBME CBSE). The two methods were used to analyze the association between attending NPT sessions and students’ performance, and to compare the performance of the three attendance (high, moderate, low) groups.

Participants

A total of 208 M1 medical students participated in this study, classes that graduated in the spring of 2021 (n = 104) and 2022 (n = 104). All students were informed and encouraged to participate in the NPT program. Students’ demographic data and their pre-matriculation characteristics, in addition to students’ performance on M1 overall basic sciences, and NBME CBSE examinations were requested from the Office of Academic Affairs at USCSOMG. The study was reviewed and approved by the University of South Carolina Institutional Review Board (IRB).

Educational Context

The M1 integrated curriculum at the USCSOMG blends basic and clinical sciences. The foundational basic sciences (biochemistry, physiology, anatomical sciences, immunology) are taught in the following modules: Molecular and Cellular Foundations of Medicine, Structure and Function of the Human Body Part 1 & 2, Neuroscience, and Defenses & Responses. In addition to these modules, running throughout M1 year is the Integrated Practice of Medicine (IPM) module, which further integrates basic and clinical sciences and promotes clinical reasoning skills. The M2 year spans for almost 8 months where the mid-year NBME CBSE was administered halfway through the calendar year.

Description of Tutoring Sessions

A total of twenty-one weekly NPT sessions were offered once a week during all M1 modules. Tutors are M4 medical students who were self-selected and their applications were reviewed by the academic success program. Two students were responsible to deliver the weekly sessions in order to allow many tutors to participate in the program, and to eliminate the reliance on a limited number of the same tutors for delivering all the sessions. Students must have earned a B or higher to participate as tutors during basic science modules. The M4 tutors were directed to write their own multiple-choice questions in a USMLE Step 1 format that were relevant to the material covered during the specific week of the modules. Using TurningPoint software on PowerPoint presentations, tutees and tutors actively interact in discussing the answers of the projected questions. Based on tutees’ answers to the questions, tutors could then engage tutees to share their perspectives on the selection of their answer choices. Accordingly, the tutors actively engaged tutees during the sessions as they explain the answers to the questions and when answering additional students’ questions to promote understanding. They also shared their experience of effective study planning and strategies, identifying high-yield topics and strategies for how to answer questions. The tutoring sessions lasted for approximately 90 min or less and covered around 30 questions per session. At the end of the session, the tutors recorded students’ attendance, the number of questions covered, and list of topics discussed.

Data Collection

Students’ attendance was recorded for all the twenty-one NPT sessions. Students who attended 15–21/21 sessions were classified as high attendance, 7–14/21 sessions moderate attendance, and 0–6/21 low attendance. The performances of the three attendance groups were collected to be analyzed to determine if there is a significant difference between the performance of the three groups on end of year M1 basic medical sciences weighted averages as an internal measure. To improve the validity of the study, we added the performance of M2 mid-year NBME CBSE as an external measure. Although NBME CBSE was administered in the first half of M2 calendar year, it is still relevant since this examination assesses the basic science materials covered in the M1 year.

Data Analysis

Data were analyzed using IBM Statistical Package for the Social Sciences (SPSS) software (IBM Corporation, Armonk, NY, USA).

As the lengths of M1 modules varied, students’ performance in each module was weighted based on the duration of the module relative to the duration of the academic year. These weighted values of each module were then averaged to generate an overall weighted score for basic sciences at the end of the M1 year. Student averages on the Medical College Admission Test® (MCAT) and cumulative undergraduate grade point averages (uGPA) were used as a baseline for comparison of each attendance group.

Pearson product-moment correlation analyses were performed to measure the strength of association between the frequency of attendance and students’ performance on overall M1 basic sciences weighted average and M2 mid-year NBME CBSE performance. Students’ overall M1 basic sciences weighted average scores and students’ M2 mid-year NBME CBSE scores based on attendance levels (high, moderate, low) of tutoring sessions were analyzed using a one-way analysis of covariance (ANCOVA). NBME CBSE scores for the class of 2022 were provided on a LCME Step 1 score equivalent scale. Therefore, NBME CBSE old scale scores for the class of 2021 were converted to match the class of 2022 scores for this analysis. The ANCOVA corrected for students’ pre-matriculation MCAT and uGPA which were used as baseline performance measures (covariates). A Bonferroni post hoc test was used to compare the scores between the high, moderate, and low attendance groups. Students with incomplete scores were removed from the data analysis.

Results

Participants were 40% males and 60% females with an average age of 23 years. Students’ pre-matriculation characteristics included an average MCAT at the 71st percentile and mean uGPA of 3.72 on a 4-point scale.

Results of the Pearson correlation indicated that there was a significant positive correlation between the frequency of attendance and students’ overall M1 basic sciences weighted average (r(196) = .240, p ≤ 0.001) and the M2 mid-year NBME CBSE (r(196) = .325, p ≤ 0.001) (Table 1). That is, an increase in number of attended peer-tutoring sessions corresponds with an increase in students’ academic performance.

Table 1.

Pearson correlations between attendance and performance on M1 overall basic science and M2 mid-year NBME CBSE

	M1 overall basic science			M2 mid-year NBME CBSE
	(N = 196)			(N = 196)
	Pearson correlation (r)	Score variation (r2)	p value	Pearson correlation (r)	Score variation (r2)	p value
Attendance	0.240	5.76%	< 0.001	0.325	10.56%	< 0.001

Correlation is significant at p < 0.01 level (2-tailed)

Descriptive statistics and comparison of students’ academic performance by the level of attendance (low, moderate, and high attendance) is summarized in Table 2. The comparison of students’ academic performance on overall M1 basic sciences weighted average showed statistically significant difference between the three attendance groups (high, moderate, low) (F (2, 192) = 8.518, p < .001) after controlling for MCAT and uGPA (Table 2). Bonferroni post hoc analysis revealed that high attendance students performed significantly higher than low attendance students (p ≤ .001) (Table 3). The comparison of students’ academic performance on the M2 mid-year NBME CBSE also showed statistically significant difference between the three attendance groups (high, moderate, low) (F (2, 192) = 15.494, p < .001) after controlling for MCAT and uGPA (Table 2). Bonferroni post hoc analysis revealed that high attendance students performed significantly higher than low attendance students (p ≤ .001) and moderate attendance students (p = .001) (Table 3). To further illustrate the differences between the three attendance groups, the percentage of the top quartile for each attendance group was calculated for the performance results on both the overall basic science average and NBME CBSE. There is an increase in the percentage of the top quartile with a higher number of sessions attendance (Fig. 1).

Table 2.

Descriptive statistics and comparison of students’ performance by the level of attendance (low, moderate, and high attendance) on the performance parameters

Variables	High attendance		Moderate attendance		Low attendance
	M (SD)	N	M (SD)	N	M (SD)	N	F	p
M1 basic science	90.06 (4.28)	45	88.00 (5.17)	56	86.37 (5.31)	96	8.518	< 0.001
M2 mid-year CBSE	185.78 (18.24)	45	172.68 (19.30)	56	167.09 (21.36)	96	15.494	< 0.001

The mean difference is significant at 0.05 level

Table 3.

Bonferroni post hoc pairwise comparison of the mean difference between the three attendance groups of NPT sessions

	NPT attendance	Mean difference	Standard error	Significance
M1 basic science	High vs low	3.45	0.84	< 0.001
	High vs moderate	2.16	0.93	0.063
	Moderate vs low	1.30	0.78	0.297
M2 mid-year CBSE	High vs low	18.58	3.34	< 0.001
	High vs moderate	13.30	3.69	0.001
	Moderate vs low	5.28	3.12	0.276

The mean difference is significant at 0.05 level

Fig. 1.

Relationship between attending NPT sessions and academic performance quantifying the percentage of the top quartile for each attendance group

Discussion

Our results show that students who frequently attended NPT sessions that utilized retrieval practice, i.e., testing effect strategy, have improved their academic performance on our internal basic sciences examinations during the first year of medical school. Similarly, high attending students of NPT sessions performed better than low attending students on the NBME CBSE administered mid-way through the second year of basic sciences coursework. These findings are consistent with data published on the positive effects of retrieval-based learning. Retrieval-based learning helps direct future studies [23] and improves learning [22]. Retrieval-based learning is a technique that has reportedly been used by high-performing learners along with spaced repetition [24].

Our findings are consistent with previous reports showing students who participate in NPT or peer-led tutoring programs improve their performance on course grades and yearly weighted averages [5, 7, 25–28]. The improved performance was shown to occur through different variations of peer-led tutoring and different lengths of peer-led tutoring programs. The current study was focused specifically on first-year cumulative performance on basic science coursework on internally developed assessments and on an external measure, i.e., NBME CBSE, that is used to determine preparedness for student performance on the USMLE Step 1. In our program, the tutoring sessions were offered weekly throughout the first year with an emphasis on retrieval-based learning. Swindle and Wimsatt [26] reported changes in performance after 3 to 4 months and 8 to 9 months for students that had at least 1 failing test grade. Interestingly, they also reported no improvement in grades for students that were in good academic standing. Opposite to our findings, Batchelder et al. [29] reported no change in performance between students attending peer-led tutoring and students not attending peer-led tutoring; however, they reported that students attending the sessions felt more prepared for exams compared to students not attending the sessions. Although the scores did not change with peer tutoring [29], the perception of feeling more prepared for exams could lower student stress which was reported by Swindle and Wimsatt [26]. In addition, the feeling of being more prepared would potentially allow the student to feel as though they have the ability to perform at their highest level.

Previous studies [28, 30] showed that performances on M1 basic science weighted average and M2 mid-year NBME CBSE were strongly correlated with USMLE Step 1 performance. These results are consistent with previous findings, which show performances on overall basic sciences [31, 32] and NBME CBSE [33, 34] were correlated with USMLE Step 1 performance. Since our study showed that attending NPT sessions improves academic performance in basic sciences, NPT programs could be used to prepare students for taking USMLE Step 1 examination.

There are conflicting reports concerning student attendance in NPT sessions and performance on the USMLE Step 1. One study reported USMLE Step 1 scores were similar between students that did attend and students that did not attend an NPT program [35] while others have reported improved Step 1 scores for students that attend NPT sessions compared to students that do not attend NPT sessions [8, 28].

The limitation of this study is it demonstrates the experience of only one medical school, which could affect the generalizability of the results. In addition, the low number of the high attendance group compared to the high number of low attendance group could negatively affect the overall statistical power of the results. Additionally, the collected data of this study was not sufficient to explain the reasons most low performing students do not attend the NPT sessions. More research is needed regarding the reason why low performing students do not attend NPT sessions, although participation would improve their academic performance.

Conclusions

Attending NPT sessions has improved academic performance of first-year medical students. It is significantly correlated with students’ performance on basic medical sciences examinations, in which high attending students consistently outperform low attending students. The results are testaments that NPT program is an important component of academic success programs. This information suggests that NPT programs could be implemented early within medical school curricula to improve students’ performance.

Author Contribution

All authors contributed to the conception and design of the work, data analysis, drafting or revising the article, have agreed on the journal to which the article will be submitted, gave final approval of the version to be published, and agreed to be accountable for all aspects of the work.

Declarations

Ethics Approval and Consent to Participate

This study received exemption from Human Research Subjects on 2/4/2019 by the University of South Carolina institutional review board. The research was exempted in accordance with 45 CFR 46.104(d)(2). The research is compliant with the Declaration of Helsinki. Student survey data was gathered voluntarily and anonymously. The de-identified data was housed in a password protected electronic file. The reference number is Pro00086173.

Conflict of Interest

The authors declare no competing interests.