Small group effectiveness in a Caribbean medical school’s problem-based learning sessions

Purpose: The Tutorial Group Effectiveness Instrument was developed to provide objective information on the effectiveness of small groups. Student perception of small group effectiveness during the problem base learning (PBL) process has not been previously studied in Xavier University School of Medicine (Aruba, Kingdom of the Netherlands); hence, the present study was carried out. Methods: The study was conducted among second and third semester undergraduate medical students during the last week of September 2013, at Xavier University School of Medicine of the Netherlands. Students were informed about the objectives of the study and invited to participate after obtaining written, informed consent. Demographic information like gender, age, nationality, and whether the respondent had been exposed to PBL before joining the institution was noted. Student perception about small group effectiveness was studied by noting their degree of agreement with a set of 19 statements using a Likert-type scale. Results: Thirty-four of the 37 (91.9%) second and third semester medical students participated in the study. The mean cognitive score was 3.76 while the mean motivational and de-motivational scores were 3.65 and 2.51, respectively. The median cognitive category score was 27 (maximum score 35) while the motivation score was 26 (maximum score 35) and the de-motivational score was 12 (maximum score25). There was no significant difference in scores according to respondents’ demographic characteristics. Conclusion: Student perception about small group effectiveness was positive. Since most medical schools worldwide already have or are introducing PBL as a learning modality, the Tutorial Group Effectiveness Instrument can provide valuable information about small group functioning during PBL sessions.


INTRODUCTION
Group work plays an important role in problem base learn ing (PBL) and is critical to its success [1]. Ensuring small groups are effective and functioning is critical to the success of a PBL program. Interactions within the small group provide students with opportunities to ask questions, receive explanations, and discuss disagreements, which can lead to a deeper understand ing of the subject matter [2]. Group learning environments like PBL also have to deal with dysfunctional groups and ritu al behavior where students pretend to be actively involved in group work while in reality they are not so engaged [3]. Dol mans et al. [4] found a linear relationship between the success of the small group and several motivational and cognitive di mensions. Students' backgrounds and cultures can also influ ence both the motivational and cognitive dimensions and the success of the small group [5]. Attention should also be focus ed towards students who respond negatively to group work as they may be an important contributing factor to group dys function [6].
The Tutorial Group Effectiveness Instrument (TGEI) was developed to provide objective information on the effective ness of small groups in the tutorial process [1]. TGEI  to provide valid and reliable factor scores. According to the authors, their PBL groups were mixed and well balanced and their findings may be generalizable towards other PBL curri cula. The instrument was developed based on the motivation al and cognitive perspectives involved in small group PBL [7] and some items in the instrument were modified from the study by Dolmans et al. [3]. Student perception of the effec tiveness of small groups during the PBL process has not been previously studied at Xavier University School of Medicine (XUSOM), Aruba, Kingdom of the Netherlands. Hence the present study was carried out to obtain information about student perception of PBL small group effectiveness at XU SOM.

PBL program at Xavier University School of Medicine
XUSOM is a private medical school. Students in the under graduate medical (MD) course are mainly from the United States (US) and Canada [8]. In the summer 2013 semester, which commenced on the first Monday of May, the school shifted to an integrated organ systembased curriculum. The first five semesters of the MD program are in Aruba and stu dents complete their clinical rotations in the US. Normal hu man structure and function are taught during the first two se mesters, abnormal human structure and function during se mesters 3 and 4, followed by a hybrid curriculum. During the MD 5 semester students mainly prepare for the step 1 United States Medical Licensing Exam (USMLE), take various exams, and participate in live online lectures. They also prepare for the clinical semesters and further develop their clinical skills through interaction with standardized patients. The major learning methodology during semesters 1 to 4 is didactic lec tures, though recently faculty have been motivated to make the sessions more interactive through training workshops. PBL sessions are conducted for approximately twelve weeks during the fifteenweek semester. The first session concen trates on identifying the learning objectives and analyzing and working on the problem, while during the second session each group makes a presentation according to the learning objec tives they established in the first session. The facilitator evalu ates students and provides feedback to further improve the group dynamics. Most topics covered during the PBL session have already been covered during the didactic lectures but at least two topics are new. Six disease conditions are covered during each semester for the different cohorts of students. The PBL sessions primarily focus on learning objectives from anat omy, physiology, biochemistry, history taking, and social is sues during the first two semesters and, during the next two, on pathology, microbiology, pharmacology, clinical examina tion skills, social issues, and diagnosis. Student performance during the small group sessions is assessed using a standard ized instrument. The small groups and the faculty facilitator are kept constant over all four semesters.

Subjects
The study was conducted among second and third semester undergraduate medical students at XUSOM during the last week of September 2013. The study was approved by the Insti tutional Review Board vide notification number XUSOM/ IRB/2013/08. The students were explained the objectives of the study and invited to participate. Written permission was obtained from Dr. Singaram, the first author of the instrument [1]. Written informed consent was obtained from all partici pants.

Study design
Student perception of small group effectiveness during the PBL sessions was assessed using the TGEI developed by Sing aram et al [6]. By the time the instrument was administered, respondents had completed six PBL sessions. The question naire used is shown in the Appendix. Demographic informa tion like gender, age, nationality, and whether the respondent had been exposed to PBL before joining XUSOM was noted. Student perception about small group effectiveness was stud ied by noting their degree of agreement with a set of 19 state ments using a Likerttype scale ranging from 1 (strongly dis agree) to 5 (strongly agree). Student perception of overall group productivity was scored from 1 (insufficient) to 5 (excellent). The cognitive aspects were calculated by adding the scores of statements 1 to 7, motivational aspects by summing the scores of statements 8 to 14 and demotivational aspects by adding the scores of statements 15 to 19. Free text comments about PBL were invited from the respondents and common ones were tabulated.

Statistical analysis
The median scores were calculated and compared among different subgroups of respondents using appropriate nonpara metric tests. The MannWhitney U test was used to compare medians where the respondents could be divided into two sub groups (for example, by gender) while the KruskalWallis test was used where the respondents could be divided into more than two categories (for example, by nationality). A Pvalue of less than 0.05 was taken as statistically significant.

RESULTS
Thirtyfour of the 37 (91.9%) second and third semester stu dents participated in the study. The final numbers did not reach http://jeehp.org 34 since some respondents did not provide complete informa tion about their demographic characteristics. More respon dents were male, of American nationality, and had not been exposed to PBL before joining the institution ( Table 1). The median score for statements one to ten, statements 12 to 15, and statement 20 (overall rating of group productivity) was four. The median score for statement 11 (The small group stim ulated my selfstudy activities) was 3.5 while the score for state ments 16 to 19, which dealt with demotivational aspects was 2. The mean cognitive score was 3.76 while the mean motiva tional and demotivational scores were 3.65 and 2.51, respec tively.
The median cognitive category score was 27 (maximum score 35) while the motivation score was 26 (maximum score 35) and the demotivational score was 12 (maximum score 25). Table 2 shows the cognitive category scores according to re spondents' personal characteristics. There was no significant difference in scores according to respondents' demographic characteristics. Tables 3 and 4 show the motivational and de motivational scores according to respondents' characteristics. The difference in median scores was not statistically signifi cant. Analyzing the individual statements contributing to the median score, the score for statement 15 'During the course of the PBL some group members contributed less to the group discussion' was 4 while the median scores for the other state ments (16 to 19) was 2. This was a problem that had been no ticed in certain groups and counseling and support was pro vided to the group members by the facilitator and the PBL di rector.
There were positive comments: PBL sessions are very effective in a small group setting and heightens our overall learning;

DISCUSSION
Student perception about small group dynamics during the PBL sessions at the institution was good overall and there were opinions that the PBL process and small group dynamics con tributed to their learning. Research has shown that there are two theoretical perspectives on group learning. The first is a theoretical one while the second is related to motivation [9]. During small group processes like PBL, interactions between group members facilitate learning. A recent review concluded group discussion positively influences students' intrinsic inter est in the subject matter under discussion [10]. The authors also concluded that studies demonstrate that a haphazard dis cussion in the tutorial group or a 'surface' discussion, probably caused by students being less motivated, inhibits student learn ing. A recent study that examined the verbal interactions of one group of PBL students during an entire PBL cycle conclud ed that 53.3% of episodes were collaborative, 27.2% were self directed, while 15.7% were constructive [10]. A PBL group motivates its members to exert maximum effort because a member can attain his/her personal goals only if the group succeeds, so group members help each other because it is in their own interest to do so [4]. The group provides students an opportunity to discuss, argue, present and hear each other's viewpoints and they obtain an opportunity to explain their concepts and what they have learned to others in their group, which stimulates learning. The motivational domain is con cerned with the extent to which students show concern, moti vate, and help each other learn while the demotivational do main deals with the extent to which nonparticipation of stu dents in the group processes affects learning.
TGEI was developed to create a better understanding of wellfunctioning and problematic PBL groups and to help fa cilitators and students develop and implement strategies to improve group functioning during PBL [1]. In the study con ducted at KwaZuluNatal, South Africa the mean cognitive, motivational, and demotivational scores were 3.12, 3.32, and 3.17 respectively [1]. In our study the mean cognitive score was 3.76 while the mean motivational and demotivational scores were 3.65 and 2.71, respectively. The cognitive and mo tivational scores were higher while the demotivational score was lower. The sample size of our study was much smaller than that in the South African study and the student population may have been more homogeneous. At XUSOM, the majority of students are of South Asian origin, although they are now US or Canadian citizens.
The strength of the study was the high response rate; how ever, the sample size was small. Not many studies have been conducted using TGEI. The information obtained using TGEI was not triangulated with that obtained from other sources. Many respondents did not complete all demographic details and so the validity of the information about the influence of gender, nationality, and age on student perception of the small group process could be compromised. This was not a problem during a previous study [11], and it is possible that the small student size at XUSOM made students apprehensive about being identified. The median score for statement 15 about oth er group members contributing less to the discussion was high and we did not enquire about possible reasons for this during the study.
In conclusion, student perception about small group effec tiveness was positive. No major problems were noted but one of the groups had problems with small group dynamics, which was resolved. PBL at XUSOM is still new and sessions have been conducted for only six months. Further studies as PBL progresses and all cohorts of undergraduate medical students have weekly PBL sessions will be required. As most medical schools worldwide already have or are introducing PBL as a learning modality, TGEI can provide valuable information about small group functioning during PBL sessions.

CONFLICT OF INTEREST
No potential conflict of interest relevant to this article was reported.

SUPPLEMENTARY MATERIAL
Audio recording of the abstract.