Self-directed learning versus lecture in medicine.

Abstract

Despite the predominance of large-group lecture in medical education, little research demonstrates its value over more self-directed learning. Medical students were randomly assigned either to attend lectures or not to attend lectures. Those not attending lectures were provided a weekly session to ask questions of the course director. Results on a content test showed no difference between the groups. Students in the self-directed group seldom attended the weekly sessions but liked the flexibility of the self-directed approach.

Introduction

The traditional didactic curriculum dominated by the large group lecture has a long tradition in medical education (Ludmerer, 1999). The typical medical school lecture involves as single lecturer in a large amphitheater supported by some degree of audiovisual technology such as PowerPoint. Efforts to introduce new curricular formats and instructional methods such as problem-based, self-directed learning (Barrows, 1983) have met with limited success (Albanese & Mitchell, 1993; Colliver, 2000; Vernon & Blake, 1993), and some hybrids of traditional and self-directed methods have emerged (e.g., Clough et al., 2004). In these hybrid curricula, lecture typically plays a reduced role or is completely eliminated (e.g., Crowder, Miller, Sadler, & Mohl, 1996). However, experimental tests of specific aspects of such new curricula are often lacking, specifically between lectures and more self directed alternatives.

Several relevant studies in this area have been reported. Peng (1989) experimentally compared self-directed learning with a traditional curriculum approach in a three year study. The students in the self-directed learning condition received self-learning guidelines including how to organize their own discussion sessions and had a 30-57 percent reduction in lecture hours. Over the three years, the self-directed learning groups performed equal to or higher than the students in the control groups.

Lake (2001) described an experimental study in which a lecture section of a physiology course was compared with sections in which the same content was covered using small group discussion. Results indicated that course grades were higher in the group discussion sections than in the lecture section. However, students in the group discussion sections perceived that they had learned less even though there was no difference in their perception of course difficulty. Furthermore, students in the group discussion sections had lower perceptions of course and instructor quality. Although it may not be perceived so by students, replacing lecture with self-directed learning in the form of group discussion may offer some learning advantages.

Haidet, Richards, Morgan, Wristers, and Moran (2003) conducted a controlled trial of self-directed learning versus lecture with medical residents studying evidence based medicine. All instruction took place in large group sessions, but in one segment lecture was broken up by small group interactive discussion assignments. Content was carefully standardized and tested and interactive discussion sessions were observed. Students were observed to be more active during the interaction sessions, but there was no difference in knowledge gain between sessions immediately or one month later. However, students rated the didactic lecture as more valuable. These results are similar to those of Lake (2001).

However, experimental tests of replacing lecture with other self-directed approaches besides group discussion appear to be lacking. The present study was a randomized experiment comparing traditional lectures to a self-directed learning approach that involved replacing all lectures with free student study time and the availability of weekly question-answer session with the course coordinator. Both quantitative measurements of learning and student perceptions of the self-directed approach were employed.

Method

Our sample was drawn from second year students at a U. S. college of medicine who were taking the three-week internal medicine component of their gastroenterology course (studying diseases of the intestine and the colon). Institutional review board approval was gained. Of 65 eligible students, 41 consented to participate. The participants were randomly assigned to either an experimental group (n = 21) or a control group (n = 20).

The control group received twelve one-hour lectures delivered by medical doctors with appropriate specializations. The lecturers were cautioned not to indicate any material that would definitely be on the test. The experimental group did not attend the lectures but was provided with weekly two-hour question-answer sessions with the course coordinator. The coordinator likewise avoided any discussion of what would definitely be on the test. The burden of responsibility was placed on the students to bring questions to the session, and attendance was voluntary. Access to instructors outside class was otherwise the same for both groups including email, phone, and office visits. Both groups were assigned the same readings and had electronic access to course notes and PowerPoint lecture slides on the course website.

A 36-item multiple-choice test was administered to all students at the end of the three weeks. The test was comprised of material taught during the three weeks only, and all test items were approved by the course coordinator. Only content taught in the course was included on the test, ensuring content validity.

An eight-item questionnaire was also administered to the experimental group. The questionnaire was designed to investigate if and how students in the experimental group altered their study habits and their general impressions of self-directed learning. The items included ratings of whether the students (1) read more, (2) studied with others more, (3) spent more time studying, (4) gained in self-reliance as a student, (5) believed they learned more, (6) believed the question-answer sessions were helpful, (7) preferred self-directed learning to lecture, and (8) liked the flexibility offered by self-directed learning. The questionnaire used Likert scales of Strongly Agree, Agree, Undecided, Disagree, Strongly Disagree, or Not Applicable (N/A).

Results

Randomization resulted in the assignment of 11 females and 9 males to the control group, and 11 females and 10 males to the experimental group. Statistical comparison of the control and experimental groups showed no statistically significant difference in aptitude for medical education as measured by their Medical College Admission Test total scores (p = .59). Hence, the two groups were equivalent in gender and general aptitude with randomization accounting for any other factors. Lecture attendance was observed to be normal in the control group. However, few students in the experimental group attended the voluntary question-answer sessions. A total of five students attended for very brief periods during the three sessions. Means and standard deviations on the multiple-choice content test were almost identical. Of a possible score of 100 percent, the control group had a mean score of 86.8 (SD = 6.5), and the experimental group had a mean score of 87.3 (SD = 7.6). A one-way analysis of variance test showed no statistically significant difference between the groups (p = .83).

Results of the questionnaire indicated that students in the experimental group did not read more, study more with other students, or believe question-answer sessions were helpful. Students were split on whether they spent more time studying, felt that they learned more, gained in self-reliance, or found the self-directed approach preferable to lecture. However, most liked the flexibility in studying provided by the self-directed approach. Specific differences by item are as follows: Of the students in the experimental group, 67 percent disagreed or strongly disagreed that they read more, 76 percent disagreed or strongly disagreed that they studied with others more, and 67 percent were undecided or marked N/A about the usefulness of the question-answer sessions, reflecting the fact that few attended these sessions.

The results for four other items were generally more mixed (totals of less than 100% indicate indications of N/A and rounding error). On the item regarding spending more time studying, 43 percent disagreed or strongly disagreed, 24 percent agreed or strongly agreed, and 29 percent were undecided. On the item regarding whether they felt they learned more, 38 percent disagreed or strongly disagreed, 38 percent agreed or strongly agreed, and 19 percent were undecided. On the item about gaining self-reliance, 29 percent disagreed or strongly disagreed, 38 percent agreed or strongly agreed, and 24 percent were undecided. On the item regarding preference for self-directed learning over attending lectures, 33 percent disagreed or strongly disagreed, 43 percent agreed or strongly agreed, and 19 percent were undecided. However, on the item regarding liking the flexibility in studying provided by the self directed approach, 67 percent agreed or strongly agreed, 14 percent disagreed, and 14 percent were undecided. No one strongly disagreed.

Discussion

The key instructional variable in this study was attendance at live lectures. All participants otherwise had equal access to written lecture notes, PowerPoint slides, assigned readings, and contact with instructors. Students in the self-directed group did not use the weekly question-answer sessions or see them as especially helpful. Our key finding was that attendance at live lectures produced virtually identical test performance to no live lecture attendance. Therefore, the cost effectiveness of live lectures may be questionable. Moreover, students strongly liked the flexibility of the self-directed approach.

However, students in the self-directed group were generally split on whether they spent more time studying overall, learned more, gained in self-reliance, or found the self-directed approach preferable to lecture. These results bear comparison to the results of Lake (2001) and Haidet et al. (2003) who found improved or equivalent learning but unimproved student perceptions for self-directed learning over traditional lecture.

Specifically, several studies have now shown that self-directed learning, operationally defined as either small group discussion or as replacing lecture with free time and access to question-answer sessions, results in equivalent or superior learning but either a preference for traditional lecture or no clear preference for any approach. The reasons for this are not immediately clear. Perhaps students feel some comfort in traditional, didactic lecture and gain direction from it about what content to study and learn. But ultimately, all learning is personal learning and it is done alone. Norman (2004, p. 1) speculated about this very point, suggesting that learning takes place "around, under and beside the curriculum, not as part of it" (cf. Norman & Schmidt, 2000). Norman called for a more eclectic and flexible approach to learning and less reliance on dogmatic adherence to any one teaching approach. This seems to be a reasonable position, but the issue of cost effectiveness may be on the side of allowing students to control input of the information themselves.

Our results deserve replication in other contexts. Generalizable conclusions are limited by the sample and scope of our study, but our results do have the strength of a randomized trial, and they tend to agree with the results of other studies. The lecture approach is one of many instructional methods in medical education. Perhaps this established tradition might yield time to more flexible, student-directed methods with no loss in learning and gains in cost effectiveness.

References

Alabanese, M. A., & Mitchell, S. (1993). Problem-based learning: A review of the literature on its outcomes and implementation issues. Academic Medicine, 68, 52-81.

Barrows, H. S. (1983). Problem-based, sell-directed learning. Journal of the American Medical Association, 250, 3077-3080.

Clough, R.W., Shea, S. L., Hamilton, W. R., Estavillo, J. A., Rupp, G., Browning, R. A., & Lal, S. (2004). Weaving basic and social sciences into a case-based, clinically oriented medical curriculum: One school's approach. Academic Medicine, 79, 1073-1083.

Colliver, J. A. (2000). Effectiveness of problem based learning curricula: theory and practice. Academic Medicine, 75, 259-266.

Crowder, D. J., Miller, D. A., Sadler, J. Z., & Mohl, P. C. (1996). Self-directed learning in a psychopathology course. Academic Psychiatry, 20, 101-110.

Haidet, P., Richards, B. F., Morgan, R. O., Wristers, K., & Moran, B. J. (2003). A controlled trial of active versus passive learning strategies in a large group setting. Advances in Health Sciences Education, 9, 15-27.

Lake, D. A. (2001). Student performance and perceptions of a lecture-based course compared with the same course utilizing group discussion. Physical Therapy, 81, 896-902.

Ludmerer, K., M. (1999). Time to heal. New York: Oxford University Press.

Norman, G. (2004). Editorial--What's the active ingredient in active learning? Advances in Health Sciences Education, 9, 1-3.

Norman, G. R., & Schmidt, H. G. (2000). Effectiveness of problem-based learning: theory, practice, and paper darts. Medical Education, 34, 721-728.

Peng, W. (1989). Self-directed learning: A matched control trial. Teaching and Learning in Medicine, 1, 78-81.

Vernon, D. T. & Blake, R. L. (1993). Does problem-based learning work? A meta-analysis of evaluative research. Academic Medicine, 68, 550-563.

Richard Dusold, Texas A&M Health Science Center

Mark Sadoski, Texas A&M University, Texas A&M Health Science Center

Dusold, M.D. is an Assistant Professor in the College of Medicine, and Sadoski Ph.D., is a Professor with joint appointments in the College of Education and the College of Medicine.