Verbal discourse and clinical reasoning in problem-based medical tutorials: Developing a multifaceted observation system.

Abstract

Traditional learning and teaching methods in medical schools are being challenged by an instructional system that features a problem-focused, patient-oriented, and student-centered learning experience. This approach for improving medical education, "problem-based learning" intends to promote students' active participation in small groups and practice in clinical reasoning. The process is increasingly used by medical educators but the nature of the group interaction is not well-understood. This study developed The Multifaceted Observation System for Tutorials (MOST) to systematically study verbal discourse in problem-based small group tutorials. By providing measures and data collection strategies to observe three process outcomes--cognitive interaction, social interaction, and group process--the system can help describe and understand the nature of this educational process. The measures, some created anew and others constructed by modifying or adapting measures used in other settings, were found to be face-valid through expert coding, and reliable when used in trials by trained observers. The system was constructed using a sample of eight videotapes of tutorial groups produced and archived at the Bowman Gray School of Medicine of Wake Forest University. The videotapes represented several variations, including first- and second-year students, experienced and inexperienced tutors, and tutorials at different points in a ten-week term. Thirty minutes (6 five-minute segments) of tutorial discourse from each of four tutorial groups were coded, analyzed, and graphed using The Event Recorder, software developed at The University of Michigan. The MOST has potential to predict clinical reasoning as well as capture the dynamics of tutorial interaction, if the sample is expanded and fewer factors are allowed to vary simultaneously. Continued research using a larger sample and controlled experimentation is needed to more fully investigate the system's predictive validity. The research concluded: (1) the Multifaceted Observation System for Tutorials (MOST) is a reliable system with face validity that allows the user to observe and analyze three aspects of group functioning in medical tutorials; (2) observers can be acceptably trained to use this computer-assisted observation system; (3) a brief preliminary guide for users can be helpful and is available for continued development; and (4) the MOST is available for further testing and has potential for broader use.