POSTER

Does time spent on Computer-assisted Instruction (CAI) correlate with exam performance?

Norm Berman, M.D; Leslie Fall, MD Dartmouth Medical School, Chris Maloney, MD University of Utah School of Medicine, Ben Siegel, MD Boston University School of Medicine

Objectives: Understanding how student use of CAI affects learning is essential to making evidence-based decisions about CAI integration. In this pilot study we evaluated the correlation between time spent using CLIPP (Computer-assisted Learning In Pediatrics Project) cases with scores on a final exam based on the content of the CLIPP cases.

Methods: CLIPP cases were integrated into the clerkship at 3 institutions, each of which used the CLIPP final exam. The exam included 4 questions from each of the 20 CLIPP cases studied. Computer log data was queried to find the time spent on each case. Pearson's product-moment correlation test with Bonferroni adjustment was used to assess the correlation between time spent on each case and the percent of correct answers on that case, as well as comparing total time spent on all cases to the overall exam score.

Results: Data were collected on sixty-one medical students. Correlations between the time spent and percent correct answers for each case showed no practical or statistical significance ® range -0.071 to 0.32). The same conclusion was obtained for the correlation between time spent and percent correct answers for the overall exam scores (r=0.17, p=0.19). These correlations were not adjusted for other potential confounders.

Conclusions: This pilot study suggests that students who spend longer times working on CLIPP CAI cases do not score significantly higher on an exam that tests the content of the cases. Variations in learning style, clinical exposure and other teaching may affect these results. This can be tested in the larger dataset that is accruing.