Data has been described as the “new natural resource” and is transforming many fields from business to sport. Team-based learning (“TBL”) with its frequent assessments has the potential to generate over 100,000 data points in a course. With technology being more frequently used in TBL, data can be more easily tracked and analyzed. It is an opportune time to identify approaches to using TBL data.
TBL data can be used to analyze student performance, predict outcomes and optimize learning.
Three case studies of how TBL educators use TBL data.
Case one compared IRAT, TRAT and Final Examination scores. Three effects were identified: i) an increase in TRAT versus IRAT scores of over 20%; ii) a narrowing of the range between the highest and lowest scores between the IRAT and the Final Examination and iii) Final Examination scores closer to TRAT than IRAT scores.
Case two attempted to predict student final course scores by analyzing the impact of various assessments on final grades and found that IRAT and midterm performance was one of the best indicators of final course grades.
Case three analyzed RAT performance. RAT scores and discrimination index at the item or question level were used to identify tough questions or concepts. Once identified, the tough questions were repeated on subsequent RATs so that students would have more practice with tough concepts.
There is enormous potential to use TBL data to analyze, predict and optimize outcomes. However, care must be taken not to blindly apply analysis from one set of learners to another. What could be done is to develop several frameworks for TBL data analysis that could be easily tested, customized and applied in various populations.