CSME Exchange – C. Atwood (Feb 11)


The Center for Science and Mathematics Education (CSME) is pleased to present the CSME Exchange, an interactive brown bag discussion for faculty, students, and staff. The next CSME Exchange is:

“How We Have Used Item Response Theory and Classroom Management to Improve Student Success Rates in Large General Chemistry Classes at the University of Utah”

Charles H. Atwood, Brock L. Casselman, Braden R. Ohlsen,  Chemistry Department, University of Utah, 315 South 1400 East, Salt Lake City, UT 84112, USA, chatwood@chem.utah.edu



Presented by:
Charles Atwood, Ronald and Eileen Ragsdale Professor of Chemical Education
Thursday, February 11, 2016
INSCC room 345
Lunch will be provided!


One of the goals of the University of Utah is to improve student retention and success rates. Historically, general chemistry classes at the University of Utah have been a stumbling block for many students. Since 2011 we have analyzed the attributes of successful students to determine what traits they have which contribute to their success. We recognized that involving students in their work, especially in discussion sections would lead to improved success. To address this we implemented a flipped classroom model along with better-modeled discussion sections. Both measures increased student success (grades of C- or higher) by 12.8% in one year. American Chemical Society nationally normed test scores in all three sections of CHEM 1210 and two sections of 1220 now average 80th percentile.

In 2015 we turned our attention to increasing metacognitive monitoring and control in our poorer performing students. Research, primarily in psychology,1,2,3 has shown that poorer performing students’ ability to accurately assess their content knowledge is deficient. In fact, Dunning and Kruger2 showed that poor students consistently overrate their ability while better students underrate their ability. To address students’ metacognitive skills we implemented in our homework system the ability to give students practice tests where they must estimate their score on the practice test prior to taking it.  In our system, students of all abilities initially overestimate their scores on the practice test but within a practice test cycle or two all students are more accurately assessing their test ability, Figure 1.  Furthermore, initial results indicate that using this system our poorest students are performing 2.5 to 5.0% better than similar students in our control group.  In this talk we will provide details of these improvements and their effect on our classes.

Figure 1. Comparison of students’ accuracy of prediction for the first set of practice tests, as calculated by ‘predicted minus actual.’

Figure 1. Comparison of students’ accuracy of prediction for the first set of practice tests, as calculated by ‘predicted minus actual.’












1) Dunlosky, J.; Metcalfe, J. Metacognition, Sage Publications: Thousand Oaks, CA, 2009.
2) Kruger, J.; Dunning, D., Journal of Personality and Social Psychology, 1999, 77, 1121-1134.
3) Koriat, A.; Sheffer, L.; Ma’ayan, H. Journal of Experimental Psychology, 2002, 131, 147-162.


Join us for an informal discussion and exchange of ideas!

Visit our CSME Exchange page for future meeting times and topics.