Predicting FCI gain with a nonverbal intelligence test

We have administered both a commercial, nonverbal intelligence test (the GAMA) and Lawson’s Classroom Test of Scientific Reasoning to students in two introductory physics classes to determine if either test can successfully predict normalized gains on the Force Concept Inventory. Since gain on the FCI is known to be related to gender, we adopted a linear model with gain on the FCI as the dependent variable and gender and a test score as the independent variables. We found that the GAMA score did not predict a significant amount of variation beyond gender. Lawson’s test, however, did predict a small but significant variation beyond gender. When simple linear regressions were run separately for males and females with the Lawson score as a predictor, we found that the Lawson score did not significantly predict gains for females but was a marginally significant predictor for males.