Skip to main content


Fig. 5 | International Journal of STEM Education

Fig. 5

From: Multiple-true-false questions reveal more thoroughly the complexity of student thinking than multiple-choice questions: a Bayesian item response model comparison

Fig. 5

a Multiple-choice (MC) data alone fails to predict the endorsement rates of multiple-true-false (MTF) statements, while b MTF data alone more closely predicts MC results. The black filled circles represent predicted values using only data from the a MC or b MTF format. For comparison purposes, the gray open circles represent predicted values using the full data set (i.e., including both MC and MTF response data). One-to-one lines are shown in solid gray. Root-mean-square deviation (RMSD) is the mean deviation from the one-to-one line in the units of endorsement or selection rate

Back to article page