Skip to main content


Fig. 3 | International Journal of STEM Education

Fig. 3

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

Fig. 3

Multiple-choice (MC) correct response rates provide a poor estimation of inferred mastery, while multiple-true-false (MTF) fully correct response rates provide a close approximation of inferred mastery. The x-axis represents the inferred mastery levels for each question derived from student responses in both formats. The y-axis represents the proportion of correct (A) or fully correct (TFFF) responses for MC and MTF questions, respectively. Closed circles represent observed values, and open circles represent predicted values. Vertical lines connect the observed and predicted values for an individual question in a particular format. The long solid lines represent the best fit of the parameter averages for each format. The one-to-one line is shown in solid gray

Back to article page