Modeling Item Response Times With a Two-State Mixture Model: A New Approach to Measuring Speededness (CT-96-02)
by Deborah L. Schnipke and David J. Scrams
Speededness refers to the extent to which time limits affect test takers' performance. With regard to the Law School Admission Test (LSAT), speededness is currently measured by calculating the proportion of test takers who do not reach each item on the test. These proportions typically increase slightly toward the end of the test, indicating that the LSAT is partially speeded. Because the LSAT is number-right scored (i.e., no points are subtracted for incorrect responses), test takers are encouraged to guess on items rather than leave them blank. Therefore, this measure of speededness for the LSAT probably underestimates the true amount of speededness on the test. A more accurate assessment of speededness should also reflect the tendency of test takers to rapidly guess on items as time expires. This "rapid-guessing" component of speededness can be estimated by modeling response-times with a two-state mixture model, as demonstrated with data from a computer-administered reasoning test. The combined effect of unreached items and rapid guessing provides a more complete measure of speededness than has previously been available.