Skip to main content
. 2016 Jul 28;40(6):387–404. doi: 10.1177/0146621616648931

Table 4.

Proportion of Correct Decisions for Different SPRT Settings for UCCT and MCCT.

Condition α=0.05
α=0.10
δ*=1.25 δ*=2.50 δ*=5.00 δ*=1.25 δ*=2.50 δ*=5.00
Random item selection
 MIRT 0.865 0.866 0.865 0.866 0.865 0.865
 UIRT 0.850 0.851 0.851 0.848 0.850 0.850
Item selection by maximization at the ability estimate
 MIRT 0.894 0.894 0.894 0.894 0.894 0.894
 UIRT 0.837 0.838 0.837 0.837 0.837 0.837
Item selection by maximization at the cutoff point
 MIRT 0.895 0.895 0.895 0.895 0.895 0.895
 UIRT 0.845 0.845 0.845 0.845 0.845 0.845

Note. Simulations for classifications with UIRT and MIRT with a flexible test length. SPRT = sequential probability ratio test; UCCT = unidimensional computerized classification testing; MCCT = multidimensional computerized classification testing; MIRT = multidimensional item response theory; UIRT = unidimensional item response theory; α=β = acceptable error rates; δ* = percentile point that was used to calculate the boundary of the indifference region.