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

Table 2.

Proportion of Correct Decisions for Different SPRT Settings and Item-Selection Methods.

ρ α=0.05
α=0.10
δ=0.1 δ=0.2 δ=0.3 δ=0.1 δ=0.2 δ=0.3
Random item selection
 0.0 0.865 0.866 0.868 0.866 0.867 0.867
 0.3 0.880 0.882 0.882 0.883 0.882 0.882
 0.6 0.892 0.894 0.893 0.893 0.893 0.890
Item selection by maximization at the ability estimate
 0.0 0.895 0.896 0.897 0.895 0.896 0.895
 0.3 0.908 0.907 0.906 0.909 0.908 0.908
 0.6 0.918 0.917 0.918 0.916 0.915 0.915
Item selection by maximization at the projected estimate
 0.0 0.896 0.895 0.895 0.897 0.895 0.894
 0.3 0.903 0.901 0.904 0.904 0.904 0.901
 0.6 0.910 0.912 0.912 0.912 0.912 0.911
Item selection by maximization at the cutoff point
 0.0 0.898 0.896 0.898 0.897 0.898 0.897
 0.3 0.909 0.910 0.909 0.908 0.910 0.909
 0.6 0.919 0.920 0.918 0.917 0.919 0.919

Note. SPRT = sequential probability ratio test; ρ = correlation between the abilities; α=β = acceptable error rates; δ = distance between the cutoff point and boundary of the indifference region.