Skip to main content
. Author manuscript; available in PMC: 2013 Aug 5.
Published in final edited form as: Appl Psychol Meas. 2012 Aug 16;36(7):548–564. doi: 10.1177/0146621612456591

Table 4.

Numbers of Correctly Estimated Q-Matrices Out of 100 Simulations With K = 3, 4, 5

N = 250
 N = 500
 N = 1,000
 N = 2,000
VJ Q̂ = Q Q̂ ≠ Q Q̂ = Q Q̂ ≠ Q Q̂ = Q Q̂ ≠ Q Q̂ = Q Q̂ ≠ Q
(1 0 0) 91 9 98 2 100 0 100 0
(1 1 0) 82 18 97 3 99 1 100 0
(1 1 1) 70 30 83 17 100 0 100 0
N = 500
 N = 1,000
 N = 2,000
 N = 4,000
VJ Q̂ = Q Q̂ ≠ Q Q̂ = Q Q̂ ≠ Q Q̂ = Q Q̂ ≠ Q Q̂ = Q Q̂ ≠ Q
(1 0 0 0) 91 9 98 2 100 0 100 0
(1 1 0 0) 84 16 94 6 100 0 100 0
(1 1 1 0) 71 29 87 13 99 1 100 0
(1 1 1 1) 39 61 62 38 94 6 100 0
N = 1,000
 N = 2,000
 N = 4,000
 N = 8,000
VJ Q̂ = Q Q̂ ≠ Q Q̂ = Q Q̂ ≠ Q Q̂ = Q Q̂ ≠ Q Q̂ = Q Q̂ ≠ Q
(1 0 0 0 0) 95 5 100 0 100 0 100 0
(1 1 0 0 0) 88 12 99 1 100 0 100 0
(1 1 1 0 0) 77 23 98 2 100 0 100 0
(1 1 1 1 0) 47 53 76 24 92 8 100 0
(1 1 1 1 1)a 29 71 37 63 56 44 88 12
a

In the case of (1 1 1 1 1), Q̂ recovers Q 100 times when N = 12,000.