Skip to main content
. 2019 Jan 11;9:2298. doi: 10.3389/fpsyg.2018.02298

Table 4.

Mean Kendall's τ for 1,000 replications between the ordering based on the population test-score reliability and the ordering produced by the three item-score reliability methods and the corrected item-total correlation (CITC), for the bottom-up and the top-down procedure in the six different conditions.

N = 200 N = 1, 000
Small variance Average variance Large variance Small variance Average variance Large variance
of α of α of α of α of α of α
BOTTOM-UP PROCEDURE
Method MS 0.44 (0.13) 0.67 (0.08) 0.80 (0.06) 0.69 (0.08) 0.83 (0.05) 0.89 (0.04)
Method λ6 0.55 (0.11) 0.75 (0.07) 0.83 (0.05) 0.80 (0.05) 0.91 (0.03) 0.94 (0.03)
Method CA 0.58 (0.10) 0.78 (0.06) 0.87 (0.05) 0.81 (0.05) 0.92 (0.03) 0.96 (0.02)
CITC 0.58 (0.10) 0.78 (0.06) 0.87 (0.04) 0.81 (0.05) 0.92 (0.03) 0.96 (0.02)
TOP-DOWN PROCEDURE
Method MS 0.46 (0.14) 0.64 (0.10) 0.75 (0.08) 0.61 (0.11) 0.73 (0.09) 0.80 (0.08)
Method λ6 0.55 (0.10) 0.75 (0.07) 0.83 (0.06) 0.81 (0.05) 0.91 (0.03) 0.94 (0.03)
Method CA 0.59 (0.10) 0.78 (0.06) 0.87 (0.05) 0.81 (0.05) 0.92 (0.03) 0.96 (0.02)
CITC 0.59 (0.10) 0.78 (0.06) 0.87 (0.04) 0.81 (0.05) 0.92 (0.03) 0.96 (0.02)

The Standard Deviation is in Parentheses.