Skip to main content
. Author manuscript; available in PMC: 2010 Feb 10.
Published in final edited form as: Psychol Methods. 2002 Mar;7(1):83. doi: 10.1037/1082-989x.7.1.83

Table 7.

Type I Error Rates of Mixed Effects for Causal Steps Methods

Sample size
α value/β value 50 100 200 500 1,000
Judd & Kenny (1981a, 1981b)
Large/zero .0440 0 0 0 0
Medium/zero .0020 .0020 .0020 .0020 0
Small/zero 0 0 0 0 .0020
Zero/large .0100 .0080 .0060 .0100 .0060
Zero/medium .0480 .0040 .0040 .0020 .0080
Zero/small 0 0 0 .0080 0
Baron & Kenny (1986)
Large/zero .0040 0 0 0 0
Medium/zero .0020 .0040 .0020 0 0
Small/zero 0 0 0 0 0
Zero/large .0020 .0060 .0080 .0020 .0040
Zero/medium 0 .0020 .0100 .0020 .0060
Zero/small 0 0 0 .0040 .0020
Joint significance of α and β
Large/zero .0400 .0420 .0500 .0480 .0380
Medium/zero .0400 .0560 .0460 .0500 .0480
Small/zero .0100 .0200 .0300 .0380 .0360
Zero/large .0520 .0520 .0460 .0520 .0340
Zero/medium .0480 .0540 .0620 .0340 .0500
Zero/small .0060 .0160 .0280 .0400 .0420

Note. For all analyses, τ′ = 0. Small value = .14, medium value = .39, and large value = .59. Tests are two-tailed, p = .05. For each method, values in each row are empirical estimates ot the Type I error rate.