Table 4.

Parameter distributions and statistical tests in Datasets 1 and 2

σ2(0)ωlog(ν)log(κ1)
Dataset 1 (baseline, n = 80)
    Nonclinical controls: mean (SD)2.5 (3.9)−1.3 (2.4)4.1 (1.0)−0.8 (1.4)
    Psychotic: mean (SD)3.0 (3.9)−1.4 (2.0)3.1 (1.1)−0.2 (0.8)
    Clinical controls: mean (SD)1.4 (1.9)−1.2 (2.0)3.3 (1.3)−0.1 (1.4)
    Kruskal–Wallis χ2(2,80)2.33, p = 0.310.22, p = 0.911.9, p = 0.0039.6, p = 0.008
η2 = 0.02η2 = 0.0η2 = 0.15η2 = 0.12
Post hoc Dunn tests
    Psychotic versus nonclinical controlsp(adj) = 0.3p(adj) = 1p(adj) = 0.002p(adj) = 0.01
    Clinical versus nonclinical controlsp(adj) = 0.2p(adj) = 0.7p(adj) = 0.01p(adj) = 0.01
    Psychotic versus clinical controlsp(adj) = 0.2p(adj) = 0.5p(adj) = 0.3p(adj) = 0.4
Dataset 1 (follow-up, n = 55)
    Nonclinical controls: mean (SD)2.8 (3.4)−0.9 (2.0)3.6 (0.8)−1.2 (1.1)
    Psychotic: mean (SD)3.2 (3.7)−1.4 (1.5)2.5 (1.2)−0.3 (0.8)
    Clinical controls: mean (SD)1.2 (0.9)−1.1 (2.0)3.5 (1.1)−0.5 (1.4)
    Kruskal–Wallis χ2(2,80)2.35, p = 0.32.32, p = 0.38.5, p = 0.018.0, p = 0.02
η2 = 0.04η2 = 0.04η2 = 0.16η2 = 0.15
Post hoc Dunn tests
    Psychotic versus nonclinical controlsp(adj) = 0.4p(adj) = 0.2p(adj) = 0.01p(adj) = 0.007
    Clinical versus nonclinical controlsp(adj) = 0.2p(adj) = 0.3p(adj) = 0.5p(adj) = 0.1
    Psychotic versus clinical controlsp(adj) = 0.3p(adj) = 0.3p(adj) = 0.01p(adj) = 0.1
Dataset 2 (n = 167)
    Nonclinical controls: mean (SD)3.1 (2.6)−2.3 (2.0)2.8 (1.0)−0.8 (0.9)
    Scz: mean (SD)1.9(1.5)−2.1 (1.8)2.1 (1.2)0.2 (1.0)
    Mann–Whitney U testZ = 3.1, p = 0.002, r = 0.24Z = −0.6, p = 0.6, r = 0.04Z = 3.9, p = 0.0001, r = 0.3Z = −5.6, p = 3 × 10−8, r = 0.43
Dataset 2 (better-matched controls, n = 116)
    Nonclinical controls: mean (SD)2.8 (2.7)−2.2 (2.1)2.9 (1.1)−0.6 (1.0)
    Scz: mean (SD)1.9 (1.5)−2.1 (1.8)2.1 (1.2)0.2 (1.0)
    Mann–Whitney U testZ = 1.9, p = 0.056, r = 0.18Z = 0.12, p = 0.9, r = 0.01Z = 3.4, p = 0.0007, r = 0.31Z = −4.1, p = 0.00004, r = 0.38