Table 5.
Systematic splits (trial grain size) |
Random splits |
||||||||
---|---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 6 | 12 | Mdn | 95% HDI |
||
low | high | ||||||||
n_corr | .949 | .847 | .901 | .799 | .678 | .600 | .862 | .777 | .928 |
n_cat | na | .908 | na | .859 | .613 | .556 | .903 | .795 | .973 |
n_PE | .937 | .913 | .896 | .878 | .814 | .813 | .898 | .855 | .935 |
n_SE | na | .498 | .554 | .592 | .521 | .536 | .509 | .314 | .668 |
n_PE + n_SE | na | .747 | .803 | .823 | .701 | .732 | .767 | .699 | .828 |
n_corr + n_cat | na | .879 | na | .827 | .647 | .598 | .890 | .794 | .957 |
n_corr + n_PE | .967 | .920 | .921 | .890 | .779 | .752 | .913 | .857 | .953 |
n_corr + n_SE | na | .639 | .744 | .730 | .618 | .599 | .668 | .555 | .765 |
n_corr + n_PE + n_SE | na | .824 | .872 | .871 | .739 | .735 | .835 | .781 | .886 |
n_cat + n_PE | na | .938 | na | .913 | .763 | .757 | .927 | .865 | .969 |
n_cat + n_SE | na | .798 | na | .812 | .636 | .598 | .802 | .715 | .870 |
n_cat + n_PE + n_SE | na | .873 | na | .895 | .730 | .725 | .878 | .820 | .923 |
n_corr + n_cat + n_PE | na | .921 | na | .889 | .741 | .720 | .918 | .850 | .965 |
n_corr + n_cat + n_SE | na | .807 | na | .819 | .651 | .616 | .822 | .740 | .887 |
n_corr + n_cat + n_PE + n_SE | na | .874 | na | .886 | .728 | .714 | .881 | .820 | .929 |
Note. M-WCST = Modified Wisconsin Card Sorting Test; HDI = highest density interval (low = lower limit at 95% credibility; high = higher limit at 95% credibility); n_corr = number of correct sorts; n_cat = number of categories; n_PE = number of perseverative errors; n_SE = number of set-loss errors; na = not applicable. Median and HDI of Spearman–Brown reliability coefficients from 100,000 random splits per cell are reported.