Table 1.
Descriptions of statistical tests
| Data structure | Type of test | Result | Effect size | Power | |
|---|---|---|---|---|---|
| a | Non-normal US condition: W(11) = 0.7185, p = 0.0018 ED condition: W(11) = 0.6417, p = 4.38e-04 |
Wilcoxon rank-sum test | Z = 2.6275, p = 0.0086 | d = 1.3018 | 0.8438 |
| b | Non-normal US condition: W(9) = 0.7890, p = 0.0141 ED condition: W(9) = 0.5623, p = 2.6799e-04 |
Wilcoxon rank-sum test | Z = 0.1890, p = 0.8501 | d = 0.0135 | 0.0501 |
| c | Normal US condition: W(9) = 0.9659 , p = 0.8508 ED condition: W(9) = 0.9713 p = 0.9027 |
Welch’s t test |
t(17.3329) = 0.2777, p = 0.7845 |
d = 0.0343 | 0.0506 |
| d | Non-normal US condition: W(7) = 0.8499, p =0.0951 Control condition: W(7) = 0.9543 p = 0.7547 |
Wilcoxon rank-sum test | Z =100, p = 1.554E-4 | d = 3.613 | 0.99 |
| e | Non-normal US condition: W(7) = 0.8802, p = 0.189 Control condition: W(7) = 0.8802, p = 0.0274 |
Wilcoxon rank-sum test | p = 0.1605 | d = 1.3432 | 0.68 |
Letters (leftmost column) correspond to p values of statistical tests as reported in Results. The data structure, test type, result, effect size, and statistical power of these tests are described. Results of Shapiro–Wilk test for normality of data in US and electrode displacement (ED) conditions (α = 0.05) are reported under Data structure. Normally distributed data were compared with Welch’s t test, and non-normal data were compared with the nonparametric Wilcoxon rank-sum test. Effect sizes were calculated as Cohen’s d with correction for small sample sizes as described by Durlak (2009).