Table 1 Statistical power analysis showing the effect of sample size (n = 6) and test significance level (5%) on statistical power. The effect size was calculated as the difference between two population means (e.g. control and treatment groups) divided by the standard deviation of either population. The analysis was done with nQuery Advisor®, version 5.1, computer program for Wilcoxon (Mann-Whitney) rank-sum test.
Test significance level, α | Power (%) | n per group | Effect size, δ = μ1 - μ2/σ | |||||
|---|---|---|---|---|---|---|---|---|
IL-6 | NO | IL-6 | NO | IL-6 | NO | IL-6 | NO | |
C vs. LPS | 0.05 | 0.05 | 85 | 85 | 6 | 6 | -11.2 | -9.2 |
LPS vs. LPS+TSA | 0.05 | 0.05 | 78 | 4 | 6 | 6 | -2.4 | -0.2 |
LPS vs. LPS+Hele | 0.05 | 0.05 | 53 | 57 | 6 | 6 | 1.4 | 1.5 |
LPS+TSA vs. LPS+TSA+Hele | 0.05 | 0.05 | 77 | 80 | 6 | 6 | 2.3 | 2.7 |