- There are 4 groups of patients, separated by study ID:
- K - dialysis patient with acute infection
- J-dialysis patient without acute infection
- C - non-dialysis patient with acute infection
- D- non-dialysis patient without acute infection
- ADDED: log transform ll37_1borre
- The overall F test for ll37_1borre was borderline significant;
- Tukey’s post-hoc pairwise comparisons shows borderline significance for K versus J;
- If we don’t apply penalty for multiple testing, comparison between K and J is statistically significant, even without log transformation.
## Analysis of Variance Table
##
## Response: log_ll37_1borre
## Df Sum Sq Mean Sq F value Pr(>F)
## group 3 2.880 0.95994 2.5843 0.05722 .
## Residuals 104 38.631 0.37145
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = fit1)
##
## $group
## diff lwr upr p adj
## D-C -0.15970818 -0.624904997 0.3054886 0.8067138
## J-C -0.21245915 -0.659141564 0.2342233 0.6017757
## K-C 0.18680940 -0.256814454 0.6304333 0.6906241
## J-D -0.05275097 -0.483693999 0.3781921 0.9886430
## K-D 0.34651759 -0.081254368 0.7742895 0.1551014
## K-J 0.39926856 -0.008292415 0.8068295 0.0570939
##
## Welch Two Sample t-test
##
## data: log_ll37_1borre by group
## t = -2.4847, df = 58.781, p-value = 0.01583
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.72083606 -0.07770106
## sample estimates:
## mean in group J mean in group K
## 6.018324 6.417592
##
## Welch Two Sample t-test
##
## data: ll37_1borre by group
## t = -2.3177, df = 52.059, p-value = 0.02443
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -455.15707 -32.74099
## sample estimates:
## mean in group J mean in group K
## 495.800 739.749