I used Kruskal Wallis test to compare three data groups pra testar comparações em três grupos (b,c,s). The p-valor was < 0,05 in all cases (So, there is difference among at least two groups ). So, I compared pair the paires of groups (to know how groups are different) and it retruned FALSE in all comparisons oO.

> c = c(0,4,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0);

> s = c(0,1,0,1,0,1,0,5,12,0,0,0,0,0,1,0,0,0,1,0,0,0,0,0,0,0,2,0,0,0,1,1,1,0,0,0,0,1,0,1);

> b = c(7,0,0,0,1,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0);

> kruskal.test(list(b,c,s));

Kruskal-Wallis rank sum test

data: list(b, c, s)

Kruskal-Wallis chi-squared = 12.0952, df = 2, p-value =

**0.002364**

> categ = factor(rep(1:3,c(20,40,40)),labels = c("g1","g2","g3"));

> data = c(b,c,s);

> kruskalmc(data,categ);

Multiple comparison test after Kruskal-Wallis

p.value: 0.05

Comparisons

obs.dif critical.dif difference

g1-g2 6.225 19.02047

**FALSE**

g1-g3 8.475 19.02047

**FALSE**

g2-g3 14.700 15.53015

**FALSE**

I didn´t understand those results. It is inconsistent. So, I used U Test with the objective of compare two groups of each time.

wilcox.test(b,c);

Wilcoxon rank sum test with continuity correction

data: b and c

W = 449.5, p-value = 0.07556

alternative hypothesis: true location shift is not equal to 0

Warning message:

cannot compute exact p-value with ties in: wilcox.test.default(b, c)

> wilcox.test(b,s);

Wilcoxon rank sum test with continuity correction

data: b and s

W = 332.5, p-value = 0.1742

alternative hypothesis: true location shift is not equal to 0

Warning message:

cannot compute exact p-value with ties in: wilcox.test.default(b, s)

> wilcox.test(c,s);

Wilcoxon rank sum test with continuity correction

data: c and s

W = 564.5, p-value =

**0.0006209**

alternative hypothesis: true location shift is not equal to 0

**Like you can see, one of comparisons returned c and s groups are statically different. So, please, what can I conclude?**

Thanks in advance!