Non-Parametric Data

Necessary to use non-parametric statistics when the data is not normal.

Usually the Mann-Whitney U test is recommended.

In R can use the Wilcoxon Signed Rank Test which is very similar.

wilcox.test(data.new[,2], mu=11.066)

     Wilcoxon signed rank test with continuity correction

data:  data.new[, 2]
V = 518, p-value = 0.4092
alternative hypothesis: true location is not equal to 11.066

Warning message:
In wilcox.test.default(data.new[, 2], mu = 11.066) :
  cannot compute exact p-value with ties

---

Transcript

We can also compute statistics on nonparametric data, which means the data isn't normal. And one of the most common methods for doing this is the Mann-Whitney U test.  And so in R, we can use the function call for the Wilcoxon Signed Rank Test - which is very similar to Mann Whitney by now passing our data in, passing our mu value in, and now we can see that (yes) we get a p-value of 0.4092 The alternate hypothesis is that the true location isn't equal to 11.066.  And we got a warning message saying that the Wilcoxon with this argument can't compute that exact p-value, because there ties in the data.