I have a dataset consisting of x number of variables, var1, var2, ..., varx
I am interested in variable 1, var1, a factor with three levels;"a","b","c", and how this is affected by the other variables.
I have done some descriptive statistics with
and now I want to look at each variable and see if the means for varj in the different grops "a","b","c".
Sorry for not being as clear as I thought I was..
Actually my problem is this, cause I do know that I want to use a two-sided t-test
t.test(varj, alternative="two.sided",conf.level = 0.95,
I do get a result, but only for the mean of varj, and not for the difference in the different groups.
What am I doing wrong?
Think I am a bit to tired for this.
But this is what I have figured out I have to do
anova(lm(var1 ~ varj))
And it seems to work out just fine.
It sounds like a One-Way ANOVA!?
model1 = aov(var1 ~ varj)
If you want to know which pairs/groups are significantly different from each other you can use for instance Bonferroni
pairwise.t.test(var1, varj, p.adj = "bonf")