topic Re: Comparing means between several groups in Statistical Procedures

Comparing means between several groups

yael — Thu, 30 Mar 2017 08:04:58 GMT

Hello,

I am tring to understand the results I got.

My code is:

proc ttest data=sasuser.sasfile010916;
class stage;
format stage firstgroup.;
var ROA1 ROA2 ROA3 ROA4 ROE1 ROE2 ROIC Q LNQ ;
run;

I copmare between the mean of group 1 to the mean of the other 3 groups mean together.

The result I got for the variable ROA1: T value = 0.06 Pr = 0.94 while F value is 2.77 pr<0.0001

According to the t value, I understand that there is not difference between group 1 to the other 3 groups together. But what is the meaning of the F value here?

Thanks for any clue...

Yael

Re: Comparing means between several groups

PaigeMiller — Thu, 30 Mar 2017 12:18:51 GMT

It would help if we could see the results.

Re: Comparing means between several groups

ballardw — Thu, 30 Mar 2017 14:50:45 GMT

Generally the F test is related to whether your variances are equal. The Ttest is supposed to test if the means differ given that the variances in the two groups are similar. When the variances are not then you maybe shouldn't be using a ttest. A large F value indicates that the variances may not be equal and the p-value is the probabiity they aren't equal given some other assumptions such as normality of the data.