How to perform a right tailed test in SAS?

valarievil — Mon, 16 Dec 2019 01:14:40 GMT

Hello! I am trying to perform a right tailed test on a set of data. Its for average gpa of 700 students.

Here is my code:

libname soapy '\\Client\D$\SAS\Data';
data FINAL;
infile '\\Client\D$\SAS\Data\finaldata.txt';
input ID GPA HSRANK ACT YEAR;
run;
proc means data= FINAL n mean stderr alpha=0.02 median p50 clm maxdec=2;
var GPA;
run;
proc univariate data=FINAL alpha=0.02 cibasic mu0=3.03;
var GPA;
run;

I get a confidence interval for the mean which is [2.92, 3.03]. Then I tested for Mu0=3.03 since SAS can't do Mu0>3.03. from the proc univariate command I get the test statistic -2.20461 with an associated p-value 0.0278. (alpha=0.02). These results are for a 2 tailed test with the alternative hypothesis Ha: mean does not equal 3.03. How to get the p-value for Ha: Mu>3.03? Can I get SAS to print it?

Any help would be appreciated! I've attached the text file with the data if it helps...

Re: How to perform a right tailed test in SAS?

PaigeMiller — Mon, 16 Dec 2019 02:21:54 GMT

So you want a 98% 1-sided confidence interval, instead of a 98% 2-sided confidence interval which is what SAS provides by default.

A 98% 1-sided confidence interval has 2% of the distribution in the right tail; a 98% 2-sided confidence interval has 1% of the distribution in the right tail and 1% in the left tail.

So to get 2% in the right tail, you ask SAS for a 96% confidence interval, which produces a test with 2% in the right tail and 2% in the left tail. Then you ignore the left tail, and only use the right tail.

topic How to perform a right tailed test in SAS? in SAS Programming

How to perform a right tailed test in SAS?

Re: How to perform a right tailed test in SAS?