topic Re: Is a two sample t-test different from a two-tailed t-test? Is code below correct for 2-tailed T- in New SAS User

Is a two sample t-test different from a two-tailed t-test? Is code below correct for 2-tailed T-test

gtucke1 — Thu, 08 Dec 2022 01:39:30 GMT

I want to run two-tailed t-tests using the following code:

Title "Age and Dementia Education";
Proc ttest data = X sides = 2 alpha = 0.05 h0=0;
Class h1_unmet;
Var cg_age;
Run;

Is this the correct code?

I've seen both two-tailed t-test and two sample t-tests mentioned. Are they different? I yes, what is the difference.

Re: Is a two sample t-test different from a two-tailed t-test? Is code below correct for 2-tailed T-

PGStats — Thu, 08 Dec 2022 04:23:01 GMT

If h1_unmet takes two values, the statement Class h1_unmet; defines two samples. The option sides=2 requests a two-sided test, i.e. the cg_age difference will be significant if it falls on the extreme left tail or the extreme right tail of the T distribution.

Re: Is a two sample t-test different from a two-tailed t-test? Is code below correct for 2-tailed T-

PaigeMiller — Thu, 08 Dec 2022 12:33:20 GMT

If you have a data set and you want to compare the mean to a specific value (usually zero, but other values are possible), this is a one sample t-test. The hypothesis you are testing is H0: Mean = k

If the data set has two groups (such as male and female), and you want to compare the mean of group 1 to the mean of group two, this is a two sample t-test. The hypothesis you are testing is H0: Mean 1 = Mean 2

Both of the above are two-sided tests, because either the mean > k or mean < k would be a violation of H0

If you want to test to see if the mean is greater than or equal to some value k (as opposed to the mean being equal to k), this is a one sided test. For the two sample case, the one sided test would be H0: Mean1 > Mean 2