Difference between 2 unrelated samples

Question 1

What are the t-test and M-W U test used to test?

Answer 1

To test 2 samples of UNRELATED data

e.g. weights of frogs in 2 difference ponds

Question 2

which test is the parametric and non-parametric?

Answer 2

t-test = pararmetric

Man-Whitney U = Non-parametric

Question 3

What can these tests NOT be used for?

Answer 3

Nominal level data

Question 4

With regard to ordinal data, which test are you better off using and why?

Answer 4

N-P as the t-test tests for difference in MEANS. For ordinal data you should report the MEDIAN.

Question 5

if your data is unbalanced (e.g. Sample 1: N=20 Sample 2: N=24), can you use the t-test?

Answer 5

Yes. You can use the t-test in both balanced and unbalanced data.

Question 6

What results do you get from running a t-test?

Answer 6

1. t-test statistic
2. df (degree of freedom)
3. P-value
   (always ignore the minus sign before the t-test statistic)