One tailed or two-tailed test
Hello.
I have a question regarding the last lecture in statistics.
In the video, p-value was calculated as 0.182 with t-score of 1.34.
I did the calculation as well but I got 0.090, because I chose one-tailed test instead of two-tailed test.
I thought it was more appropriate to use one-tailed test, because the H0 is expressed as 'greater than' rather than 'equal or not equal'.
Could you explain why you used two-tailed test, and therefore got p-value of 0.182 instead of 0.090?
Thank you.
1 answers ( 0 marked as helpful)
Hi Minhee,
Your understanding of one and two-tailed tests is correct. However, in the practical example we're trying to find out if there is a wage gap in the firm, not if male employees earn more than female ones or viceversa. That's why the null hypothesis was actually an equality rather than an inequality. Therefore this would be a two-tailed test and the P-Value would end up being (1 - the critical value from the table) x 2. Here's the problem's null hypothesis:
Hope this helps!
Hope this helps!