"Fail to Reject the Null Hypothesis" is a confusing triple negative. This video uses a methodical approach and creative graphics to dissect the concept and make it understandable and "rememberable".
This was extremely helpful as I am taking an online course and confused by this concept. Thank you for the real-world examples for those of us for whom math is not the easiest!
I am taking an advanced statistics course and was confused on when to accept or fail to reject the null hypothesis. After viewing this video, I feel much more confident in knowing the difference. Your way of explaining, the pace at which you spoke, and the use of graphics is what made the difference between me learning or not learning. I had watched several other videos on this same subject and was still confused. Thank you very much for this video!!
Thank you, Furquan. I intend to keep uploading videos on the individual articles within my book, Statistics from A to Z -- Confusing Concepts Clarified. Currently I'm uploading about once every two weeks. This content has the scope of concepts in 1st and 2nd semester statistics plus statistics used at the Six Sigma Black Belt level. There are over 75 articles covering 60+ concepts.
Taking a stats course so I can get into a good nursing program. I was confused by the terminology in my assignment and this has helped so much! I appreciate the clarification.
Sir I had one question : I am a stock analyst and i want to prove that the stock A's return equals the return of the stock index, wouldn't my alternate hypothesis be Stock A's return = Return of index? I am confused as I have an understanding that the equal to sign is always with the null hypothesis and the null hyp is the opposite of what i am trying to prove? Could you please clarify. I couldn't find its answer nowhere!
This is best explained by slides 10, 11, and 12 in the Alternative Hypothesis video. If we only want to determine whether or not two things are equal, that is like example 1 on slide 10. We would just state the Null Hypothesis as an equation: Stock A's return = Return of Index. That is a 2-sided/ 2-tailed test, as illustrated in slide 12. We only care WHETHER there is a difference and not if the difference is in one direction or the other. End of story. We don't get into the Maintained Hypothesis being the Alternative Hypothesis. It is only when we DO care about the direction of the difference that we get into the Maintained Hypothesis. (Say we want to prove that Stock A 's return is better than the index. Then, we need to use a 1-tailed test. Only then do we start with the Alternative Hypothesis as the Hypothesis which we maintain. See slide 14
If I understand correctly your question is more about what is acceptable wording rather than what the test conclusion is. As Key to Understanding #1 says, if p > Alpha, then we Fail to Reject the Null Hypothesis. And 5% seems to be the most common value selected for Alpha. So, would it be right to say "retain the Null Hypothesis"? It appears that a fair number of experts use that wording,and it may be more acceptable for them than "accept the Null Hypothesis."