# Hypothesis Testing terminology [duplicate]

I am studying now Hypothesis Testing. I use for it book High-Yield Biostatistics, Epidemiology, and Public Health and videos of Brandon Foltz (recommending these videos for everyone, they are great!) from here https://www.youtube.com/channel/UCFrjdcImgcQVyFbK04MBEhA .

In High-Yield Biostatistics, Epidemiology, and Public Health author uses terms like Accept or Reject hypothesis. But Brandon Foltz in his video here says:

As researchers we either reject the null hypothesis or fail to reject the null hypothesis; we do not accept the null.

So, which terminology is more right way to talk about hypothesis: Accept and Reject or Reject and Fail to reject ?

• Yes it is duplicate. I did not found these question before asking. Sorry for that. But I also want to point, that the link provided in the answer of this question by Pio gives better explanation of the topic then answers in other questions. Jul 9, 2014 at 11:54

Reject and Fail to reject since when you're testing your hypothesis you basically challenge it based on some data. It might happen, that your data is just not the right kind/proper sample that could Reject the null hypothesis. This does not mean that the null hypothesis is true, it just means that the null hypothesis describes better your data, than your alternative hypothesis.
In case you Reject the null hypothesis you are Accept-ing the alternative hypothesis. Here is a short article about this topic.
• So we can say we accept alternative hypothesis but we cant say we accept null hypothesis. Instead of we accept null hypothesis we have to say we reject null hypothesis. Am I right? Jul 9, 2014 at 11:40
• I think you wanted to say we fail to reject the null hypothesis. Just think about the the actual meaning of the words accept and fail to reject and how are they different.
• My mistake. I wanted to say: So we can say we accept alternative hypothesis but we cant say we accept null hypothesis. Instead of we accept null hypothesis we have to say we failed to reject null hypothesis. Am I right? Jul 9, 2014 at 11:48