 Hi everyone, it's MJ and in this video I want to answer what is the null hypothesis and why can't we accept it? Now with all of these stats questions I like to start by first looking at the big picture and the big picture is that in statistics We have some data. We calculate some things known as statistics in order to get information About parameters and distributions of this thing called the random variable that generated the data in the first place Now where hypothesis testing comes in is hypothesis testing says hold on Let's actually so yeah hypothesis testing. Let's actually make a guess Okay, it's a fancy scientific term for guessing and what we're doing by guessing is we're kick-starting an Investigation we're then gonna carry out a whole bunch of you know tests and We're then gonna gather some evidence from them and we're gonna see if we can either reject or Fail to reject so this brings us to the second part of the question if we know that the null hypothesis You know, what is it? It is a guess that kick starts an investigation We now I need to ask why can't we accept it? Why do we use this language of reject and fail to reject and This is because when we carry out these statistical tests. We are not 100% certain and This is where stats does become a little bit philosophical because if we had to start using the word accept It means that what we're accepting is there for the truth So let's maybe take a step back from philosophy and just look at a very quick example Let's say our null hypothesis is that the parameter that we're looking after is going to be equal to the value of 20 and then let's say we create a confidence interval where the statistical test that generates the confidence interval Gives us the following we have 25 and 30 and this is at a 95% confidence interval because our Parameter value or our guess of 20 does not lie within the confidence interval We can safely say that we reject it But if the confidence interval was say 18 to 23 With the say a 95% confidence interval. We now say fail to reject and the reason for this language like we said is Accept means truth and we're not a hundred percent certain In fact, we know we're not a hundred percent certain because of the fact that these confidence intervals are only at 95% and and what this basically is telling us is that five times out of 100 The confidence interval is going to be letting us down and we've got two types of errors So we've got these errors types of errors We have type one and we have type two Now type one error is when we reject the the null hypothesis But it is actually true. So over here. We reject Our null hypothesis even though it is true Type two is when we fail to reject The null hypothesis Even when it is false and because this is seen as more of a problem or because we want to try and Communicate the fact that we don't have a hundred percent certainty in our statistical test We prefer to use the language of reject and fail to reject rather than the word Accept as this would imply that we're certain that the parameter that we've guessed is true And because we're not a hundred percent certain we therefore use this language of fail to reject Now in most exams you will lose marks if you say accept rather than saying fail to reject So this is an important thing that you do understand so you don't make a mistake in the test that you write But as always if you guys have got any questions, please feel free to ask me and y'all check out the Udemy course for more videos on Statistics keep well. Cheers