Unit 5 Concepts of Hypothesis Testing The null hypothesis which is symbolized with H0 always represents the skeptical view It is a statement that nothing interesting is going on The alternative hypothesis which is symbolized with HA represents the alternative claim under consideration In reality researchers are almost always hoping to reject the null hypothesis If your alternative hypothesis does have a or we call that a one sided alternative If it has a sign we call that a two sided alternative The basic steps of null hypothesis significance testing or NHST are as follows I ll lay it out and then we ll go step by step We compute the probability of getting a sample statistic at least as extreme as the one we got if the null hypothesis were true This is called the p value Like all probabilities the p value ranges from 0 to 1 A small p value means that if H0 were true our sample statistic would not be very likely Thus a small p value suggests that H0 may not be true We compare the p value to a predetermined cutoff called the level If our p value is less than our level we reject the null hypothesis in favor of the alternative hypothesis Each of the probabilities that we ve computed here is a p value data as extreme as what we got or more extreme if the null hypothesis were true It is the probability of getting How do we decide whether the p value is in fact low enough to reject the null hypothesis We compare it to a pre specified cutoff called the level In psychology the most commonly used level is 05 So if p 05 you can reject the null If our p value is less than our level we say that we reject the null hypothesis If our p value is not less than our level we say that we fail to reject the null When we have rejected the null we refer to our result as statistically significant Running one test and reaching the 0 05 level is reasonably good evidence against H0 running 20 tests and reaching that level only once is not Incorrect rejection of the null Type I Error Failure to reject the null Type II Error In psychology power of 8 is considered fairly good This means that you have a 4 in 5 chance of rejecting the null