This is a practice question for the CFA Level 1 exam, but it's applicable to general statistics as well. Every hypothesis test is based on a level of significance to determine whether we will reject, or fail to reject the null hypothesis. That level of significance makes up the probability that the test statistic falls in the tail(s). And the boundaries of those tails are called the critical values. So in this case, the researcher wanted to set a 5% significance level, that means 2.5% in each tail (because it's a two-tailed test) and from our z-table, that results in a critical value of 1.96. Therefore, if our test statistic falls beyond 1.96 or -1.96, we will reject the null hypothesis. For more questions, please subscribe to our channel. Comment below if you need a different question answered or a topic explained. Questions can also be submitted via our free app: https://itunes.apple.com/us/app/at-no...