PSYC 217 Lecture Notes - Lecture 24: Homeopathy, Publication Bias, Null Hypothesis
Document Summary
Under one condition that f = t square. When you run a test with two groups. Whether the obtained t is greater than the critical t value. The rule works for non-directional research hypothesis. 1. 65 is the stat to reject null hypothesis. You can have a large t value. If it"s not in the region of your distribution, then it is not significant t test and f test. If the probably that your obtained stat lies on your null distribution is less than 0. 05, then you will reject your null hypothesis. Very unlikely that the result is due to chance. We have a lots distribution but which one. It"s unlikely the difference between two groups has a t of zero. But for sure not a part of the null distribution. Everyone"s svalue can vary as much as they want, except the last one. Group 1 has one less degree of freedom.