PSYC 3000 Lecture Notes - Lecture 15: A Priori And A Posteriori, Sampling Distribution, Type I And Type Ii Errors
4 Apr 2016
School
Department
Course
Professor
Document Summary
Statistical power= probability that a test will successfully reject a false hypothesis. Lots of power= statistically significant even in small effects. If a test lacks power= it will be difficult to show that there is a significant difference between 2 means even in a large effect size. Type i error: rejection region chosen by alpha level you choose, for 1 tail test- rejection region is all in the tails of the distribution. Reject null hypothesis if it"s out in the tails. Sometimes, a hypothesis can be true even if it"s out in the tails. Falsely rejecting the hypothesis when it"s true is called a type i error. The decision for rejecting or failing to reject depends on the rejection region and is a property of alpha. We want to minimize type i and type ii errors. Reducing type ii error is the same as increasing power of a test. We want to minimize type ii errors and maximize power.
