Psychology 2810 Lecture Notes - Lecture 10: Null Hypothesis, Point Estimation, Truth Table

• School: Western University
• Department: Psychology
• Course: Psychology 2810
• Professor: Stephen Lupker

Document Summary

Hypothesis (null and alternative) are about a parameter, not a statistic (population, not a sample). Decision: reject (1 - ) (alpha, )--- type 1 error type 2 error (beta) If a sample falls here, then you reject the null. Alpha also represents the probability of making a type 1 error. Eg. ho: mu < 50, ha: mu > 50. 66: in this circumstance, how do we calculate beta, beta is the probability that i will get a probability x bar less than 54. 66, you need more info you need mu, which you don"t know. Here, mu is greater than 50 in ha: you need to be given an assumption of mu, find some statement about the true value of mu, here, mu = 53. Three things that affect power: 1. Power increases as ho is more wrong (more different than null: 2. If you make alpha level bigger, you make beta smaller.