PSY 230 Lecture Notes - Lecture 6: Normal Distribution, Statistical Power, Null Hypothesis
Document Summary
So even if an event is probable (or improbable), it does(cid:374)"t eli(cid:373)i(cid:374)ate all error: example: p < . 05 means probability is less than 5%. Also called alpha ( ). (telling a man that he is pregnant: type 2 error: failing to find a statistically significant result, even though the research hypothesis is true. Instead of p < . 05, we might lower it to p < . 01. This means we have less than a 1% chance of having a. Population 2 mean : sample from population 1 (n = 64) mean = 220, following 5 steps of hypothesis testing, we conclude the scores are significantly different at a p < . 05 level (we reject the null hypothesis). We conclude there is an effect of the experimental condition. Rules of thumb: small d = . 2, medium, large d = . 8 d = . 5, note: these are absolute values. An effect size of -. 8 is still large: effect sizes in psychology.