PSYB04H3 Lecture Notes - Lecture 11: Type I And Type Ii Errors, Null Hypothesis, Repeated Measures Design
Document Summary
Psyb04h3 week 10 statistical significance, effect size, power, sample size. P less than 0. 05 is good, meaning 5% or less of sampling error. Probability of finding effect of a specified size. I type ii error (1-a: probability of not rejecting the null hypothesis if it is true. I type i error (1-b: probability of rejecting the null hypothesis if it is false. Need to consider independent effect of each iv and their joint effects. Experimental design names: gender x exam times, 2x3 design, there could also be 3x2x4. All iv"s repeated (cid:373)easures: repeated measure designs. If one iv is independent and one iv is repeated measures: mixed or split-plot design. Main effect = effect of iv on the dv, collapsed across levels of the other iv. I(cid:374)teractio(cid:374) = joi(cid:374)t effect of the 2(cid:894)+(cid:895) iv"s o(cid:374) the dv.
