PSY-B - Psychology PSY-B 311 Lecture Notes - Lecture 15: Type I And Type Ii Errors, Multiple Comparisons Problem, Multivariate Analysis Of Variance

• School: IUPUI
• Department: PSY-B - Psychology
• Course: PSY-B - Psychology PSY-B 311
• Professor: Robert Stewert

Document Summary

= variability between the groups + variability within the groups. Tests whether the between-groups variance is larger than we would expect based on the amount of within-groups (error) variance in the data. If iv had no effect the numerator and denominator would be estimates of the same thing (error variance) and the value of f would be around 1. If the iv had an effect the numerator will be larger than the denominator and f will be greater than 1. Spss will give probability for your f value so you don"t have to use tables. Reject null hypothesis if p <0. 05 (iv had a significant effect) (end of review) Two-way anova e. g. , a 2 x 2 factorial design. Total variance = systematic variance + error variance. Sstotal = ssa + ssb + ssaxb + sswg. The two-way anova will generate an f statistic and probability (p value) for each of these 3 questions.