PSYCH 100A Lecture Notes - Lecture 11: Multiple Comparisons Problem, F-Distribution, Type I And Type Ii Errors
Document Summary
Anova vs. t-test t-test can only compare 2 treatments or groups. Anova can compare 2 or more treatments of groups. Why not just do multiple t-tests: type i error inflation (5% possible error, anova adjusts for this much better when 3 or more groups are being compared. Anova also compares groups to see if they are significantly different from each other t-test accomplished this by looking at mean differences. Anova looks at the amount of overlap between group variances (hence. Factor: the variable (independent or quasi-independent) that designates the groups being compared. Levels: the individual conditions or values that make up the factor. Appropriate research designs for anova: independent-measures design (aka one-way anova, repeated-measures design (not studied in this class, studies that involve more than one factor (factorial anova) Null hypothesis: the treatment had no effect on the dependent variable, h0: 1 = 2 = 3.