Presentation on theme: "Chapter 13 and 14. Error Type 1: False positive. The effect was not really present, but our statistics lead us to believe that it is. Type 2: False."— Presentation transcript:
Error Type 1: False positive. The effect was not really present, but our statistics lead us to believe that it is. Type 2: False negative. The effect really is present, but our statistics are not able to detect it.
Sample size effects In small sample sizes, it is more difficult to find true effects (type II error is greater) In large sample sizes, it is easier to find true effects, but also easier to find false effects (type I error is greater, type II error is smaller)
Test statistics Most common are F, t, chi-square Looking for a significance level p of less than.05.
ANOVA: Analysis of Variance Do the group means differ significantly? ONE dependent variable MANOVA: multiple analysis of variance – two or more dependent variables