## ABSTRACT

Alpha (α) Matched Groups Design Region of Rejection Alternative Hypothesis (H1) Matching Variable Relative Power Area in Smaller Proportion Meaningful Difference Repeated Measure Asymmetric Transfer Effect Nonparametric Sequence Effect Beta (β) Normality Serendipitous Finding Carryover Effect Null Hypothesis (H0) Signicant Difference Central Limit Theorem (CLT) One-Sample Chi-Square Spurious Result/Findings Chi-Square Test ( χ2 ) One-Tailed Hypothesis Standard Error of Differences Contingency Table One-Tailed Test Between Means (σDX) Counterbalancing Overestimate Standard Error of Mean (SEM) ( )σX Critical Ratio Parametric Transfer Effect Degrees of Freedom df Pool/Pooling t-Test Experimental Hypothesis Power Type I Error (p = α) Homogeneity of Variance Power Efciency Type II Error (p = β) Independent Data Practice Effect Wilcoxon Matched-Pairs Learning Effect Probability Signed-Ranks Test Mann-Whitney U Randomization Within-Subject Measure

z-Test

We noted in Chapter 1 that the great value of the normal curve in statistical hypothesis testing is that sampling distributions of many statistics are normal in shape. Let us see why this is the case.