Chapter 15 Strategies When Population Distributions are Not Normal: Data Transformations and Rank-Order Tests
Assumptions in the Standard Hypothesis-Testing Procedures Populations follow a normal curve Populations have equal variances Ceiling and floor effects
Data Transformations Data transformation Square-root transformation
Data Transformations Legitimacy of data transformations Kinds of data transformations Square root transformation Log transformation Inverse transformation Arcsine transformation
Rank-Order Tests Rank-order transformation Rank-order tests Nonparametric tests Parametric tests
Rank-Order Tests Basic logic The null hypothesis Normal curve approximations Using parametric tests with rank-transformed data
Comparison of Methods Advantages and disadvantages Relative risk of Type I and Type II errors
Controversies and Recent Development Computer-intensive methods Randomization tests Proposed alternative to parametric and nonparametric methods Widely applicable Unfamiliar to researchers
Reporting in Research Articles Data transformations Described just prior to analyses using them Rank-order methods Described much like any other kind of hypothesis test