My data aren’t normal, now what?
The validity of all statistical tests rests on underlying assumptions. Typical assumptions of familiar testing procedures include normality of the dependent variables, equal variances among groups and independence of observations. When the data do not meet assumptions of a testing procedure, the p-values can be erroneous leading to incorrect inference. In this seminar, we will review assumptions of commonly used statistical tests and provide insight into how violating the assumptions impact results. We will present various approaches for analyzing data that violate assumptions of standard procedures such as using non-parametric methods and transforming the dependent variable. Participants in this seminar will have a better understanding of what the underlying assumptions are for common statistical tests and possible approaches for data analysis when the data do not conform to these assumptions. Learning objectives: Know what the underlying assumptions are of common statistical procedures Appreciate how violation of assumptions can impact inferential results Be able to employ appropriate strategies to analyze data that do not meet distributional assumptions
451 Health Sciences Drive , Room 4202