Abstract and Keywords
Traditional statistical methods are built on strong assumptions, such as normality and homoscedasticity. These assumptions are frequently violated in practice. This can lead to undesirable consequences such as the inaccurate estimation of parameters and confidence intervals, inaccurate calculation of p-values, inflated rates of type I error, and low statistical power. Modern robust statistical methods typically overcome these problems. They are designed to work well both when traditional assumptions are satisfied and when they are not. Using robust methods increases the likelihood of discovering genuine differences between groups and associations among variables. We provide a nontechnical introduction to robust measures of location and scale, bootstrapping, outlier detection, significance testing, and other procedures that have practical value to applied researchers. We discuss software that can be used to conduct robust analyses. Psychological research would benefit from the greater use of robust methods.
Access to the complete content on Oxford Handbooks Online requires a subscription or purchase. Public users are able to search the site and view the abstracts and keywords for each book and chapter without a subscription.
If you have purchased a print title that contains an access token, please see the token for information about how to register your code.