Conventional And Robust Paired And Independent-Samples t Tests: Type I Error And Power Rates
- 1 November 2003
- journal article
- Published by Scilight Press Pty Ltd in Journal of Modern Applied Statistical Methods
- Vol. 2 (2) , 481-496
Abstract
Monte Carlo methods were used to examine Type I error and power rates of 2 versions (conventional and robust) of the paired and independent-samples t tests under nonnormality. The conventional (robust) versions employed least squares means and variances (trimmed means and Winsorized variances) to test for differences between groups.Keywords
This publication has 19 references indexed in Scilit:
- A Power Comparison of Robust Test Statistics Based On Adaptive EstimatorsJournal of Modern Applied Statistical Methods, 2004
- Comparing Trimmed or Least Squares Means of Two Independent Skewed PopulationsBiometrical Journal, 2002
- Testing treatment effects in repeated measures designs: Trimmed means and bootstrappingBritish Journal of Mathematical and Statistical Psychology, 2000
- Simulating Correlated Multivariate Nonnormal Distributions: Extending the Fleishman Power MethodPsychometrika, 1999
- Power, Type I, and Type III Error Rates of Parametric and Nonparametric Statistical TestsThe Journal of Experimental Education, 1999
- Robust Nonorthogonal Analyses Revisited: An Update Based on Trimmed MeansPsychometrika, 1998
- To Trim or Not to Trim: Tests of Location Equality Under Heteroscedasticityand NonnormalityEducational and Psychological Measurement, 1998
- The unicorn, the normal curve, and other improbable creatures.Psychological Bulletin, 1989
- A Method for Simulating Non-Normal DistributionsPsychometrika, 1978
- Robustness?British Journal of Mathematical and Statistical Psychology, 1978