An Empirical Evaluation of Robust Regression Applied to Reaction-Time Data

During the past two decades there has been a vast growth in the use of reaction-time methods in psychological research, largely because psychologists have recognized that the analysis of reaction-time data can lead to powerful inferences about the structure ofmental processes. However, the problem of contamination of RT distributions by aberrant observations is one that has not been solved either experimentally or analytically, and as inferences become more subtle, even small distortions in the data take on added importance.