Effect of Outliers on One Sample t-test

Mintu Kr. Das

Abstract: The one sample t-test is a parametric test used to determine whether a sample mean differs significantly from the hypothesized value of the population mean. This one sample location test helps us to generalize our findings to the concerned population. The process of generalizing the results from any research study mostly depends on some summary measures of the test procedure. One such measure is the p-value whose smaller values leads to rejection of null hypothesis. The validity of the decision either to reject or accept the null hypothesis is subject to the fulfillment of the underlying assumptions. Sometimes the presence of outlying observations in a data set violates the assumptions leading to inappropriate p-values. In this paper an attempt has been made to illustrate the effect of outliers on one sample t-test for different sample sizes and varying magnitude of the outliers.

Keywords: outlier, t-test, p-value, location test, hypothesis