ABSTRACT
This chapter considers inferential tests involving the difference between two means. The underlying sampling distribution for the tests is known as the sampling distribution of the difference between two means. This makes sense, as the hypotheses examine the extent to which two sample means differ. The hypotheses to be evaluated for detecting a difference between two means. The Welch t' test is usually appropriate when the population variances are unequal and the sample sizes are unequal. Power for the independent t test can be determined based on reviewing power tables or using statistical software. The assumptions of the independent t test are that scores on the dependent variable Y are normally distributed within each of the two populations, are independent, and have equal population variances. The simplest methods for detecting violation of the normality assumption are graphical methods, such as stem-and-leaf plots, box plots, histograms, as well as statistical procedures such as the Shapiro-Wilk test and skewness and kurtosis statistics.
