How to Compute Welch’s t-Test in R? скачать в хорошем качестве

How to Compute Welch’s t-Test in R?

Many methods have been proposed for testing the equality of J means without assuming equal variances (e.g., S. Chen & Chen, 1998; Mehrotra, 1997; James, 1951; Krutchkoff, 1988; Alexander & McGovern, 1994; Fisher, 1935, 1941; Cochran & Cox, 1950; Wald, 1955; Asiribo & Gurland, 1989; Scariano & Davenport, 1986; Matuszewski & Sotres, 1986, Pagurova, 1986; Weerahandi, 1995). Unfortunately, all of these methods, plus many others, have been found to have serious practical problems (e.g., Keselman, Wilcox, Taylor, & Kowalchuk, 2000; Keselman & Wilcox, 1999). One of these problems is poor control over the probability of a Type I error and another is low power under nonnormality, a problem that cannot be escaped when using sample means. The method described here performs reasonably well under normality and heteroscedasticity and it forms the basis of a technique that deals with nonnormality. The method is due to Welch (1951) and it generally outperforms the F-test. Welch’s t-test also known as unequal variances t-test is used when you want to test whether the means of two population are equal. This test is generally applied when the there is a difference between the variations of two populations and also when their sample sizes are unequal. To understand when you will apply the Welch’s t-test, let’s take a look at some scenarios which require you to use Welch’s t-test for better results: 1. When it is assumed that the distribution is normal. 2. When the samples have different variances. 3. Sample sizes of the populations are different.