Hypothesis Testing of the Difference Between Two Dependent Means using the t-test
When two samples are "dependent" (correlated or linked), then each data point in one sample can be associated in some natural, nonarbitrary way with each data point in the second sample. For example, one of the most common applications of dependent samples is "pretest" vs. "posttest." A Male sample and a Female sample where each Male has one sister in the Female sample is another example. To test the significance of the difference between two sample means when the samples are dependent, you must first calculate for each PAIR of scores the difference between the two scores. Then you must calculate the mean and standard deviation of these differences; then you can enter these summary statistics into the form below.