PSY 207 Lecture 11: Ch 11: t-Test for Two Related Samples (a:k:a Dependent-Sample t-test)
Document Summary
Ch 11: t-test for two related samples (a/k/a dependent-sample t-test) Each observation in one sample is paired with another single observation in the other sample. Repeated measures: within subjects and pre-post designs. D = x1 - x2 (xtx xcontrol is advised) Difference between each pair of scores in the two dependent samples. Computations are simplified by working with difference scores. Everyone is put in both the caffeine & placebo condition, their memory is measured each time, & their 2 memory scores are subtracted from one another. There would be a variety of d scores: some positive, some negative some small and some large difference scores some would be 0 (if there is no difference) Despite the variety, if there were no effect, on average the d values would be 0. But if there were an effect(say, caffeine placebo generally = 5), the mean would be different than 0.