PSYC 301 Lecture Notes - Lecture 7: Null Hypothesis, Sampling Distribution, Test Statistic
Document Summary
In a one-sample t-test, we compare one sample mean against a hypothesized population mean. When you have degrees of freedom equal to infinity. Within participant design (same person is in every group) Test mean of the difference scores against the null hypothesis that the population mean of the difference is 0, d = 0. Formula: is the mean of the difference scores, is the standard error of the mean of the difference scores. Example df = n - 1 (or can think of it as df = # of pairs - 1) This is just a one-sample t-test with the one sample consisting of difference scores and h = 0. Microsoft is interested in whether computer monitor size influences performance. Researchers collected data on the reaction time to complete a visual search task on a 15-inch monitor and on a 42-inch monitor from the same 5 individuals. Participant 15-inch 42-inch difference difference - mean difference squared deviation.