MAT 2379 Lecture Notes - Lecture 11: Blood Sugar, Null Hypothesis, Statistical Hypothesis Testing
Document Summary
We discussed the case unpaired data, i. e. independent samples. We will now discuss the case of paired data. A typical example of dependent data sets are repeated observations that are taken on the same individuals. In this case the two samples have the same size. Scenario: we denote with x1,,xn the observations from a population with mean x and with y1,,yn the observations from a population with mean y . The two data sets are supposed to be dependent. We would like to compare x with y . In other words, (cid:498)paired samples(cid:499) because they always arise in pairs. we want to decide whether the difference d = x y is positive or negative. For this we calculate the differences d1 = x1 y1, d2 = x2 y2, etc. These differences sample mean d and the sample standard deviation sd calculated from the (cid:498)difference(cid:499) will constitute a new data set that will be used for drawing statistical conclusions.