This preview shows page 1. Sign up to view the full content.
Unformatted text preview: ost generally: the slope of a given line is different from 0 (which means no relationship between X and Y) • Type 1: Paired (two measures of the same item) o E.g. Before and After. Does heart rate increase after drinking a cup of coffee? Does plant growth increase after adding fertilizer to pots of soil? • Types 2 and 3: Unpaired o E.g. Do MSE students downtown consume more coffee than MSE students at Mac? o Type 2 if we are sure the variances are equal, Type 3 if we are not. • Comparing distributions: concepts o The difference between means of two distributions might be the same even if the variability changes considerably (tighter/looser distributions). o We need to judge the difference to their means relative to the spread or variability of their scores. • Two
sided vs one
sided t
tests o One
sided: The mean of group 1 is greater than the mean of group 2. Used if we are given clear information about which group is which. o Two
sided: Used if we know the means of the two groups are different, but we are not sure column of data corresponds to which group, or whether we are “blind” to...
View Full
Document
 Winter '13
 JeffreyCardille

Click to edit the document details