STATS 2B03 Lecture Notes - Variance
Document Summary
Existing data can be used to formulate a hypothesis, but should not be used to test it. In example 14, we need a method that simultaneously compares all 6 pairs of means while controlling the eer at the specified value . X. j = total of the jth treatment. Xavg. j = average of the jth treatment. Xii = the ith data value in sample. Nj = # of observations in the jth sample. Sj2= sample variance of the jth treatment. Terminology the total sum of squares (sst) is defined by (d. f. = n 1) and is a measure of the total variance in the combined data. It is the numerator of the sample variance for the combined data the denominator would be n 1. N 1 is called the degrees of freedom for sst. In our example, the within groups or error sum of squares (sse) is defined by measure of the amount of variation within each sample.