MATH 1050Y Lecture Notes - Lecture 5: Bonferroni Correction, Aldrin, Multiple Comparisons Problem
14 Apr 2018
School
Department
Course
Professor
Document Summary
Thiamin content was measured on randomly selected samples of four types of central grain. Ha: at least one mean is different from another. Check requirements: random samples, normality, equal variance. Plot supports normality, iqr is about the same. The data provide convincing evidence that the average thiamin content is different for at least one group. It has to be independent sample, different sopts and 3 different samples. Ha: at least one mean is different. Mean of sq= sum sq / df. The data provide convincing evidence that the average aldrin concentration is different for at least one group. Overlap do not- are different construction confidence interval estimates of the means for the different samples, then compare those so dffidence intervals to see if one or more of them does not overlap with the others. There are several formal procedures to determine which means are different : called post-hoc analysis :
