Angles between subspaces in two-way designs

UNCG Author/Contributor (non-UNCG co-authors, if there are any, appear on document)
Allen Clement Kendall (Creator)
The University of North Carolina at Greensboro (UNCG )
Web Site:
David Herr

Abstract: An experimenter is interested in the effect of several factors, such as age, sex, number of eggs eaten per day, and weight, on blood serum cholesterol. Define as a cell, all observations taken from people who have been identically classified with respect to all factors. In the above case, the observations are blood serum cholesterol counts. An observation may be thought of as an observed value of a random variable, and all of the observations in a cell may be modeled by random variables which are assumed to be independent and identically distributed with unknown mean and variance. The unknown mean and variance are, respectively, the cell mean and cell variance. Suppose one wishes to compare the mean, or average, amount of blood serum cholesterol of two or more cells to each other. For example, compare men over 30 years of age and under 175 pounds who eat one egg per day with women over 30 years of age and under 175 pounds who eat one egg per day, and determine if the blood serum cholesterol count is different for these two cells. This kind of comparison is accomplished by an analysis of variance (ANOVA). An ANOVA is an analysis of the measurements in the various cells, for the purpose of determining if these observed differences represent actual differences between cell means or merely reflect random fluctuations in the data.

Additional Information

Language: English
Date: 1976
Functions, Orthogonal

Email this document to