Statistical Methods for Quality Improvement, Third Edition

CHAPTER 16

Analysis of Means

To some people the term analysis of means conjures up notions of analysis of variance (ANOVA), which is also concerned with the analysis of means, and which is much better known and more widely used than the statistical technique known as analysis of means (ANOM). (The relative obscurity of ANOM is indicated by the fact that it is not included among the approximately 2700 articles on statistical methods and other statistical topics that are included in the Encyclopedia of Statistical Sciences.) It is apt to have more appeal to engineers and other industrial personnel than does ANOVA, however, since ANOM is inherently a graphical procedure and is somewhat similar to a control chart.

Since ANOM is essentially a multiple comparison procedure, its potential applications extend far beyond engineering and industry in general, although such applications have not been plentiful. For example, Homa (2007) stated that ANOM “is underutilized in health care improvement work.”

For decades, only a very small number of people have been active in publishing research articles about ANOM and writing about ANOM in books. Consequently, there is not broad knowledge about ANOM, either in the statistics community or other fields, or in books. Consequently, when someone who has not previously written about ANOM comments on it, those comments may need some clarification, as in Ryan (2006).

ANOM was developed by E. R. Ott and presented in Ott (1958). It was introduced into the ...