oneway.test, aov, anovaの違い

RではANOVAを実行する際にoneway.test, aov, anovaを利用できる。違いが判らない。調べてみるとなんかややこしそうなのでとりあえず資料だけ集めてみた。

ONEWAY BETWEEN SUBJECTS ANOVAより

The advantage of using the oneway.test( ) function is obviously the Welch correction for nonhomogeneity. The disadvantages are it doesn’t give you a lot of information in the output, and there is no post hoc test available. The standard R function for all kinds of ANOVA is aov( ). It’s best to store the output of this function and then to extract the information you want by way of various extractor functions. This is the way we will be working with all future statistical procedures. Here is an example using the InsectSprays data…

http://tolstoy.newcastle.edu.au/R/e6/help/09/01/0939.html

The main difference is that aov() *fits* a single model and anova() *summarizes* a single model or compares two or more models.

http://r.789695.n4.nabble.com/lm-ANOVA-vs-AOV-td819452.html

The difference between the two is
intent of the analysis and the default presentation of the results.

With lm [Linear Model], the focus is on the effect of the individual
columns of the predictor matrix.  The columns are usually interpreted
as values of real-valued observations.  The regression coefficients
are usually meaningful and interesting.

With aov [Analysis Of Variance], the focus is on the effects of
factors.  These are multi-degree of freedom effects associated with
categorical variables.  The arithmetic is based on a set of dummy
variables constructed from a contrast matrix.  The individual
regression coefficients themselves are not easily interpretable.