[R] Learning ANOVA

[Martin Maechler]

>>>>> Liaw, Andy <andy_liaw at merck.com>
>>>>>     on Mon, 16 Aug 2010 08:22:33 -0400 writes:

    > From: Stephen Liu
    >> 
    >> Hi JesperHybel,
    >> 
    >> Thanks for your advice.
    >> 
    >> >If you're trying to follow the youtube video you have a 
    >> typing mistake here:
    >> 
    >> >InsectSprays.aov <-(test01$count ~ test01$spray)
    >> 
    >> >I think this should be:
    >> 
    >> >InsectSprays.aov <-aov(test01$count ~ test01$spray)
    >> 
    >> 
    >> Your advice works for me.  Sorry I missed "aov" before(test01$count ~ 
    >> test01$spray)

    > I just want to offer another point:  If you see any
    > tutorial/document/book advising you to use model formula as above; e.g.,
    > anything like

    > df$var1 ~ df$var2 + df$var3

    > Just run away from it as fast as you can, and try to wipe it from your
    > memory.  That's about the worst way to use a model formula, and very
    > likely to give you what may seem to be strange problems down the road.
    > Well-written model fitting functions should be called like this:

    > modelfn(var1 ~ var2 + var3, data=df, ...)

YES, indeed, do run away from such bad R teaching...
and maybe really consider reading the tutorials and manuals we
list on  http://www.r-project.org/ or buy a good (not thick!)
book introducing R rather than watching some youtube videos...

Do people now go to Youtube instead of to School and University?
O tempora, o mores!

Martin