R help - Mar 2009 - re form data for aov()?

I have data in a file named hands.dat, which is given at the end of this
question.  (It's from a stats textbook example on anova).  I'd like to
do an
aov on this, which I guess would be

d <- read.table("~/hands.dat", header=TRUE)

aov(Bacterial.Counts ~ Water + Soap + Antibacterial.Soap + Alcohol.Spray,
data=d)

but this fails.  Do I need to break d$Method up into columns for each
category, with boolean entries?  Or is there a way to do this more cleanly?

Data file (hands.dat)
===========
Bacterial.Counts        Method
74      Water
84      Soap
70      Antibacterial.Soap
51      Alcohol.Spray
135     Water
51      Soap
164     Antibacterial.Soap
5       Alcohol.Spray
102     Water
110     Soap
88      Antibacterial.Soap
19      Alcohol.Spray
124     Water
67      Soap
111     Antibacterial.Soap
18      Alcohol.Spray
105     Water
119     Soap
73      Antibacterial.Soap
58      Alcohol.Spray
139     Water
108     Soap
119     Antibacterial.Soap
50      Alcohol.Spray
170     Water
207     Soap
20      Antibacterial.Soap
82      Alcohol.Spray
87      Water
102     Soap
95      Antibacterial.Soap
17      Alcohol.Spray

-- 
View this message in context:
http://www.nabble.com/reform-data-for-aov%28%29--tp22769061p22769061.html
Sent from the R help mailing list archive at Nabble.com.

I'm trying to follow along in a text by Velleman and others, with the
'handwashing' example of anova.   I used read.table() to read the data,
and
now I have an object d (put below the dots here), with an entry Method that
has possible values "Water", "Soap", etc.  What I can't
figure out is how to
take this d$Method and use it in an aov call.  I tried

aov(Bacterial.Counts ~ Water + Soap + Antibacterial.Soap + Alcohol.Spray,
data=d) 

but this fails.  Do I need to break d$Method up into columns for each
category, with boolean entries?  Or is there a way to do this more cleanly? 

... the data ...

`d` <-
structure(list(Bacterial.Counts = c(74L, 84L, 70L, 51L, 135L, 
51L, 164L, 5L, 102L, 110L, 88L, 19L, 124L, 67L, 111L, 18L, 105L, 
119L, 73L, 58L, 139L, 108L, 119L, 50L, 170L, 207L, 20L, 82L, 
87L, 102L, 95L, 17L), Method = structure(c(4L, 3L, 2L, 1L, 4L, 
3L, 2L, 1L, 4L, 3L, 2L, 1L, 4L, 3L, 2L, 1L, 4L, 3L, 2L, 1L, 4L, 
3L, 2L, 1L, 4L, 3L, 2L, 1L, 4L, 3L, 2L, 1L), .Label =
c("Alcohol.Spray",
"Antibacterial.Soap", "Soap", "Water"), class =
"factor")), .Names c("Bacterial.Counts",
"Method"), class = "data.frame", row.names = c(NA, -32L))

-- 
View this message in context:
http://www.nabble.com/reform-data-for-aov%28%29--tp22769850p22769850.html
Sent from the R help mailing list archive at Nabble.com.

Dan Kelley <kelley.dan <at> gmail.com> writes:
> 
> 
> I have data in a file named hands.dat, which is given at the end of this
> question.  (It's from a stats textbook example on anova).  I'd like
to do an
> aov on this, which I guess would be
> 
> d <- read.table("~/hands.dat", header=TRUE)
> 
> aov(Bacterial.Counts ~ Water + Soap + Antibacterial.Soap + Alcohol.Spray,
> data=d)
> 
> Data file (hands.dat)
> ===========> 
> Bacterial.Counts        Method
> 74      Water
> 84      Soap
> 70      Antibacterial.Soap
> 51      Alcohol.Spray
To be sure that data are ok, first try

str(d)

Method should be factor.

aov(BacterialCounts~Method,data=d)

Dieter