Am I misunderstanding loop variable assignment or how to use print()?

Given this interactive session:

an<-ls(pat="anova.ag.m2529")
an

?[1] "anova.ag.m2529.az" ? "anova.ag.m2529.can" ? "anova.ag.m2529.fl"

print(anova.ag.m2529.az)

Analysis of Variance Table

Response: year
? ? ? ? ? ? ? ? Df Sum Sq Mean Sq F value ? Pr(>F)
time ? ? ? ? ? ? 1 14.823 ?14.8235 ? ?10.598 0.004392 **
Residuals ? 18 ?25.177 1.3987
---
Signif. codes: ? 0 '***'' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

How come the following just prints the names and not the contents of
the existing variable/data frame with the name assigned to x? ?Does it
need to be dereferenced in some way to indicate that it's not a string
but an actual variable living inside the workspace?:

for(x in an) {print(x)}

[1] "anova.ag.m2529.az"
[1] "anova.ag.m2529.can"
[1] "anova.ag.m2529.fl"

I'm trying to get this working interactively first, but ultimately I
need to put it into an Rscript so that I can use dynamic listings and
for loops to print everything. ?I'd also be happy if I could pull just
the Pr out, as I could interactively like so:

print(anova.ag.m2529.az$Pr)

[1] 0.004392059 ? ? ? ? ? ? ? NA

So what am I misunderstanding about how the language works? ?I've been
all over the web and through the usingR.pdf file but can't find an
example that shows something like what I'm trying to do. ?What exactly
(data frame, table, function, character string, etc.) is stored in
'anova.ag.m2529.az' if the commands that created it were:

aov.m2529.az=aov(year~time,data=ag.m2529.az)
anova.ag.m2529.az=anova(aov.m2529.az)

Much thanks in advance.

But "anova.ag.m2529.az" ?is a character string that happens to be the
*name* of an anova object, but R has no way to know that unless you
specifically tell it that your character string is an object by using
get().

Something like print(get(x)) would work.

It's often neater and more efficient to store your anova objects in a
list, though.

is.list(an)

alist<-list(an)
is.list(alist)

alist

On Thu, Dec 15, 2011 at 09:51, Sarah Goslee <sarah.goslee at gmail.com> wrote:

But "anova.ag.m2529.az" ?is a character string that happens to be the
*name* of an anova object, but R has no way to know that unless you
specifically tell it that your character string is an object by using
get().

Something like print(get(x)) would work.

Sarah - Thanks very much! ?That did indeed work great at printing the
entire contents out. ?I couldn't do print(get(x$Pr)), but I can live
with that for now.

It's often neater and more efficient to store your anova objects in a
list, though.

So if I were to do:

is.list(an)

[1] FALSE

alist<-list(an)
is.list(alist)

[1] TRUE

alist

[1] "anova.ag.m2529.az" ? "anova.ag.m2529.can" ? "anova.ag.m2529.fl"

I would have created a list, but I'm assuming that you mean something
different than that since I'm not sure how that functionally changed
anything since it's still a set of character strings. ?Could you
elaborate a bit on what you mean by storing the anova objects as
lists?

print(get(x)[["Pr"]]) maybe. Do the get(), then do the subsetting.

It's often neater and more efficient to store your anova objects in a
list, though.

anything since it's still a set of character strings. ?Could you
elaborate a bit on what you mean by storing the anova objects as
lists?

Yes: not the names, but the anova objects themselves. Rather than
creating a bunch of individual objects, store them in a list when
created:

myanova <- list()
myanova[["ag.m2529.az"]] <- anova(whatever)
myanova[["ag.m2529.can"]] <- anova(whatever)
...

Then you can quite elegantly use lapply() across all of the anovas at once,
and don't have so many objects in your workspace.