[R] How to obtain a mean by more than one other variable

[arun]

Hi,
If `dat` is the dataset:

with(dat,ave(ABUNDHA,SITE,REP,FUN=mean))
#[1] 0.46 0.46 0.14 0.03 0.22 0.67 0.01

I am not sure how you got 0.45 when REP's are not the same.

A.K.

On Sunday, March 9, 2014 11:29 PM, Erynn Call <erynn.call at maine.edu> wrote:
I'm trying to create a new variable (meanABUNDHA) by obtaining a mean value
*by* other variables.  I surveyed birds (multiple species) at multiple sites
and have multiple surveys within a REP (i.e. in example data below, SURVEY
55 and 57 are in REP 11).  I need the mean ABUNDHA (pooling SURVEY) for each
BIRD species by REP and by SITE. Below, I would average 0.77 and 0.15 for
MALL in REP 11 at SITE 1 to get 0.46 meanABUNDHA.

An example of my data are as follows, including the new variable I hope to
create meanABUNDHA:
SITE    REP     SURVEY  BIRD    ABUNDHA meanABUNDHA
1       11            57             MALL      0.77            0.46
1       11            55             MALL      0.15            0.46
1       12            58             MALL      0.14            0.14
1       2              58             BAEA     0.03            0.03
3       4              64             AMCR     0.22            0.45
3       5              65             AMCR     0.67            0.45
7       11             56             MALL      0.01            0.01

I'm not sure how to incorporate more than one "by="?

data.all4[,mean(ABUNDHA),by=
REP]

#Add new column containing the by-REP mean of ABUNDHA
data.all5<-data.all4[,ABUNDHAmean:=mean(ABUNDHA),by=REP]

Thank you for any guidance!!

    [[alternative HTML version deleted]]

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.