R help - Jan 2003 - newbie: data frame to vector

Sorry, but I'm very new to R.  I'm trying to figure out how to convert a
column from a data frame to a vector.  More specifically, I have read in a
comma separated value table which contains a number of variables in columns,
plots in rows...

t <- read.table("G:/R/table1.txt", sep=",", header=TRUE)

the data comes in just fine.  Now I want to do an XY plot of 2 variables and
will be doing some statistical analyses on them.  But

plot(t["Var1"], t["Var2"])

Gives me the error, "Error in pmatch(x, table, duplicates.ok) :
        argument is not of mode character"

and

v1 <- t["Var1"]
v2 <- t["Var2"]
plot(v1, v2)

gives me the same error (of course) and v1 does not display the same as a
vector; it looks like it is an array(?)

Thanks,
-Eric

---
Eric B. Peterson
Vegetation Ecologist & Lichenologist
Nevada Natural Heritage Program
Department of Conservation and Natural Resources
1550 E. College Pkwy #137
Carson City, NV 89706
775-687-4245 x236
http://heritage.nv.gov/

On Tuesday 21 January 2003 03:47 pm, Eric Peterson
wrote:
> Sorry, but I'm very new to R.  I'm trying to figure out how to
convert a
> column from a data frame to a vector.  More specifically, I have read in a
> comma separated value table which contains a number of variables in
> columns, plots in rows...
>
> t <- read.table("G:/R/table1.txt", sep=",",
header=TRUE)
>
> the data comes in just fine.  Now I want to do an XY plot of 2 variables
> and will be doing some statistical analyses on them.  But
>
> plot(t["Var1"], t["Var2"])
You want t$Var1 instead of t["Var1"], etc. 

t, incidentally, is the name of an R function, so you might want to avoid 
using it as a variable name.

Deepayan

Hi,

----- Original Message -----
From: "Eric Peterson" <peterson at heritage.nv.gov>
To: <r-help at stat.math.ethz.ch>
Sent: Wednesday, January 22, 2003 10:47 AM
Subject: [R] newbie: data frame to vector

> Sorry, but I'm very new to R.  I'm trying to figure out how to
convert a
> column from a data frame to a vector.  More specifically, I have read in a
> comma separated value table which contains a number of variables in
columns,
> plots in rows...
>
> t <- read.table("G:/R/table1.txt", sep=",",
header=TRUE)
>
> the data comes in just fine.  Now I want to do an XY plot of 2 variables
and
> will be doing some statistical analyses on them.  But
>
> plot(t["Var1"], t["Var2"])

I'm assuming your data frame, t, looks like:
  Var1  Var2  Var3
        1        2        3
        4        5        6
        7        8        9

Then there are (at least) two ways.  The first is type:
  attach(t)
right after your
  > t <- read.table("G:/R/table1.txt", sep=",",
header=TRUE)
this will allow you to use the column names directly, i.e. you can do:
  plot(Var1, Var2)

If you do not attach it, you can still achieve your goal.  Remember that the
data frame is two dimensional, to subset a two dimensional object you need a
comma, i.e
  t[1, ]  # give you row 1
  t[, 1]  # give you column 1
  t[1, 1] # give you the first value in row 1, column 1

So to do what you want you need:
  plot(t[, 1], t[, 2])  # Plot Column 1 and 2

Hope this helps,

Kevin Wang


------------------------------------------------
Ko-Kang Kevin Wang
Master of Science (MSc) Student
Department of Statistics
University of Auckland
New Zealand
www.stat.auckland.ac.nz/~kwan022