On Mar 18, 2010, at 7:04 PM, David Winsemius wrote:
(I am attaching your original question so it will be easier for me
and those who choose to comment further,)
Hi all,
I am very new to R and I'm trying to import data from Microsoft
Access. So
far, I've managed to do so successfully using the following code:
testdb <- file.path("c:\Databse.accdb")
I am surprised that worked. I was under the impression won needed
to double the back-slashes or use "/"'s. But then I don't use
Access or odbc, so I may be completely off base.
channel2 <- odbcConnectAccess2007(testdb)
data.table <- sqlFetch(channel2,"data")
This successfully imports a table(?) called "data.table".
But when I try to run basic stats or manipulate the data I've
imported, I
can only do so when specifying "data.table", rather than the
variables
names (like one can do when importing Excel data. For instance, I
need to
specify:
Please post the results of str(data.table)
There will almost certainly be variable names. They may need to be
quoted if you have spaces or other special characters in them.
--
David.
var_3<-data.table[[1]]-data.table[[2]]
Generally one would want to create a new column within the dataframe:
data.table$var_3 <- data.table$'Fish (PCOrd)' - data.table$Location
or
lm.1<-lm(data.table[[1]]~data.table[[4]])
Your use of parentheses in the variable names may cause problems I
am not skilled enough to work around. Such a practice is not typical
R programming. I would have tried:
lm.1 <- lm( 'Fish (POrf)' ~ 'Treatment Location', data = data.table)
But when I hacked the column names on the lm help page I could not
get names with parens and or space to work properly. My advice
would be that you make your variable names more in conformance to
standard R practices, i.e no parens , dashes, or spaces in variable
names.
It is "possible":
> names(dfrm) <- c('Trt Group', 'Wt (grp)')
> dfrm
Trt Group Wt (grp)
1 Ctl 4.17
2 Ctl 5.58
3 Ctl 5.18
4 Ctl 6.11
5 Ctl 4.50
6 Ctl 4.61
7 Ctl 5.17
8 Ctl 4.53
9 Ctl 5.33
10 Ctl 5.14
11 Trt 4.81
12 Trt 4.17
13 Trt 4.41
14 Trt 3.59
15 Trt 5.87
16 Trt 3.83
17 Trt 6.03
18 Trt 4.89
19 Trt 4.32
20 Trt 4.69
> anova(lm.D9 <- lm(`Wt (grp)` ~ `Trt Group`, data= dfrm))
Analysis of Variance Table
Response: Wt (grp)
Df Sum Sq Mean Sq F value Pr(>F)
`Trt Group` 1 0.6882 0.68820 1.4191 0.249
Residuals 18 8.7292 0.48496
Notice the use of back-ticks in the formula. Mind you my advice still
stands. Your R-life will be easier if you don't need to decide whether
ot not to us back=ticks.
(I would also find a different name for your dataframe. "data.table"
is a function in a package of that name, and you will make some of us
think you are using that package.
--
David.
Any way I can change the import process so that R recognizes
variable
names? Or at least change what I'm importing so that I can simply
use the
variables names stored in Access? I've tried various combinations
of
"columnnames=TRUE" and "rownames=FALSE" with no success (or at
least
anything that makes sense).
Thanks ahead of time for any input or advice...
Ryan
On Mar 18, 2010, at 6:38 PM, Ryan Utz wrote:
David,
Thanks for responding. You were right; I misplaced a backspace (\)
with a
forward space (/) in my posting. The c:\ doesn't actually work; in
my true
code there is a forward space.
Anyway, the results of str(data) appears as follows:
'data.frame': 90 obs. of 9 variables:
$ Fish (PCOrd) : Factor w/ 90 levels "Fish1","Fish10",..:
1 2 3
4 5 6 7 8 9 10 ...
$ Location : Factor w/ 10 levels "16","18","24",..: 4
4 4 10
10 4 4 4 4 4 ...
$ Treatment : Factor w/ 2 levels "RES","UN": 1 1 1 2 2
1 1 1
1 1 ...
$ Treatment Location : Factor w/ 3 levels "IN","TOP","UN": 2 2 2
3 3 2
2 2 2 2 ...
$ Fish Length (cm) : num 7.8 6 7.1 6.6 7.2 6.6 6.4 6 6.15
5.7 ...
$ Fish Weight (g) : num 5.72 3.27 4.46 3.57 4.78 3.41 3.3
2.3 2.42
2.4 ...
$ Dried contents : num 0.024 0.0349 0.017 0.0254 0.0456 ...
$ Combusted contents : num 0.00203 0.00282 0.00155 0.00226
0.00353 ...
$ Estimated content mass: num 6.66e-05 2.91e-03 7.45e-03 1.28e-02
5.46e-03 ...
Any thoughts now?
David Winsemius, MD
West Hartford, CT
______________________________________________
R-help@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.
David Winsemius, MD
West Hartford, CT
______________________________________________
R-help@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.
