[R] Contingency Table - output to odfWeave not working for matrix
Polwart Calum (County Durham and Darlington NHS Foundation Trust)
calum.polwart at nhs.net
Sat Feb 20 14:49:11 CET 2010
Hi guys I'm hoping someone can help me with this. It should be easy but it seems to get stuck for no obvious reason! I am trying to set a report up in odfWeave so that we can re-run the same analysis at 3 time points as the data matures and provide an 'instant' report.
To simplify the situation we have two pieces of data: site_id (numerical value 1-9) and reaction (categorical Y or N).
mydata=""
mydata$site_id = rep(1:9, 15)
mydata$reaction = rep(c("Y","N","N","Y","N"),27)
until now I've simply used the command:
with (mydata, table (site_id, reaction))
to give me a 'table' that looks like this:
reaction
site_id N Y
1 9 6
2 9 6
3 9 6
...
7 9 6
8 9 6
9 9 6
I can pass that as plain text to odf, but I want to make it a properly formatted table.
Executing odfTable(with (mydata, table (site_id, reaction))) results in: Error in UseMethod("odfTable") : no applicable method for "odfTable", which I assume is because the output of table is not a dataframe, vector or matrix which is what odfTable needs. So I got more creative - but nothing is working they way I want it :-(
odfTable(
as.vector (
with (mydata, table (site_id, reaction))
)
)
Produces a table with 9 9's and 9 6's in a vertical column...
odfTable(
as.data.frame (
with (mydata, table (site_id, reaction))
)
)
Doesn't error but produces a table similar to this:
site_id reaction Freq
1 1 N 9
2 2 N 9
3 3 N 9
4 4 N 9
...
15 6 Y 6
16 7 Y 6
17 8 Y 6
18 9 Y 6
Both of the above are how they appear at the command prompt so not odfWeave doing anything wrong...
odfTable(
as.matrix (
with (mydata, table (site_id, reaction))
)
)
At the command prompt this looks the same as the original output so looks like it might work! BUT odfWeave complains Error in UseMethod("odfTable") : no applicable method for "odfTable" -- but its a matrix and odfWeave is supposed to be able to handle matrices? Anyone got any suggestions?
********************************************************************************************************************
This message may contain confidential information. If yo...{{dropped:21}}
More information about the R-help
mailing list