[R] Sorting a data frame with values of different lengths

[Sarah Goslee]

Hi,

It looks to me that your data frame is being sorted as text.

What does
str(W_table)
show?

How was W_table created? Your W column appears to not be numeric.

Sarah

On Tue, Jun 7, 2011 at 9:51 AM, William Armstrong
<William.Armstrong at noaa.gov> wrote:
> Hi all,
>
> I am attempting to run a script in which I permute my data and run a
> Wilcoxon rank sum test on the data 1000 times and compare my original test
> statistic to the permuted test statistics to more accurately estimate a
> significance level for the trends I am observing.
>
> In the process of doing this, I need to sort a data frame that measures 3 x
> 1001 based on the value of the test statistic (W).  I am trying to get the
> data in ascending order based on W.  My data frame looks something like
> this:
>
>> W_table
>   pds_gagehandles.i.  p     W
> W                mibe  1   746
> 2                mibe  2 870.5
> 3                mibe  3   767
> 4                mibe  4  1066
> 5                mibe  5   885
> 6                mibe  6 931.5
> 7                mibe  7   765
> 8                mibe  8   930
> 9                mibe  9 696.5
> 10               mibe 10 711.5
> 11               mibe 11  1006
>
> I am trying to sort it using the command: > W_table[order(W_table$W),],
> which is spitting out:
>
>   pds_gagehandles.i.  p     W
> 11               mibe 11  1006
> 4                mibe  4  1066
> 9                mibe  9 696.5
> 10               mibe 10 711.5
> W                mibe  1   746
> 7                mibe  7   765
> 3                mibe  3   767
> 2                mibe  2 870.5
> 5                mibe  5   885
> 8                mibe  8   930
> 6                mibe  6 931.5
>
> I want this data frame to have the values that are currently the first two
> as the last two, i.e. I need it in ascending order.  I am thinking that I am
> having this problem because W is different lengths and ?order states the
> data should be "a sequence of numeric, complex, character or logical
> vectors, all of the same length, or a classed R object".  Maybe because the
> values over 1000 have an extra digit R is seeing them as '100' for some
> reason? Does anyone know of another function I can use to accomplish this
> task or a way to work around this error?
>
> Thank you very much,
>
> Billy
>
> --


-- 
Sarah Goslee
http://www.functionaldiversity.org