R help - Feb 2007 - Chi Square with two tab-delimited text files

Hi,

I want to do a chi square test and I have two tab delimited text files with
Expected and Observed values to compare.  Each file contains only the values
and are 48 rows by 116 columns.  I have managed to do something with them,
but I don't think it is right as I got a p value of 1.  In this case I used
the read.table() function to read the values from the files.  But I don't
know if this was right.

 

 
> x=read.table(file="C:/Program Files/R/R-2.2.1/Projects/Stats
EU/Expected
input.txt")

 
> y=read.table(file="C:/Program Files/R/R-2.2.1/Projects/Stats
EU/Observed
input.txt")

 
> chisq.test(x,y)
 

 

 

 

Pearson's Chi-squared test

 

data:  x 

X-squared = 4.4602, df = 5405, p-value = 1

 

Warning message:

Chi-squared approximation may be incorrect in: chisq.test(x, y)

 

 

 

 

Maybe the scan() function is more correct??  Using this I got:

 

 
> x=scan(file="C:/Program Files/R/R-2.2.1/Projects/Stats EU/Observed
input.txt")

Read 5568 items

 
> y=scan(file="C:/Program Files/R/R-2.2.1/Projects/Stats EU/Expected
input.txt")

Read 5568 items

 
> chisq.test(x,y)
 

        Pearson's Chi-squared test

 

data:  x and y 

X-squared = 172306.4, df = 13880, p-value < 2.2e-16

 

Warning message:

Chi-squared approximation may be incorrect in: chisq.test(x, y)

 

 

 

Any help would be much appreciated.

Regards,

 

Carina


	[[alternative HTML version deleted]]

Carina Brehony napsal(a):
> Hi,
> I want to do a chi square test and I have two tab delimited text files with
> Expected and Observed values to compare.  Each file contains only the
values
<snip>

There are a lot of chi^2 tests, most of them compare O&E quantities and 
it is not clear which one you want to use. I'd guess a goodness of fit 
test, but who knows? See ?chisq.test and the examples given there. It 
also tells you that the y-argument is ignored if x is a matrix (that's 
probably the reason why you get different results using read.table and 
scan).
Petr
-- 
Petr Klasterecky
Dept. of Probability and Statistics
Charles University in Prague
Czech Republic