R help - Dec 2010 - Problem applying McNemar's - Different values in SPSS and R

Hi friends,
I get different values for McNemar's test in R and SPSS. Which one should i
rely on when the p values differ.
I came across this problem when i started learning R and seriously give up
on SPSS or any other proprietary software.
Thank u in advance

Output in SPSS follows

*Crosstab*


               hsc

Total

     ABN

NE

ABN

tvs

ABN

Count

40

3

43

     Row %

93.0%

7.0%

100.0%

     COL%

78.4%

30.0%

70.5%

  NE

Count

11

7

18

     Row %

61.1%

38.9%

100.0%

     COL%

21.6%

70.0%

29.5%

Total

Count

51

10

61

  Row %

83.6%

16.4%

100.0%

  COL%

100.0%

100.0%

100.0%



 * Chi-Square Tests*


      Value

Exact Sig. (2-sided)

McNemar Test

  .057(a)

N of Valid Cases

61

   a Binomial distribution used.

Output from R is as follows....
> tvshsc<-
+ matrix(c(40,11,3,7),

+ nrow=2,

+ dimnames=list("TVS"=c("ABN","NE"),

+ "HSC"=c("ABN","NE")))
> tvshsc
     HSC

TVS   ABN NE

  ABN  40  3

  NE   11  7
> mcnemar.test(tvshsc)

McNemar's Chi-squared test with continuity correction


data:  tvshsc

McNemar's chi-squared = 3.5, df = 1, p-value = 0.06137

Regards

Dr. B Manoj Aravind

	[[alternative HTML version deleted]]

On Dec 28, 2010, at 11:05 AM, Manoj Aravind wrote:
> Hi friends,
> I get different values for McNemar's test in R and SPSS. Which one
should i
> rely on when the p values differ.
> I came across this problem when i started learning R and seriously give up
> on SPSS or any other proprietary software.
> Thank u in advance
> 
> Output in SPSS follows
> 
> *Crosstab*
> 
> 
>               hsc
> 
> Total
> 
>     ABN
> 
> NE
> 
> ABN
> 
> tvs
> 
> ABN
> 
> Count
> 
> 40
> 
> 3
> 
> 43
> 
>     Row %
> 
> 93.0%
> 
> 7.0%
> 
> 100.0%
> 
>     COL%
> 
> 78.4%
> 
> 30.0%
> 
> 70.5%
> 
>  NE
> 
> Count
> 
> 11
> 
> 7
> 
> 18
> 
>     Row %
> 
> 61.1%
> 
> 38.9%
> 
> 100.0%
> 
>     COL%
> 
> 21.6%
> 
> 70.0%
> 
> 29.5%
> 
> Total
> 
> Count
> 
> 51
> 
> 10
> 
> 61
> 
>  Row %
> 
> 83.6%
> 
> 16.4%
> 
> 100.0%
> 
>  COL%
> 
> 100.0%
> 
> 100.0%
> 
> 100.0%
> 
> 
> 
> * Chi-Square Tests*
> 
> 
>      Value
> 
> Exact Sig. (2-sided)
> 
> McNemar Test
> 
>  .057(a)
> 
> N of Valid Cases
> 
> 61
> 
>   a Binomial distribution used.
> 
> Output from R is as follows....
> 
>> tvshsc<-
> 
> + matrix(c(40,11,3,7),
> 
> + nrow=2,
> 
> + dimnames=list("TVS"=c("ABN","NE"),
> 
> + "HSC"=c("ABN","NE")))
> 
>> tvshsc
> 
>     HSC
> 
> TVS   ABN NE
> 
>  ABN  40  3
> 
>  NE   11  7
> 
>> mcnemar.test(tvshsc)
> 
> 
> McNemar's Chi-squared test with continuity correction
> 
> 
> data:  tvshsc
> 
> McNemar's chi-squared = 3.5, df = 1, p-value = 0.06137
> 
> Regards
> 
> Dr. B Manoj Aravind

The SPSS test appears to be an exact test, whereas the default R function does
not perform an exact test, so you are not comparing Apples to Apples...

Try this using the 'exact2x2' CRAN package:
> require(exact2x2)
Loading required package: exact2x2
Loading required package: exactci
> mcnemar.exact(matrix(c(40, 11, 3, 7), 2, 2))
	Exact McNemar test (with central confidence intervals)

data:  matrix(c(40, 11, 3, 7), 2, 2) 
b = 3, c = 11, p-value = 0.05737
alternative hypothesis: true odds ratio is not equal to 1 
95 percent confidence interval:
 0.04885492 1.03241985 
sample estimates:
odds ratio 
 0.2727273 


HTH,

Marc Schwartz