R help - Jun 2009 - Automatically Adding Categories

Greetings,

I have two files which contain responses to a series of multiple  
choice questions.  One
file contains responses before an "intervention" and the other  
contains the responses afterward.

There were three possible responses to each question: D, F, T (for  
Don't Know, False, and True).

I would like to try McNemar's test to determine if there was any  
significant difference between before
and after.

I read the files.  I create tables using:

firstQuestion <- table( PreSurveyData$q1, PostSurveyData$q2)

for example.

The problem is that for several of the questions not all of the  
possible responses appear.  So I get
a table like this:

	T
D	6
F	2
T	12

Which cannot be used in mcnemar.test because it is not a square table  
and does not have enough rows and columns.

Is there some way to specify that R should count the occurrences of  
D, F, and T even though they do not appear
in the Data?  Or some easy way to add the missing columns?

Thank you,

Jeffrey Edgington

Try this:

table(factor(pre, levels = c("D", "F", "T")),
        factor(post, levels = c("D", "F", "T")))

On Mon, Jun 22, 2009 at 8:56 PM, Jeffrey Edgington <jedgingt@du.edu>
wrote:
> Greetings,
>
> I have two files which contain responses to a series of multiple choice
> questions.  One
> file contains responses before an "intervention" and the other
contains the
> responses afterward.
>
> There were three possible responses to each question: D, F, T (for
Don't
> Know, False, and True).
>
> I would like to try McNemar's test to determine if there was any
> significant difference between before
> and after.
>
> I read the files.  I create tables using:
>
> firstQuestion <- table( PreSurveyData$q1, PostSurveyData$q2)
>
> for example.
>
> The problem is that for several of the questions not all of the possible
> responses appear.  So I get
> a table like this:
>
>        T
> D       6
> F       2
> T       12
>
> Which cannot be used in mcnemar.test because it is not a square table and
> does not have enough rows and columns.
>
> Is there some way to specify that R should count the occurrences of D, F,
> and T even though they do not appear
> in the Data?  Or some easy way to add the missing columns?
>
> Thank you,
>
> Jeffrey Edgington
>
> ______________________________________________
> 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.
>


-- 
Henrique Dallazuanna
Curitiba-Paraná-Brasil
25° 25' 40" S 49° 16' 22" O

	[[alternative HTML version deleted]]

On Jun 22, 2009, at 7:56 PM, Jeffrey Edgington wrote:
> Greetings,
>
> I have two files which contain responses to a series of multiple  
> choice questions.  One
> file contains responses before an "intervention" and the other  
> contains the responses afterward.
>
> There were three possible responses to each question: D, F, T (for  
> Don't Know, False, and True).
>
> I would like to try McNemar's test to determine if there was any  
> significant difference between before
> and after.
>
> I read the files.  I create tables using:
>
> firstQuestion <- table( PreSurveyData$q1, PostSurveyData$q2)
>
> The problem is that for several of the questions not all of the  
> possible responses appear.  So I get
> a table like this:
>
Missing cells added:
	T   D  F
D	6   0  0
F	2   0  0
T	12  0  0
>
> Which cannot be used in mcnemar.test because it is not a square  
> table and does not have enough rows and columns.
>
> Is there some way to specify that R should count the occurrences of  
> D, F, and T even though they do not appear in the Data?  Or some  
> easy way to add the missing columns?

You might want to exercise appropriate care regarding the validity  
concerns, since you obviously have quite a few zero entries.

-- 

David Winsemius, MD
Heritage Laboratories
West Hartford, CT