Testing the difference between two correlated proportions from one sample

[mils]

Hi SPSS gurus!

Does anyone know how to Test the difference between two correlated proportions from one sample where the proportions are mutually exclusive, using SPSS?

I was wondering if I can test the following proportions:

Product A 235 78.33%
Product B 65 21.67%
       
n=300
p1=78.33%
p2=21.67%
95% level of significance

Let me know if you need more information. Thanks in advance.

[Bruce Weaver]

Maybe I just need more coffee.  I know what mutually exclusive categories are, but what are mutually exclusive proportions?  

Your description suggests a 2x2 table like this:

          B-Yes  B-No
A-Yes    a        b
A-No     c        d

For that table, the usual test is the McNemar chi-square (available via CROSSTABS).  It tests the null hypothesis that b = c.*  Therefore, it also tests the null that (a+b)/N = (a+c)/N.

* In fact, it is equivalent to a chi-square goodness of fit test on the b and c cells with (b+c)/2 as the expected count in each cell.

HTH.

mils wrote

Hi SPSS gurus!

Does anyone know how to Test the difference between two correlated proportions from one sample where the proportions are mutually exclusive, using SPSS?

I was wondering if I can test the following proportions:

Product A 235 78.33%
Product B 65 21.67%
       
n=300
p1=78.33%
p2=21.67%
95% level of significance

Let me know if you need more information. Thanks in advance.

[Ata Ghaderi]

Hej/Hi,

Jag har semester och är åter 19 augusti.

I will be out of office until August 19.

Med bästa hälsningar/Best wishes,

Ata

-------------------------------------------------------------------------------

Ata Ghaderi , Ph.D.

Clinical Psychologist, Licensed Psychotherapist

Professor

Department of Clinical Neuroscience, Karolinska Institutet

SE-171 77 Stockholm

Sweden

[hidden email]

[phone]

[phone]

--------------------------------------------------------------------------------

[Maguin, Eugene]

Please explain 'proportions are mutually exclusive'.  Was this the design: people were asked, "Do you buy product A?" Then, "Do you buy product B?" If so, I believe you will have to use either Genlinmixed or genlin. Both have repeated measures capability.
Gene Maguin

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of mils
Sent: Thursday, July 04, 2013 5:39 AM
To: [hidden email]
Subject: Testing the difference between two correlated proportions from one sample

Hi SPSS gurus!

Does anyone know how to Test the difference between two correlated proportions from one sample where the proportions are mutually exclusive, using SPSS?

I was wondering if I can test the following proportions:

Product A       235     78.33%
Product B       65      21.67%

n=300
p1=78.33%
p2=21.67%
95% level of significance

Let me know if you need more information. Thanks in advance.



-----
mils
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Testing-the-difference-between-two-correlated-proportions-from-one-sample-tp5720999.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Ware, William B]

Sounds like a two by two table... Each case can only be tallied in one of the four cells:  yes/yes, yes/no, no/yes, or no/no.  What about McNemar's test?

Bill

William B. Ware, Ph.D.
Professor of Learning Sciences and Psychological Studies
Adjunct Professor, School of Social Work
McMichael Term Professor of Education 2011-2013
Academy of Distinguished Teaching Scholars at UNC-Chapel Hill - Charter Member
CB #3500 - 118 Peabody Hall
University of North Carolina at Chapel Hill
Chapel Hill, NC     27599-3500
Office:   (919)-962-0132  (Sharon Powell - 309 Peabody)
Fax:        [phone]
Office:  118 Peabody Hall
EMAIL: [hidden email]



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Maguin, Eugene
Sent: Saturday, July 06, 2013 4:52 PM
To: [hidden email]
Subject: Re: Testing the difference between two correlated proportions from one sample

Please explain 'proportions are mutually exclusive'.  Was this the design: people were asked, "Do you buy product A?" Then, "Do you buy product B?" If so, I believe you will have to use either Genlinmixed or genlin. Both have repeated measures capability.

Gene Maguin

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of mils
Sent: Thursday, July 04, 2013 5:39 AM
To: [hidden email]
Subject: Testing the difference between two correlated proportions from one sample

Hi SPSS gurus!

Does anyone know how to Test the difference between two correlated proportions from one sample where the proportions are mutually exclusive, using SPSS?

I was wondering if I can test the following proportions:

Product A       235     78.33%
Product B       65      21.67%

n=300
p1=78.33%
p2=21.67%
95% level of significance

Let me know if you need more information. Thanks in advance.



-----
mils
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Testing-the-difference-between-two-correlated-proportions-from-one-sample-tp5720999.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Rich Ulrich]

For p(A)=p(B)  where the sum is 1?

That is called the "sign test" - a simple binomial proportion.
Are the Ns equal for two groups?

Nonparametric tests.

--
Rich Ulrich

> Date: Thu, 4 Jul 2013 02:39:03 -0700

> From: [hidden email]
> Subject: Testing the difference between two correlated proportions from one sample
> To: [hidden email]
>
> Hi SPSS gurus!
>
> Does anyone know how to Test the difference between two correlated
> proportions from one sample where the proportions are mutually exclusive,
> using SPSS?
>
> I was wondering if I can test the following proportions:
>
> Product A 235 78.33%
> Product B 65 21.67%
>
> n=300
> p1=78.33%
> p2=21.67%
> 95% level of significance
>
> Let me know if you need more information. Thanks in advance.
>

...

[GauravSrivastava]

it seems u have binominal data set. u can choose chi square test for this.

Gaurav

On Jul 7, 2013 10:20 AM, "Rich Ulrich" <[hidden email]> wrote:

For p(A)=p(B)  where the sum is 1?

That is called the "sign test" - a simple binomial proportion.
Are the Ns equal for two groups?

Nonparametric tests.

--
Rich Ulrich

> Date: Thu, 4 Jul 2013 02:39:03 -0700
> From: [hidden email]
> Subject: Testing the difference between two correlated proportions from one sample
> To: [hidden email]
>
> Hi SPSS gurus!
>
> Does anyone know how to Test the difference between two correlated
> proportions from one sample where the proportions are mutually exclusive,
> using SPSS?
>
> I was wondering if I can test the following proportions:
>
> Product A 235 78.33%
> Product B 65 21.67%
>
> n=300
> p1=78.33%
> p2=21.67%
> 95% level of significance
>
> Let me know if you need more information. Thanks in advance.
>
...

[mils]

Hi,


I'm trying to approach this in a different way by using ctables, but it seems that the fact that my data is weighted SPSS won't let me run the columm sig test using the row proportions.

I've attached the error I'm getting.

Any suggestions?


Thanks in advance,

significance_test.PNG