# Thread: Stata rookie - McNemar's chi2

1. ## Stata rookie - McNemar's chi2

Hey,
I've just got to grips with the basics of Stata (for my dissertation). My tutor asked me to use the command 'mcc' to compare binary variables, which is fine - but I'm not sure how to interpret the results.
All I'm really after is a p value!
It offers a whole host of results;

McNemars chi2(1)=2.48
Prob>chi2=0.1151
Exact McNemar significance probability=0.1480

Proportion with factor
Cases .3298
Controls .3085_______________95% Conf Interval

difference .0213______________-.0069 _______ .0495
ratio 1.069__________________.9839_________1.161
rel difference .0308__________-.0069_________.0684

odds ratio 1.522 ______________.8741_________2.697________(exact)

Basically I have no idea what any of these numbers are telling me. Is any one of them the p value?! Sorry for such a basic question from someone with a very poor grasp of stats!

Thanks

2. ## Re: Stata rookie - McNemar's chi2

Originally Posted by ha08255
McNemars chi2(1)=2.48
Prob>chi2=0.1151
Exact McNemar significance probability=0.1480
The 0.1151 and 0.1480 are both p-values, calculated using different methods (that's statistics for you!) I would report the "exact" probability, p=0.15

3. ## Re: Stata rookie - McNemar's chi2

Thank you! Another quick question;

With the logistic regression analysis, is the P>|z| value the p value?!

4. ## Re: Stata rookie - McNemar's chi2

Yes - the p-value is the probability of observing a result of at least this magnitude under the null hypothesis. The null hypothesis is that z is zero, and in a two-tailed test you're worried about the absolute value of z (|z|) rather than its direction.

5. ## Re: Stata rookie - McNemar's chi2

How do you perform McNemar's test in stata? Thanks!!

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts