(Q) I want to test whether there is a significant decrease of smoking status in a sample over a timespan. I would like to use McNemar-ChiSquare-Test with the following syntax: NPAR TEST /MCNEMAR=smoke1 WITH smoke2 (PAIRED) /STATISTICS DESCRIPTIVES /MISSING ANALYSIS. But I donâ„–t get a ChiSquare-Value: I only get the probabilities (exact sig. 2-tailed). Can somebody tell me how to get the McNemar-ChiSquare value? (A) Posted to SPSSX-L list by Marta Garcia-Granero on 2005/01/04 If sample size is low (less than 25 discordant pairs *), NPAR TEST /MCNEMAR will always give the exact p-value. If the number of discordant pairs is GE 10, then Chi-square test (uncorrected and corrected) can be used instead. By the way, McNemar test can be obtained in SPSS with CROSSTABS too (it gives the exact p value for very big sample sizes). (*) Discordant pairs: smoke1=0/smoke2=1 and smoke1=1/smoke2=0. Note: The code assumes all 4 combinations exist in data. Here's a solution: * Simulated data (replace by your own) *. DATA LIST FREE/ smoke1 smoke2 (2 F8.0). BEGIN DATA 1 0 0 0 0 0 1 0 1 1 1 1 1 1 1 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 0 1 0 1 0 0 1 0 0 1 1 1 0 0 0 1 1 0 0 0 0 1 1 1 1 1 1 0 1 1 0 0 0 0 0 1 1 1 1 1 1 0 0 1 1 0 0 1 0 1 0 1 0 1 1 1 1 1 1 0 0 1 1 1 0 END DATA. VALUE LABELS smoke1 smoke2 0 'No' 1 'Yes'. * McNemar's exact test with SPSS (using CROSSTABS) *. CROSSTABS /TABLES=smoke1 BY smoke2 /STATISTIC=MCNEMAR. * McNemar's Chi-square tests (using MATRIX) *. TEMPORARY. SELECT IF (NOT MISSING(smoke1) AND NOT MISSING(smoke2)). AGGREGATE /OUTFILE='c:\\temp\\aggr.sav' /BREAK=smoke1 smoke2 /N=N. MATRIX. PRINT/TITLE="MC-NEMAR'S CHI-SQUARE (CORRECTED & UNCORRECTED)". GET obs /VAR=n /FILE='c:\\temp\\aggr.sav' /MISSING=0. COMPUTE a=obs(1). COMPUTE b=obs(2). COMPUTE c=obs(3). COMPUTE d=obs(4). RELEASE obs. PRINT {a,b,a+b;c,d,c+d;a+c,b+d,a+b+c+d} /FORMAT='F8.0' /RLABEL='A-','A+','Total' /CLABEL='B-','B+','Total' /TITLE='INPUT DATA'. * Corrected & Uncorrected Chi-Square values *. COMPUTE chi2=((b-c)&**2)&/(b+c). COMPUTE chi2sig=1-CHICDF(chi2,1). COMPUTE chi2cor=(ABS(b-c)-1)&**2&/(b+c). COMPUTE chi2sigc=1-CHICDF(chi2cor,1). PRINT {chi2,chi2sig;chi2cor,chi2sigc} /FORMAT='F8.4' /CLABELS='Chi^2','Sig.' /RLABELS='Uncorr.','Correct.' /TITLE="Chi-square tests (df=1) & asymptotic significance". END MATRIX.