The SAS System 01:29 Thursday, February 10, 2005 1 The FREQ Procedure Table of first by second first second Frequency| Percent | Row Pct | Col Pct | 1| 2| Total ---------+--------+--------+ 1 | 794 | 150 | 944 | 49.63 | 9.38 | 59.00 | 84.11 | 15.89 | | 90.23 | 20.83 | ---------+--------+--------+ 2 | 86 | 570 | 656 | 5.38 | 35.63 | 41.00 | 13.11 | 86.89 | | 9.77 | 79.17 | ---------+--------+--------+ Total 880 720 1600 55.00 45.00 100.00 Statistics for Table of first by second McNemar's Test ------------------------------ Statistic (S) 17.3559 DF 1 Asymptotic Pr > S <.0001 Exact Pr >= S 3.716E-05 Simple Kappa Coefficient -------------------------------- Kappa 0.6996 ASE 0.0180 95% Lower Conf Limit 0.6644 95% Upper Conf Limit 0.7348 Sample Size = 1600 The SAS System 01:29 Thursday, February 10, 2005 2 The CATMOD Procedure Data Summary Response first*second Response Levels 4 Weight Variable count Populations 1 Data Set MATCHED Total Frequency 1600 Frequency Missing 0 Observations 4 Population Profiles Sample Sample Size --------------------- 1 1600 Response Profiles Response first second --------------------------- 1 1 1 2 1 2 3 2 1 4 2 2 Response Functions and Design Matrix Function Response Design Matrix Sample Number Function 1 2 -------------------------------------------------- 1 1 0.59000 1 0 2 0.55000 1 1 Analysis of Variance Source DF Chi-Square Pr > ChiSq -------------------------------------------- Model|Mean 1 17.55 <.0001 Residual 0 . . Analysis of Weighted Least Squares Estimates Standard Chi- Effect Parameter Estimate Error Square Pr > ChiSq -------------------------------------------------------------------- Model 1 0.5900 0.0123 2302.44 <.0001 2 -0.0400 0.00955 17.55 <.0001