Conditional Independence: (MS,MW) The GENMOD Procedure Model Information Data Set WORK.COLLAR Distribution Poisson Link Function Log Dependent Variable count Number of Observations Read 8 Number of Observations Used 8 Class Level Information Class Levels Values manager 2 bad good super 2 low high worker 2 low high Criteria For Assessing Goodness Of Fit Criterion DF Value Value/DF Deviance 2 5.3871 2.6936 Scaled Deviance 2 5.3871 2.6936 Pearson Chi-Square 2 5.4104 2.7052 Scaled Pearson X2 2 5.4104 2.7052 Log Likelihood 2599.0752 Algorithm converged. Analysis Of Parameter Estimates Standard Wald 95% Chi- Parameter DF Estimate Error Confidence Limits Square Intercept 1 5.2834 0.0671 5.1519 5.4149 6201.76 Analysis Of Parameter Estimates Parameter Pr > ChiSq Intercept <.0001 Conditional Independence: (MS,MW) The GENMOD Procedure Analysis Of Parameter Estimates Standard Wald 95% Chi- Parameter DF Estimate Error Confidence Limits Square manager bad 1 -1.6954 0.1483 -1.9860 -1.4049 130.79 manager good 0 0.0000 0.0000 0.0000 0.0000 . super low 1 -0.5215 0.0974 -0.7124 -0.3306 28.67 super high 0 0.0000 0.0000 0.0000 0.0000 . worker low 1 -0.8294 0.1024 -1.0301 -0.6287 65.62 worker high 0 0.0000 0.0000 0.0000 0.0000 . manager*super bad low 1 1.4644 0.1681 1.1349 1.7939 75.88 manager*super bad high 0 0.0000 0.0000 0.0000 0.0000 . manager*super good low 0 0.0000 0.0000 0.0000 0.0000 . manager*super good high 0 0.0000 0.0000 0.0000 0.0000 . manager*worker bad low 1 0.8749 0.1601 0.5610 1.1887 29.85 manager*worker bad high 0 0.0000 0.0000 0.0000 0.0000 . manager*worker good low 0 0.0000 0.0000 0.0000 0.0000 . manager*worker good high 0 0.0000 0.0000 0.0000 0.0000 . Scale 0 1.0000 0.0000 1.0000 1.0000 Analysis Of Parameter Estimates Parameter Pr > ChiSq manager bad <.0001 manager good . super low <.0001 super high . worker low <.0001 worker high . manager*super bad low <.0001 manager*super bad high . manager*super good low . manager*super good high . manager*worker bad low <.0001 manager*worker bad high . manager*worker good low . manager*worker good high . Scale NOTE: The scale parameter was held fixed. Conditional Independence: (MS,MW) The GENMOD Procedure Observation Statistics Observation count manager super worker Pred Xbeta Std HessWgt Lower Upper Resraw Reschi Resdev StResdev StReschi Reslik 1 103 bad low low 97.159091 4.5763497 0.094248 97.159091 80.771732 116.8712 5.8409091 0.5925687 0.5867754 1.585496 1.6011497 1.5990147 2 87 bad low high 92.840909 4.5308874 0.0960584 92.840909 76.908497 112.07389 -5.840909 -0.606193 -0.612722 -1.618395 -1.60115 -1.603633 3 32 bad high low 37.840909 3.6333908 0.1308932 37.840909 29.278244 48.907797 -5.840909 -0.94951 -0.975657 -1.645241 -1.60115 -1.616792 4 42 bad high high 36.159091 3.5879284 0.1322028 36.159091 27.905275 46.854219 5.840909 0.9713409 0.9468076 1.5607091 1.6011497 1.5863863 5 59 good low low 51.033259 3.9324776 0.1050445 51.033259 41.537392 62.699977 7.9667406 1.1152032 1.0879267 1.6459556 1.6872231 1.6693197 6 109 good low high 116.96674 4.7618896 0.0831853 116.96674 99.369912 137.67969 -7.966741 -0.73663 -0.745239 -1.706941 -1.687223 -1.690999 7 78 good high low 85.966741 4.4539605 0.09282 85.966741 71.667453 103.11906 -7.966741 -0.859242 -0.873054 -1.714344 -1.687223 -1.694299 8 205 good high high 197.03326 5.2833725 0.0670894 197.03326 172.75592 224.72228 7.9667406 0.5675589 0.5637972 1.6760405 1.6872231 1.6859615 Homogeneous Association: (MS,MW,SW) The GENMOD Procedure Model Information Data Set WORK.COLLAR Distribution Poisson Link Function Log Dependent Variable count Number of Observations Read 8 Number of Observations Used 8 Class Level Information Class Levels Values manager 2 bad good super 2 low high worker 2 low high Criteria For Assessing Goodness Of Fit Criterion DF Value Value/DF Deviance 1 0.0649 0.0649 Scaled Deviance 1 0.0649 0.0649 Pearson Chi-Square 1 0.0649 0.0649 Scaled Pearson X2 1 0.0649 0.0649 Log Likelihood 2601.7363 Algorithm converged. Analysis Of Parameter Estimates Standard Wald 95% Chi- Parameter DF Estimate Error Confidence Limits Square Intercept 1 5.3266 0.0683 5.1928 5.4604 6084.25 Analysis Of Parameter Estimates Parameter Pr > ChiSq Intercept <.0001 Homogeneous Association: (MS,MW,SW) The GENMOD Procedure Analysis Of Parameter Estimates Standard Wald 95% Chi- Parameter DF Estimate Error Confidence Limits Square manager bad 1 -1.6066 0.1484 -1.8975 -1.3157 117.19 manager good 0 0.0000 0.0000 0.0000 0.0000 . super low 1 -0.6420 0.1115 -0.8606 -0.4234 33.14 super high 0 0.0000 0.0000 0.0000 0.0000 . worker low 1 -0.9794 0.1231 -1.2207 -0.7381 63.29 worker high 0 0.0000 0.0000 0.0000 0.0000 . manager*super bad low 1 1.3964 0.1705 1.0622 1.7306 67.06 manager*super bad high 0 0.0000 0.0000 0.0000 0.0000 . manager*super good low 0 0.0000 0.0000 0.0000 0.0000 . manager*super good high 0 0.0000 0.0000 0.0000 0.0000 . manager*worker bad low 1 0.7479 0.1691 0.4165 1.0792 19.57 manager*worker bad high 0 0.0000 0.0000 0.0000 0.0000 . manager*worker good low 0 0.0000 0.0000 0.0000 0.0000 . manager*worker good high 0 0.0000 0.0000 0.0000 0.0000 . super*worker low low 1 0.3847 0.1667 0.0581 0.7114 5.33 super*worker low high 0 0.0000 0.0000 0.0000 0.0000 . super*worker high low 0 0.0000 0.0000 0.0000 0.0000 . super*worker high high 0 0.0000 0.0000 0.0000 0.0000 . Scale 0 1.0000 0.0000 1.0000 1.0000 Analysis Of Parameter Estimates Parameter Pr > ChiSq manager bad <.0001 manager good . super low <.0001 super high . worker low <.0001 worker high . manager*super bad low <.0001 manager*super bad high . manager*super good low . manager*super good high . manager*worker bad low <.0001 manager*worker bad high . manager*worker good low . manager*worker good high . super*worker low low 0.0210 super*worker low high . super*worker high low . super*worker high high . Scale Homogeneous Association: (MS,MW,SW) The GENMOD Procedure NOTE: The scale parameter was held fixed. Observation Statistics Observation count manager super worker Pred Xbeta Std HessWgt Lower Upper Resraw Reschi Resdev StResdev StReschi Reslik 1 103 bad low low 102.26389 4.6275567 0.094763 102.26389 84.929756 123.13593 0.7361053 0.0727912 0.0727041 0.2544114 0.2547161 0.2546912 2 87 bad low high 87.736105 4.4743335 0.1015523 87.736105 71.901326 107.05817 -0.736105 -0.078587 -0.078697 -0.255074 -0.254716 -0.25475 3 32 bad high low 32.736106 3.4884786 0.1508448 32.736106 24.357218 43.997334 -0.736106 -0.128655 -0.129142 -0.25568 -0.254716 -0.254963 4 42 bad high high 41.263895 3.7199879 0.1390302 41.263895 31.421542 54.189227 0.7361047 0.1145921 0.1142539 0.2539642 0.2547159 0.2545639 5 59 good low low 59.736106 4.0899366 0.1199995 59.736106 47.216431 75.575435 -0.736106 -0.09524 -0.095437 -0.255242 -0.254716 -0.25479 6 109 good low high 108.26389 4.6845717 0.0923264 108.26389 90.343158 129.73944 0.7361053 0.0707454 0.0706654 0.2544283 0.2547161 0.2546939 7 78 good high low 77.263895 4.3472268 0.1074415 77.263895 62.592494 95.374206 0.7361053 0.0837436 0.0836112 0.2543132 0.2547161 0.2546726 8 205 good high high 205.73611 5.3265943 0.0682883 205.73611 179.96309 235.20015 -0.736105 -0.05132 -0.05135 -0.254868 -0.254716 -0.254722 Correlation between observed and fitted: (MS,MW) & (MS,MW,SW) The CORR Procedure 2 With Variables: mwmsCI mwmsHA 1 Variables: count Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum mwmsCI 8 89.37500 52.63569 715.00000 36.15909 197.03326 mwmsHA 8 89.37500 54.26561 715.00000 32.73611 205.73611 count 8 89.37500 54.16097 715.00000 32.00000 205.00000 Simple Statistics Variable Label mwmsCI Predicted Value mwmsHA Predicted Value count Pearson Correlation Coefficients, N = 8 Prob > |r| under H0: Rho=0 count mwmsCI 0.99063 Predicted Value <.0001 mwmsHA 0.99990 Predicted Value <.0001 The dissimilarity indices for Blue Collar Worker Example The MEANS Procedure Variable Sum ャャャャャャャャャャャャ count 715.0000000 dCI 0.0386228 dHA 0.0041181 ャャャャャャャャャャャャ