The SAS System The LOGISTIC Procedure Model Information Data Set WORK.SCOUT1 Response Variable (Events) y Response Variable (Trials) n Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 2 Number of Observations Used 2 Sum of Frequencies Read 800 Sum of Frequencies Used 800 Response Profile Ordered Binary Total Value Outcome Frequency 1 Event 97 2 Nonevent 703 Class Level Information Design Class Value Variables S nonscout 0 scout 1 Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Deviance and Pearson Goodness-of-Fit Statistics Criterion Value DF Value/DF Pr > ChiSq Deviance 0.0000 0 . . Pearson 0.0000 0 . . Number of events/trials observations: 2 The SAS System The LOGISTIC Procedure Model Fit Statistics Intercept and Covariates Intercept Log Full Log Criterion Only Likelihood Likelihood AIC 593.053 587.440 15.083 SC 597.738 596.810 24.453 -2 Log L 591.053 583.440 11.083 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 7.6126 1 0.0058 Score 7.4652 1 0.0063 Wald 7.3032 1 0.0069 Type 3 Analysis of Effects Wald Effect DF Chi-Square Pr > ChiSq S 1 7.3032 0.0069 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq Intercept 1 -1.7272 0.1357 162.1110 <.0001 S scout 1 -0.6140 0.2272 7.3032 0.0069 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits S scout vs nonscout 0.541 0.347 0.845 The SAS System The LOGISTIC Procedure Association of Predicted Probabilities and Observed Responses Percent Concordant 32.2 Somers' D 0.148 Percent Discordant 17.4 Gamma 0.298 Percent Tied 50.4 Tau-a 0.032 Pairs 68191 c 0.574 The SAS System The LOGISTIC Procedure Model Information Data Set WORK.SCOUT2 Response Variable (Events) y Response Variable (Trials) n Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 3 Number of Observations Used 3 Sum of Frequencies Read 800 Sum of Frequencies Used 800 Response Profile Ordered Binary Total Value Outcome Frequency 1 Event 97 2 Nonevent 703 Class Level Information Design Class Value Variables S low 0 0 medium 1 0 high 0 1 Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Deviance and Pearson Goodness-of-Fit Statistics Criterion Value DF Value/DF Pr > ChiSq Deviance 0.0000 0 . . Pearson 0.0000 0 . . Number of events/trials observations: 3 The SAS System The LOGISTIC Procedure Model Fit Statistics Intercept and Covariates Intercept Log Full Log Criterion Only Likelihood Likelihood AIC 593.053 560.801 20.942 SC 597.738 574.855 34.995 -2 Log L 591.053 554.801 14.942 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 36.2523 2 <.0001 Score 32.8263 2 <.0001 Wald 27.7335 2 <.0001 Type 3 Analysis of Effects Wald Effect DF Chi-Square Pr > ChiSq S 2 27.7335 <.0001 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq Intercept 1 -1.3863 0.1536 81.4848 <.0001 S medium 1 -0.5512 0.2392 5.3080 0.0212 S high 1 -1.8524 0.3571 26.9110 <.0001 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits S medium vs low 0.576 0.361 0.921 S high vs low 0.157 0.078 0.316 The SAS System The LOGISTIC Procedure Association of Predicted Probabilities and Observed Responses Percent Concordant 50.9 Somers' D 0.337 Percent Discordant 17.1 Gamma 0.496 Percent Tied 32.0 Tau-a 0.072 Pairs 68191 c 0.669 The SAS System The LOGISTIC Procedure Model Information Data Set WORK.SCOUT3 Response Variable (Events) y Response Variable (Trials) n Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 6 Number of Observations Used 6 Sum of Frequencies Read 800 Sum of Frequencies Used 800 Response Profile Ordered Binary Total Value Outcome Frequency 1 Event 97 2 Nonevent 703 Class Level Information Design Class Value Variables S low 1 0 medium 0 1 high 0 0 B scout 1 nonscout 0 Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Deviance and Pearson Goodness-of-Fit Statistics Criterion Value DF Value/DF Pr > ChiSq Deviance 0.1543 2 0.0771 0.9258 Pearson 0.1518 2 0.0759 0.9269 Number of events/trials observations: 6 The SAS System The LOGISTIC Procedure Model Fit Statistics Intercept and Covariates Intercept Log Full Log Criterion Only Likelihood Likelihood AIC 593.053 562.793 33.078 SC 597.738 581.531 51.817 -2 Log L 591.053 554.793 25.078 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 36.2604 3 <.0001 Score 32.8340 3 <.0001 Wald 27.7407 3 <.0001 Type 3 Analysis of Effects Wald Effect DF Chi-Square Pr > ChiSq B 1 0.0080 0.9286 S 2 23.3406 <.0001 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq Intercept 1 -3.2214 0.3755 73.6163 <.0001 B scout 1 -0.0225 0.2512 0.0080 0.9286 S low 1 1.8396 0.3842 22.9258 <.0001 S medium 1 1.2937 0.3802 11.5762 0.0007 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits B scout vs nonscout 0.978 0.598 1.600 S low vs high 6.294 2.964 13.366 S medium vs high 3.646 1.731 7.683 The SAS System The LOGISTIC Procedure Association of Predicted Probabilities and Observed Responses Percent Concordant 57.1 Somers' D 0.339 Percent Discordant 23.3 Gamma 0.421 Percent Tied 19.6 Tau-a 0.072 Pairs 68191 c 0.669 The SAS System The LOGISTIC Procedure Model Information Data Set WORK.SCOUT4 Response Variable (Events) y Response Variable (Trials) n Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 6 Number of Observations Used 6 Sum of Frequencies Read 800 Sum of Frequencies Used 800 Response Profile Ordered Binary Total Value Outcome Frequency 1 Event 97 2 Nonevent 703 Class Level Information Design Class Value Variables B scout 1 nonscout 0 S low 0 0 medium 1 0 high 0 1 Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Deviance and Pearson Goodness-of-Fit Statistics Criterion Value DF Value/DF Pr > ChiSq Deviance 0.0000 0 . . Pearson 0.0000 0 . . Number of events/trials observations: 6 The SAS System The LOGISTIC Procedure Model Fit Statistics Intercept and Covariates Intercept Log Full Log Criterion Only Likelihood Likelihood AIC 593.053 566.638 36.924 SC 597.738 594.746 65.032 -2 Log L 591.053 554.638 24.924 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 36.4147 5 <.0001 Score 32.9576 5 <.0001 Wald 27.8071 5 <.0001 Type 3 Analysis of Effects Wald Effect DF Chi-Square Pr > ChiSq B 1 0.0058 0.9392 S 2 9.0032 0.0111 B*S 2 0.1515 0.9270 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq Intercept 1 -1.3922 0.1724 65.2041 <.0001 B scout 1 0.0289 0.3793 0.0058 0.9392 S medium 1 -0.4948 0.2955 2.8048 0.0940 S high 1 -1.9921 0.7394 7.2597 0.0071 B*S scout medium 1 -0.1472 0.5315 0.0767 0.7818 B*S scout high 1 0.1568 0.8893 0.0311 0.8601 Association of Predicted Probabilities and Observed Responses Percent Concordant 57.3 Somers' D 0.343 Percent Discordant 23.1 Gamma 0.426 Percent Tied 19.6 Tau-a 0.073 Pairs 68191 c 0.671 The SAS System The LOGISTIC Procedure Model Information Data Set WORK.SCOUT5 Response Variable (Events) y Response Variable (Trials) n Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 6 Number of Observations Used 6 Sum of Frequencies Read 800 Sum of Frequencies Used 800 Response Profile Ordered Binary Total Value Outcome Frequency 1 Event 97 2 Nonevent 703 Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Deviance and Pearson Goodness-of-Fit Statistics Criterion Value DF Value/DF Pr > ChiSq Deviance 0.0000 0 . . Pearson 0.0000 0 . . Number of events/trials observations: 6 Model Fit Statistics With Covariates Without Log Full Log Criterion Covariates Likelihood Likelihood AIC 1109.035 566.638 36.924 SC 1109.035 594.746 65.032 -2 Log L 1109.035 554.638 24.924 The SAS System The LOGISTIC Procedure Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 554.3971 6 <.0001 Score 473.0913 6 <.0001 Wald 293.7531 6 <.0001 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq x1 1 -1.3922 0.1724 65.2041 <.0001 x2 1 0.0289 0.3793 0.0058 0.9392 x3 1 -1.8871 0.2399 61.8495 <.0001 x4 1 -0.1183 0.3723 0.1009 0.7508 x5 1 -3.3843 0.7190 22.1575 <.0001 x6 1 0.1857 0.8044 0.0533 0.8174 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits x1 0.249 0.177 0.348 x2 1.029 0.489 2.165 x3 0.152 0.095 0.242 x4 0.888 0.428 1.843 x5 0.034 0.008 0.139 x6 1.204 0.249 5.825 Association of Predicted Probabilities and Observed Responses Percent Concordant 57.3 Somers' D 0.343 Percent Discordant 23.1 Gamma 0.426 Percent Tied 19.6 Tau-a 0.073 Pairs 68191 c 0.671