Variable Label N Mean Std Dev Minimum Maximum ャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャ notarget =1 if does not target a species/species group 11257 0 0 0 0 dolphin =1 if targets species 11257 1.9371058 0.2427834 1.0000000 2.0000000 mackerel 11257 1.8958870 0.3054206 1.0000000 2.0000000 reddrum =1 if targets species 11257 1.7957715 0.4031546 1.0000000 2.0000000 snapgrou =1 if targets species 11257 1.8505819 0.3565160 1.0000000 2.0000000 ャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャャ The SAS System 09:36 Thursday, April 10, 2008 1052 The LOGISTIC Procedure Model Information Data Set WORK.TARGET Response Variable dolphin =1 if targets species Number of Response Levels 2 Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 11257 Number of Observations Used 11257 Response Profile Ordered Total Value dolphin Frequency 1 1 708 2 2 10549 Probability modeled is dolphin=1. Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. The SAS System 09:36 Thursday, April 10, 2008 1053 The LOGISTIC Procedure Model Fit Statistics Intercept Intercept and Criterion Only Covariates AIC 5289.590 3843.138 SC 5296.918 3909.097 -2 Log L 5287.590 3825.138 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 1462.4515 8 <.0001 Score 1284.5693 8 <.0001 Wald 602.0546 8 <.0001 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq Intercept 1 -0.7634 0.1121 46.3401 <.0001 YEARFISH 1 -0.00920 0.00298 9.5262 0.0020 boatown 1 0.0994 0.0945 1.1065 0.2928 charter 1 0.9006 0.1426 39.8937 <.0001 FFDAYS2 1 -0.0285 0.00569 25.0467 <.0001 wave4 1 -0.4461 0.0997 20.0067 <.0001 wave5 1 -1.5450 0.1273 147.2625 <.0001 wave6 1 -1.4920 0.1393 114.7950 <.0001 gulf 1 -4.2117 0.2289 338.5586 <.0001 The SAS System 09:36 Thursday, April 10, 2008 1054 The LOGISTIC Procedure Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits YEARFISH 0.991 0.985 0.997 boatown 1.104 0.918 1.329 charter 2.461 1.861 3.254 FFDAYS2 0.972 0.961 0.983 wave4 0.640 0.526 0.778 wave5 0.213 0.166 0.274 wave6 0.225 0.171 0.296 gulf 0.015 0.009 0.023 Association of Predicted Probabilities and Observed Responses Percent Concordant 87.7 Somers' D 0.760 Percent Discordant 11.7 Gamma 0.764 Percent Tied 0.5 Tau-a 0.090 Pairs 7468692 c 0.880 The SAS System 09:36 Thursday, April 10, 2008 1055 The LOGISTIC Procedure Model Information Data Set WORK.TARGET Response Variable mackerel Number of Response Levels 2 Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 11257 Number of Observations Used 11257 Response Profile Ordered Total Value mackerel Frequency 1 1 1172 2 2 10085 Probability modeled is mackerel=1. Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. The SAS System 09:36 Thursday, April 10, 2008 1056 The LOGISTIC Procedure Model Fit Statistics Intercept Intercept and Criterion Only Covariates AIC 7522.290 7380.116 SC 7529.619 7446.075 -2 Log L 7520.290 7362.116 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 158.1740 8 <.0001 Score 162.6891 8 <.0001 Wald 158.6285 8 <.0001 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq Intercept 1 -1.7177 0.0920 348.2526 <.0001 YEARFISH 1 -0.00160 0.00212 0.5677 0.4512 boatown 1 -0.4337 0.0661 43.0556 <.0001 charter 1 0.2793 0.1087 6.6061 0.0102 FFDAYS2 1 0.0156 0.00276 31.7384 <.0001 wave4 1 0.1706 0.0815 4.3868 0.0362 wave5 1 -0.3227 0.0885 13.2955 0.0003 wave6 1 -0.1772 0.0917 3.7345 0.0533 gulf 1 -0.4113 0.0623 43.5328 <.0001 The SAS System 09:36 Thursday, April 10, 2008 1057 The LOGISTIC Procedure Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits YEARFISH 0.998 0.994 1.003 boatown 0.648 0.569 0.738 charter 1.322 1.069 1.636 FFDAYS2 1.016 1.010 1.021 wave4 1.186 1.011 1.391 wave5 0.724 0.609 0.861 wave6 0.838 0.700 1.003 gulf 0.663 0.587 0.749 Association of Predicted Probabilities and Observed Responses Percent Concordant 60.4 Somers' D 0.224 Percent Discordant 38.0 Gamma 0.228 Percent Tied 1.6 Tau-a 0.042 Pairs 11819620 c 0.612 The SAS System 09:36 Thursday, April 10, 2008 1058 The LOGISTIC Procedure Model Information Data Set WORK.TARGET Response Variable reddrum =1 if targets species Number of Response Levels 2 Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 11257 Number of Observations Used 11257 Response Profile Ordered Total Value reddrum Frequency 1 1 2299 2 2 8958 Probability modeled is reddrum=1. Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. The SAS System 09:36 Thursday, April 10, 2008 1059 The LOGISTIC Procedure Model Fit Statistics Intercept Intercept and Criterion Only Covariates AIC 11398.784 11184.288 SC 11406.113 11250.247 -2 Log L 11396.784 11166.288 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 230.4956 8 <.0001 Score 217.4079 8 <.0001 Wald 211.2628 8 <.0001 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq Intercept 1 -1.9408 0.0804 582.6373 <.0001 YEARFISH 1 -0.00603 0.00163 13.6725 0.0002 boatown 1 0.4221 0.0554 57.9619 <.0001 charter 1 -0.7030 0.1128 38.8488 <.0001 FFDAYS2 1 -0.00189 0.00246 0.5919 0.4417 wave4 1 0.2964 0.0684 18.7910 <.0001 wave5 1 0.5528 0.0660 70.1380 <.0001 wave6 1 0.4332 0.0709 37.3007 <.0001 gulf 1 0.2518 0.0487 26.6935 <.0001 The SAS System 09:36 Thursday, April 10, 2008 1060 The LOGISTIC Procedure Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits YEARFISH 0.994 0.991 0.997 boatown 1.525 1.368 1.700 charter 0.495 0.397 0.618 FFDAYS2 0.998 0.993 1.003 wave4 1.345 1.176 1.538 wave5 1.738 1.527 1.978 wave6 1.542 1.342 1.772 gulf 1.286 1.169 1.415 Association of Predicted Probabilities and Observed Responses Percent Concordant 59.5 Somers' D 0.199 Percent Discordant 39.5 Gamma 0.202 Percent Tied 1.0 Tau-a 0.065 Pairs 20594442 c 0.600 The SAS System 09:36 Thursday, April 10, 2008 1061 The LOGISTIC Procedure Model Information Data Set WORK.TARGET Response Variable snapgrou =1 if targets species Number of Response Levels 2 Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 11257 Number of Observations Used 11257 Response Profile Ordered Total Value snapgrou Frequency 1 1 1682 2 2 9575 Probability modeled is snapgrou=1. Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. The SAS System 09:36 Thursday, April 10, 2008 1062 The LOGISTIC Procedure Model Fit Statistics Intercept Intercept and Criterion Only Covariates AIC 9496.119 8990.889 SC 9503.448 9056.848 -2 Log L 9494.119 8972.889 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 521.2298 8 <.0001 Score 563.9329 8 <.0001 Wald 502.5330 8 <.0001 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq Intercept 1 -2.2145 0.0902 602.8425 <.0001 YEARFISH 1 -0.00443 0.00187 5.6167 0.0178 boatown 1 -0.1450 0.0596 5.9220 0.0150 charter 1 1.1253 0.0812 191.8707 <.0001 FFDAYS2 1 -0.0102 0.00314 10.6113 0.0011 wave4 1 0.0917 0.0751 1.4914 0.2220 wave5 1 -0.0799 0.0763 1.0992 0.2944 wave6 1 0.2194 0.0773 8.0615 0.0045 gulf 1 0.8685 0.0611 202.1282 <.0001 The SAS System 09:36 Thursday, April 10, 2008 1063 The LOGISTIC Procedure Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits YEARFISH 0.996 0.992 0.999 boatown 0.865 0.770 0.972 charter 3.081 2.628 3.613 FFDAYS2 0.990 0.984 0.996 wave4 1.096 0.946 1.270 wave5 0.923 0.795 1.072 wave6 1.245 1.070 1.449 gulf 2.383 2.114 2.687 Association of Predicted Probabilities and Observed Responses Percent Concordant 65.4 Somers' D 0.320 Percent Discordant 33.4 Gamma 0.324 Percent Tied 1.1 Tau-a 0.081 Pairs 16105150 c 0.660