CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 1 THE DATA AS SAS SEES IT 23:54 Wednesday, November 9, 2005 subj yy bmicase famcase phaccase bmictrl famctrl phacctrl bmi fam sed 1 1 22.1 1 1 26.7 0 1 -4.6 1 0 2 1 31.3 0 0 24.4 0 1 6.9 0 1 3 1 33.8 1 0 29.4 0 0 4.4 1 0 4 1 33.7 1 1 26.0 0 0 7.7 1 -1 5 1 23.1 1 1 24.2 1 0 -1.1 0 -1 6 1 26.8 1 0 29.7 0 0 -2.9 1 0 7 1 32.3 1 0 30.2 0 1 2.1 1 1 8 1 31.4 1 0 23.4 0 1 8.0 1 1 9 1 37.6 1 0 42.4 0 0 -4.8 1 0 10 1 32.4 1 0 25.8 0 0 6.6 1 0 11 1 29.1 0 1 39.8 0 1 -10.7 0 0 12 1 28.6 0 1 31.6 0 0 -3.0 0 -1 13 1 35.9 0 0 21.8 1 1 14.1 -1 1 14 1 30.4 0 0 24.2 0 1 6.2 0 1 15 1 39.8 0 0 27.8 1 1 12.0 -1 1 16 1 43.3 1 0 37.5 1 1 5.8 0 1 17 1 32.5 0 0 27.9 1 1 4.6 -1 1 18 1 28.7 0 1 25.3 1 0 3.4 -1 -1 19 1 30.3 0 0 31.3 0 1 -1.0 0 1 20 1 32.5 1 0 34.5 1 1 -2.0 0 1 21 1 32.5 1 0 25.4 0 1 7.1 1 1 22 1 21.6 1 1 27.0 1 1 -5.4 0 0 23 1 24.4 0 1 31.1 0 0 -6.7 0 -1 24 1 46.7 1 0 27.3 0 1 19.4 1 1 25 1 28.6 1 1 24.0 0 0 4.6 1 -1 26 1 29.7 0 0 33.5 0 0 -3.8 0 0 27 1 29.6 0 1 20.7 0 0 8.9 0 -1 28 1 22.8 0 0 29.2 1 1 -6.4 -1 1 29 1 34.8 1 0 30.0 0 1 4.8 1 1 30 1 37.3 1 0 26.5 0 0 10.8 1 0 CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 2 VISUALIZING THE DATA: VERTICAL BAR CHARTS: POSITIVE VALUES ARE MORE COMMON AMONG THE AFFECTEDS 23:54 Wednesday, November 9, 2005 Frequency 13 + ***** | ***** 12 + ***** | ***** 11 + ***** | ***** 10 + ***** | ***** 9 + ***** | ***** 8 + ***** | ***** 7 + ***** | ***** 6 + ***** ***** ***** | ***** ***** ***** 5 + ***** ***** ***** | ***** ***** ***** 4 + ***** ***** ***** | ***** ***** ***** 3 + ***** ***** ***** ***** | ***** ***** ***** ***** 2 + ***** ***** ***** ***** | ***** ***** ***** ***** 1 + ***** ***** ***** ***** ***** ***** | ***** ***** ***** ***** ***** ***** ------------------------------------------------------------------ -12 -6 0 6 12 18 Body Mass Index difference CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 3 VISUALIZING THE DATA: VERTICAL BAR CHARTS: POSITIVE VALUES ARE MORE COMMON AMONG THE AFFECTEDS 23:54 Wednesday, November 9, 2005 Frequency 13 + ***** | ***** 12 + ***** ***** | ***** ***** 11 + ***** ***** | ***** ***** 10 + ***** ***** | ***** ***** 9 + ***** ***** | ***** ***** 8 + ***** ***** | ***** ***** 7 + ***** ***** | ***** ***** 6 + ***** ***** | ***** ***** 5 + ***** ***** ***** | ***** ***** ***** 4 + ***** ***** ***** | ***** ***** ***** 3 + ***** ***** ***** | ***** ***** ***** 2 + ***** ***** ***** | ***** ***** ***** 1 + ***** ***** ***** | ***** ***** ***** -------------------------------------------------------------------- -1.0 -0.5 0.0 0.5 1.0 Family history difference of AODM (1=Yes,0=No) CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 4 VISUALIZING THE DATA: VERTICAL BAR CHARTS: POSITIVE VALUES ARE MORE COMMON AMONG THE AFFECTEDS 23:54 Wednesday, November 9, 2005 Frequency 14 + ***** | ***** 13 + ***** | ***** 12 + ***** | ***** 11 + ***** | ***** 10 + ***** | ***** 9 + ***** ***** | ***** ***** 8 + ***** ***** | ***** ***** 7 + ***** ***** ***** | ***** ***** ***** 6 + ***** ***** ***** | ***** ***** ***** 5 + ***** ***** ***** | ***** ***** ***** 4 + ***** ***** ***** | ***** ***** ***** 3 + ***** ***** ***** | ***** ***** ***** 2 + ***** ***** ***** | ***** ***** ***** 1 + ***** ***** ***** | ***** ***** ***** -------------------------------------------------------------------- -1.0 -0.5 0.0 0.5 1.0 Sedentary (No physical activity) difference (1=Yes,0=No) CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 5 LOGISTIC REGRESSION WITH NO INTERCEPT 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Model Information Data Set WORK.CCBMI Response Variable yy Number of Response Levels 1 Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 30 Number of Observations Used 30 Response Profile Ordered Total Value yy Frequency 1 1 30 Probability modeled is yy=1. Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Without With Criterion Covariates Covariates AIC 41.589 39.123 SC 41.589 43.327 -2 Log L 41.589 33.123 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 8.4658 3 0.0373 Score 7.4691 3 0.0584 Wald 6.0094 3 0.1112 CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 6 LOGISTIC REGRESSION WITH NO INTERCEPT 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq bmi 1 0.0901 0.0652 1.9059 0.1674 fam 1 0.9681 0.5880 2.7109 0.0997 sed 1 0.5632 0.5410 1.0836 0.2979 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits bmi 1.094 0.963 1.244 fam 2.633 0.832 8.336 sed 1.756 0.608 5.071 NOTE: Since there is only one response level, measures of association between the observed and predicted values were not calculated. CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 7 LOGISTIC REGRESSION WITH STEPWISE MODEL REGRESSION SLE=0.10 for entry, SLS=0.10 for removal 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Model Information Data Set WORK.CCBMI Response Variable yy Number of Response Levels 1 Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 30 Number of Observations Used 30 Response Profile Ordered Total Value yy Frequency 1 1 30 Probability modeled is yy=1. Stepwise Selection Procedure Step 0. No covariates. Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. -2 Log L = 41.589 Residual Chi-Square Test Chi-Square DF Pr > ChiSq 7.4691 3 0.0584 Step 1. Effect bmi entered: Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 8 LOGISTIC REGRESSION WITH STEPWISE MODEL REGRESSION SLE=0.10 for entry, SLS=0.10 for removal 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Model Fit Statistics Without With Criterion Covariates Covariates AIC 41.589 38.925 SC 41.589 40.326 -2 Log L 41.589 36.925 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 4.6635 1 0.0308 Score 4.3409 1 0.0372 Wald 3.8284 1 0.0504 Residual Chi-Square Test Chi-Square DF Pr > ChiSq 3.6341 2 0.1625 NOTE: No effects for the model in Step 1 are removed. NOTE: No (additional) effects met the 0.1 significance level for entry into the model. Summary of Stepwise Selection Effect Number Score Wald Step Entered Removed DF In Chi-Square Chi-Square Pr > ChiSq 1 bmi 1 1 4.3409 0.0372 Summary of Stepwise Selection Variable Step Label 1 Body Mass Index difference CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 9 LOGISTIC REGRESSION WITH STEPWISE MODEL REGRESSION SLE=0.10 for entry, SLS=0.10 for removal 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq bmi 1 0.1181 0.0603 3.8284 0.0504 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits bmi 1.125 1.000 1.267 NOTE: Since there is only one response level, measures of association between the observed and predicted values were not calculated. CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 10 LOGISTIC REGRESSION FOR BMI ONLY WITH NO INTERCEPT 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Model Information Data Set WORK.CCBMI Response Variable yy Number of Response Levels 1 Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 30 Number of Observations Used 30 Response Profile Ordered Total Value yy Frequency 1 1 30 Probability modeled is yy=1. Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Without With Criterion Covariates Covariates AIC 41.589 38.925 SC 41.589 40.326 -2 Log L 41.589 36.925 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 4.6635 1 0.0308 Score 4.3409 1 0.0372 Wald 3.8284 1 0.0504 CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 11 LOGISTIC REGRESSION FOR BMI ONLY WITH NO INTERCEPT 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq bmi 1 0.1181 0.0603 3.8284 0.0504 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits bmi 1.125 1.000 1.267 NOTE: Since there is only one response level, measures of association between the observed and predicted values were not calculated. CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 12 LOGISTIC REGRESSION WITH BACKWARDS STEPWISE REGRESSION ADDING THREE INTERACTIONS 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Model Information Data Set WORK.CCBMI Response Variable yy Number of Response Levels 1 Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 30 Number of Observations Used 30 Response Profile Ordered Total Value yy Frequency 1 1 30 Probability modeled is yy=1. Backward Elimination Procedure Step 0. The following effects were entered: bmi fam sed bmifam bmised famsed Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Without With Criterion Covariates Covariates AIC 41.589 42.944 SC 41.589 51.351 -2 Log L 41.589 30.944 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 10.6446 6 0.1000 Score 9.1107 6 0.1674 Wald 6.6989 6 0.3496 CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 13 LOGISTIC REGRESSION WITH BACKWARDS STEPWISE REGRESSION ADDING THREE INTERACTIONS 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq bmi 1 0.0480 0.0783 0.3761 0.5397 fam 1 1.2396 0.7751 2.5578 0.1098 sed 1 0.3231 0.6035 0.2865 0.5924 bmifam 1 0.0396 0.1157 0.1172 0.7321 bmised 1 0.1220 0.1102 1.2253 0.2683 famsed 1 -1.0006 0.9658 1.0734 0.3002 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits bmi 1.049 0.900 1.223 fam 3.454 0.756 15.779 sed 1.381 0.423 4.508 bmifam 1.040 0.829 1.305 bmised 1.130 0.910 1.402 famsed 0.368 0.055 2.441 NOTE: Since there is only one response level, measures of association between the observed and predicted values were not calculated. Analysis of Effects Eligible for Removal Wald Effect DF Chi-Square Pr > ChiSq bmi 1 0.3761 0.5397 fam 1 2.5578 0.1098 sed 1 0.2865 0.5924 bmifam 1 0.1172 0.7321 bmised 1 1.2253 0.2683 famsed 1 1.0734 0.3002 Step 1. Effect bmifam is removed: Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 14 LOGISTIC REGRESSION WITH BACKWARDS STEPWISE REGRESSION ADDING THREE INTERACTIONS 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Model Fit Statistics Without With Criterion Covariates Covariates AIC 41.589 41.062 SC 41.589 48.068 -2 Log L 41.589 31.062 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 10.5264 5 0.0616 Score 9.0928 5 0.1054 Wald 6.9771 5 0.2223 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq bmi 1 0.0529 0.0767 0.4759 0.4903 fam 1 1.3277 0.7367 3.2479 0.0715 sed 1 0.3438 0.6035 0.3245 0.5689 bmised 1 0.1068 0.0993 1.1573 0.2820 famsed 1 -0.9361 0.9392 0.9934 0.3189 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits bmi 1.054 0.907 1.225 fam 3.772 0.890 15.985 sed 1.410 0.432 4.603 bmised 1.113 0.916 1.352 famsed 0.392 0.062 2.471 NOTE: Since there is only one response level, measures of association between the observed and predicted values were not calculated. Residual Chi-Square Test Chi-Square DF Pr > ChiSq 0.1177 1 0.7315 CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 15 LOGISTIC REGRESSION WITH BACKWARDS STEPWISE REGRESSION ADDING THREE INTERACTIONS 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Analysis of Effects Eligible for Removal Wald Effect DF Chi-Square Pr > ChiSq bmi 1 0.4759 0.4903 fam 1 3.2479 0.0715 sed 1 0.3245 0.5689 bmised 1 1.1573 0.2820 famsed 1 0.9934 0.3189 Step 2. Effect sed is removed: Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Without With Criterion Covariates Covariates AIC 41.589 39.386 SC 41.589 44.991 -2 Log L 41.589 31.386 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 10.2031 4 0.0371 Score 8.7645 4 0.0673 Wald 6.7765 4 0.1482 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq bmi 1 0.0609 0.0750 0.6595 0.4167 fam 1 1.2829 0.7228 3.1504 0.0759 bmised 1 0.1273 0.0925 1.8919 0.1690 famsed 1 -0.8869 0.9261 0.9172 0.3382 CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 16 LOGISTIC REGRESSION WITH BACKWARDS STEPWISE REGRESSION ADDING THREE INTERACTIONS 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits bmi 1.063 0.918 1.231 fam 3.607 0.875 14.874 bmised 1.136 0.947 1.362 famsed 0.412 0.067 2.530 NOTE: Since there is only one response level, measures of association between the observed and predicted values were not calculated. Residual Chi-Square Test Chi-Square DF Pr > ChiSq 0.4590 2 0.7949 Analysis of Effects Eligible for Removal Wald Effect DF Chi-Square Pr > ChiSq bmi 1 0.6595 0.4167 fam 1 3.1504 0.0759 bmised 1 1.8919 0.1690 famsed 1 0.9172 0.3382 Step 3. Effect bmi is removed: Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Without With Criterion Covariates Covariates AIC 41.589 38.041 SC 41.589 42.245 -2 Log L 41.589 32.041 CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 17 LOGISTIC REGRESSION WITH BACKWARDS STEPWISE REGRESSION ADDING THREE INTERACTIONS 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 9.5475 3 0.0228 Score 8.3369 3 0.0395 Wald 6.6400 3 0.0843 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq fam 1 1.4233 0.7077 4.0450 0.0443 bmised 1 0.1572 0.0831 3.5835 0.0584 famsed 1 -1.0444 0.8859 1.3898 0.2384 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits fam 4.151 1.037 16.616 bmised 1.170 0.994 1.377 famsed 0.352 0.062 1.998 NOTE: Since there is only one response level, measures of association between the observed and predicted values were not calculated. Residual Chi-Square Test Chi-Square DF Pr > ChiSq 1.1124 3 0.7741 Analysis of Effects Eligible for Removal Wald Effect DF Chi-Square Pr > ChiSq fam 1 4.0450 0.0443 bmised 1 3.5835 0.0584 famsed 1 1.3898 0.2384 Step 4. Effect famsed is removed: CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 18 LOGISTIC REGRESSION WITH BACKWARDS STEPWISE REGRESSION ADDING THREE INTERACTIONS 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Without With Criterion Covariates Covariates AIC 41.589 37.648 SC 41.589 40.451 -2 Log L 41.589 33.648 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 7.9406 2 0.0189 Score 6.7954 2 0.0335 Wald 5.4315 2 0.0662 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq fam 1 1.1028 0.5975 3.4062 0.0650 bmised 1 0.1448 0.0783 3.4253 0.0642 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits fam 3.012 0.934 9.717 bmised 1.156 0.991 1.347 NOTE: Since there is only one response level, measures of association between the observed and predicted values were not calculated. Residual Chi-Square Test Chi-Square DF Pr > ChiSq 2.5936 4 0.6279 CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 19 LOGISTIC REGRESSION WITH BACKWARDS STEPWISE REGRESSION ADDING THREE INTERACTIONS 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Analysis of Effects Eligible for Removal Wald Effect DF Chi-Square Pr > ChiSq fam 1 3.4062 0.0650 bmised 1 3.4253 0.0642 NOTE: No (additional) effects met the 0.1 significance level for removal from the model. Summary of Backward Elimination Effect Number Wald Step Removed DF In Chi-Square Pr > ChiSq 1 bmifam 1 5 0.1172 0.7321 2 sed 1 4 0.3245 0.5689 3 bmi 1 3 0.6595 0.4167 4 famsed 1 2 1.3898 0.2384 Summary of Backward Elimination Variable Step Label 1 2 Sedentary (No physical activity) difference (1=Yes,0=No) 3 Body Mass Index difference 4 CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 20 LOGISTIC REGRESSION FOR FAM BMISED WITH NO INTERCEPT 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Model Information Data Set WORK.CCBMI Response Variable yy Number of Response Levels 1 Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 30 Number of Observations Used 30 Response Profile Ordered Total Value yy Frequency 1 1 30 Probability modeled is yy=1. Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Without With Criterion Covariates Covariates AIC 41.589 37.648 SC 41.589 40.451 -2 Log L 41.589 33.648 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 7.9406 2 0.0189 Score 6.7954 2 0.0335 Wald 5.4315 2 0.0662 CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 21 LOGISTIC REGRESSION FOR FAM BMISED WITH NO INTERCEPT 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq fam 1 1.1028 0.5975 3.4062 0.0650 bmised 1 0.1448 0.0783 3.4253 0.0642 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits fam 3.012 0.934 9.717 bmised 1.156 0.991 1.347 NOTE: Since there is only one response level, measures of association between the observed and predicted values were not calculated. CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 22 LOGISTIC REGRESSION FOR BMISED ONLY WITH NO INTERCEPT 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Model Information Data Set WORK.CCBMI Response Variable yy Number of Response Levels 1 Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 30 Number of Observations Used 30 Response Profile Ordered Total Value yy Frequency 1 1 30 Probability modeled is yy=1. Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Without With Criterion Covariates Covariates AIC 41.589 39.633 SC 41.589 41.034 -2 Log L 41.589 37.633 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 3.9556 1 0.0467 Score 3.6077 1 0.0575 Wald 3.0737 1 0.0796 CASE-CONTROL LOGISTIC REGRESSION - Type 2 Diabetes - YOURNAME 23 LOGISTIC REGRESSION FOR BMISED ONLY WITH NO INTERCEPT 23:54 Wednesday, November 9, 2005 The LOGISTIC Procedure Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq bmised 1 0.1284 0.0732 3.0737 0.0796 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits bmised 1.137 0.985 1.312 NOTE: Since there is only one response level, measures of association between the observed and predicted values were not calculated.