Results: MathLogReg2.sas

Prediction of Performance in First-year Calculus

Predict Passing the course (Y-N) with Logistic Regression

HS variables

The LOGISTIC Procedure

The LOGISTIC Procedure

Model Information

Model Information
Data Set WORK.MATHEX  
Response Variable passed Passed the course
Number of Response Levels 2  
Model binary logit  
Optimization Technique Fisher's scoring  

Observations Summary

Number of Observations Read 579
Number of Observations Used 435

Response Profile

Response Profile
Ordered
Value		 passed Total
Frequency
1 No 178
2 Yes 257

Probability modeled is passed='Yes'.

Note:144 observations were deleted due to missing values for the response or explanatory variables.

Convergence Status

Model Convergence Status
Convergence criterion (GCONV=1E-8) satisfied.

Fit Statistics

Model Fit Statistics
Criterion Intercept Only Intercept and Covariates
AIC 590.611 465.610
SC 594.686 481.912
-2 Log L 588.611 457.610

Global Tests

Testing Global Null Hypothesis: BETA=0
Test Chi-Square DF Pr > ChiSq
Likelihood Ratio 131.0005 3 <.0001
Score 110.3360 3 <.0001
Wald 82.9074 3 <.0001

Parameter Estimates

Analysis of Maximum Likelihood Estimates
Parameter DF Estimate Standard
Error		 Wald
Chi-Square		 Pr > ChiSq
Intercept 1 -15.9885 2.0503 60.8083 <.0001
hsgpa 1 0.1491 0.0331 20.3218 <.0001
hscalc 1 0.0582 0.0126 21.3526 <.0001
hsengl 1 0.00437 0.0164 0.0711 0.7898

Odds Ratios

Odds Ratio Estimates
Effect Point Estimate 95% Wald
Confidence Limits
hsgpa 1.161 1.088 1.238
hscalc 1.060 1.034 1.086
hsengl 1.004 0.973 1.037

Association Statistics

Association of Predicted Probabilities and Observed Responses
Percent Concordant 81.0 Somers' D 0.620
Percent Discordant 19.0 Gamma 0.620
Percent Tied 0.0 Tau-a 0.300
Pairs 45746 c 0.810

Prediction of Performance in First-year Calculus

Predict Passing the course (Y-N) with Logistic Regression

HS gpa and calc, course2 and diagnostic test

The LOGISTIC Procedure

The LOGISTIC Procedure

Model Information

Model Information
Data Set WORK.MATHEX  
Response Variable passed Passed the course
Number of Response Levels 2  
Model binary logit  
Optimization Technique Fisher's scoring  

Observations Summary

Number of Observations Read 579
Number of Observations Used 375

Response Profile

Response Profile
Ordered
Value		 passed Total
Frequency
1 No 141
2 Yes 234

Probability modeled is passed='Yes'.

Note:204 observations were deleted due to missing values for the response or explanatory variables.

Class Level Information

Class Level Information
Class Value Design Variables
course2 Catch-up 1 0
  Elite 0 1
  Mainstrm 0 0

Convergence Status

Model Convergence Status
Convergence criterion (GCONV=1E-8) satisfied.

Fit Statistics

Model Fit Statistics
Criterion Intercept Only Intercept and Covariates
AIC 498.554 379.569
SC 502.481 407.057
-2 Log L 496.554 365.569

Global Tests

Testing Global Null Hypothesis: BETA=0
Test Chi-Square DF Pr > ChiSq
Likelihood Ratio 130.9852 6 <.0001
Score 108.3421 6 <.0001
Wald 79.9399 6 <.0001

Type 3 Tests

Type 3 Analysis of Effects
Effect DF Wald
Chi-Square		 Pr > ChiSq
hsgpa 1 14.6472 0.0001
hscalc 1 18.1207 <.0001
course2 2 0.4383 0.8032
precalc 1 8.5427 0.0035
calc 1 1.6624 0.1973

Parameter Estimates

Analysis of Maximum Likelihood Estimates
Parameter   DF Estimate Standard
Error		 Wald
Chi-Square		 Pr > ChiSq
Intercept   1 -14.8670 2.3187 41.1102 <.0001
hsgpa   1 0.1203 0.0314 14.6472 0.0001
hscalc   1 0.0596 0.0140 18.1207 <.0001
course2 Catch-up 1 0.2860 0.5602 0.2607 0.6097
course2 Elite 1 0.2244 0.5129 0.1915 0.6617
precalc   1 0.2617 0.0895 8.5427 0.0035
calc   1 0.0843 0.0654 1.6624 0.1973

Odds Ratios

Odds Ratio Estimates
Effect Point Estimate 95% Wald
Confidence Limits
hsgpa 1.128 1.060 1.199
hscalc 1.061 1.033 1.091
course2 Catch-up vs Mainstrm 1.331 0.444 3.991
course2 Elite vs Mainstrm 1.252 0.458 3.420
precalc 1.299 1.090 1.548
calc 1.088 0.957 1.237

Association Statistics

Association of Predicted Probabilities and Observed Responses
Percent Concordant 83.5 Somers' D 0.670
Percent Discordant 16.5 Gamma 0.670
Percent Tied 0.0 Tau-a 0.315
Pairs 32994 c 0.835

Wald Test for Contrasts

Contrast Test Results
Contrast DF Wald
Chi-Square		 Pr > ChiSq
Course 2 0.4383 0.8032

Prediction of Performance in First-year Calculus

Predict Passing the course (Y-N) with Logistic Regression

HS gpa and calc, precalc and total score

The LOGISTIC Procedure

The LOGISTIC Procedure

Model Information

Model Information
Data Set WORK.MATHEX  
Response Variable passed Passed the course
Number of Response Levels 2  
Model binary logit  
Optimization Technique Fisher's scoring  

Observations Summary

Number of Observations Read 579
Number of Observations Used 375

Response Profile

Response Profile
Ordered
Value		 passed Total
Frequency
1 No 141
2 Yes 234

Probability modeled is passed='Yes'.

Note:204 observations were deleted due to missing values for the response or explanatory variables.

Convergence Status

Model Convergence Status
Convergence criterion (GCONV=1E-8) satisfied.

Fit Statistics

Model Fit Statistics
Criterion Intercept Only Intercept and Covariates
AIC 498.554 376.007
SC 502.481 395.642
-2 Log L 496.554 366.007

Global Tests

Testing Global Null Hypothesis: BETA=0
Test Chi-Square DF Pr > ChiSq
Likelihood Ratio 130.5468 4 <.0001
Score 108.2737 4 <.0001
Wald 79.7057 4 <.0001

Parameter Estimates

Analysis of Maximum Likelihood Estimates
Parameter DF Estimate Standard
Error		 Wald
Chi-Square		 Pr > ChiSq
Intercept 1 -14.6351 2.2803 41.1914 <.0001
hsgpa 1 0.1181 0.0311 14.4227 0.0001
hscalc 1 0.0592 0.0136 18.9109 <.0001
precalc 1 0.1812 0.1250 2.0999 0.1473
totscore 1 0.0821 0.0650 1.5969 0.2063

Odds Ratios

Odds Ratio Estimates
Effect Point Estimate 95% Wald
Confidence Limits
hsgpa 1.125 1.059 1.196
hscalc 1.061 1.033 1.090
precalc 1.199 0.938 1.532
totscore 1.086 0.956 1.233

Association Statistics

Association of Predicted Probabilities and Observed Responses
Percent Concordant 83.5 Somers' D 0.670
Percent Discordant 16.5 Gamma 0.670
Percent Tied 0.0 Tau-a 0.315
Pairs 32994 c 0.835

Test Statement Results

Linear Hypotheses Testing Results
Label Wald
Chi-Square		 DF Pr > ChiSq
precalc_n_totscore 13.8587 2 0.0010

Prediction of Performance in First-year Calculus

Predict Passing the course (Y-N) with Logistic Regression

Try gender, ethnic and mother tongue controlling for good stuff

The LOGISTIC Procedure

The LOGISTIC Procedure

Model Information

Model Information
Data Set WORK.MATHEX  
Response Variable passed Passed the course
Number of Response Levels 2  
Model binary logit  
Optimization Technique Fisher's scoring  

Observations Summary

Number of Observations Read 579
Number of Observations Used 370

Response Profile

Response Profile
Ordered
Value		 passed Total
Frequency
1 No 138
2 Yes 232

Probability modeled is passed='Yes'.

Note:209 observations were deleted due to missing values for the response or explanatory variables.

Class Level Information

Class Level Information
Class Value Design Variables
ethnic Asian 1 0 0 0 0
  East Indian 0 0 0 0 0
  Eastern European 0 1 0 0 0
  European not Eastern 0 0 1 0 0
  Middle-Eastern and Pakistani 0 0 0 1 0
  Other and DK 0 0 0 0 1

Convergence Status

Model Convergence Status
Convergence criterion (GCONV=1E-8) satisfied.

Fit Statistics

Model Fit Statistics
Criterion Intercept Only Intercept and Covariates
AIC 490.784 379.703
SC 494.698 422.752
-2 Log L 488.784 357.703

Global Tests

Testing Global Null Hypothesis: BETA=0
Test Chi-Square DF Pr > ChiSq
Likelihood Ratio 131.0808 10 <.0001
Score 110.8226 10 <.0001
Wald 80.9993 10 <.0001

Type 3 Tests

Type 3 Analysis of Effects
Effect DF Wald
Chi-Square		 Pr > ChiSq
hsgpa 1 10.3205 0.0013
hscalc 1 24.3884 <.0001
precalc 1 13.5282 0.0002
ethnic 5 4.5364 0.4750
gender 1 0.8343 0.3610
mtongue 1 0.1917 0.6615

Parameter Estimates

Analysis of Maximum Likelihood Estimates
Parameter   DF Estimate Standard
Error		 Wald
Chi-Square		 Pr > ChiSq
Intercept   1 -14.2314 2.4312 34.2655 <.0001
hsgpa   1 0.1043 0.0325 10.3205 0.0013
hscalc   1 0.0687 0.0139 24.3884 <.0001
precalc   1 0.3205 0.0871 13.5282 0.0002
ethnic Asian 1 -0.3522 0.4642 0.5758 0.4480
ethnic Eastern European 1 -0.0314 0.5126 0.0038 0.9512
ethnic European not Eastern 1 0.2889 0.4266 0.4587 0.4982
ethnic Middle-Eastern and Pakistani 1 -0.4521 0.5419 0.6963 0.4040
ethnic Other and DK 1 0.6539 0.8954 0.5334 0.4652
gender   1 0.2465 0.2698 0.8343 0.3610
mtongue   1 -0.1485 0.3391 0.1917 0.6615

Odds Ratios

Odds Ratio Estimates
Effect Point Estimate 95% Wald
Confidence Limits
hsgpa 1.110 1.042 1.183
hscalc 1.071 1.042 1.101
precalc 1.378 1.161 1.634
ethnic Asian vs East Indian 0.703 0.283 1.746
ethnic Eastern European vs East Indian 0.969 0.355 2.647
ethnic European not Eastern vs East Indian 1.335 0.579 3.081
ethnic Middle-Eastern and Pakistani vs East Indian 0.636 0.220 1.840
ethnic Other and DK vs East Indian 1.923 0.333 11.120
gender 1.279 0.754 2.171
mtongue 0.862 0.443 1.676

Association Statistics

Association of Predicted Probabilities and Observed Responses
Percent Concordant 83.7 Somers' D 0.674
Percent Discordant 16.3 Gamma 0.674
Percent Tied 0.0 Tau-a 0.316
Pairs 32016 c 0.837

Coefficients for Contrasts

Coefficients of Contrast Demographics
Parameter Row1 Row2 Row3 Row4 Row5 Row6 Row7
Intercept 0 0 0 0 0 0 0
hsgpa 0 0 0 0 0 0 0
hscalc 0 0 0 0 0 0 0
precalc 0 0 0 0 0 0 0
ethnicAsian 1 0 0 0 0 0 0
ethnicEastern_European 0 1 0 0 0 0 0
ethnicEuropean_not_Eastern 0 0 1 0 0 0 0
ethnicMiddle_Eastern__and_Pakist 0 0 0 1 0 0 0
ethnicOther___and_DK 0 0 0 0 1 0 0
gender 0 0 0 0 0 1 0
mtongue 0 0 0 0 0 0 1

Wald Test for Contrasts

Contrast Test Results
Contrast DF Wald
Chi-Square		 Pr > ChiSq
Demographics 7 6.0125 0.5383