The MEANS Procedure

Variable Variance
Slope12
Slope26
time1
time1sq
0.1391185
2.2258953
2.9214876
79.2696970



1a: Piecewise Unconditional Model - Random Early/Later Practice Slopes

The Mixed Procedure

Model Information
Data Set WORK.TRIMMED
Dependent Variable nm3rt
Covariance Structures Unstructured, Variance Components
Subject Effects ID, ID
Estimation Method REML
Residual Variance Method Parameter
Fixed Effects SE Method Model-Based
Degrees of Freedom Method Satterthwaite

Dimensions
Covariance Parameters 7
Columns in X 3
Columns in Z per Subject 3
Subjects 101
Max Obs per Subject 6

Number of Observations
Number of Observations Read 606
Number of Observations Used 606
Number of Observations Not Used 0

Iteration History
Iteration Evaluations -2 Res Log Like Criterion
0 1 9188.48345679  
1 1 8275.37431715 0.00000000

Convergence criteria met.

Estimated G Correlation Matrix
Row Effect Participant ID Col1 Col2 Col3
1 Intercept 101 1.0000 -0.4025 -0.3902
2 Slope12 101 -0.4025 1.0000 -0.1293
3 Slope26 101 -0.3902 -0.1293 1.0000

Covariance Parameter Estimates
Cov Parm Subject Estimate Standard Error Z Value Pr Z
UN(1,1) ID 284312 42731 6.65 <.0001
UN(2,1) ID -54270 18230 -2.98 0.0029
UN(2,2) ID 63954 13244 4.83 <.0001
UN(3,1) ID -10644 3791.26 -2.81 0.0050
UN(3,2) ID -1672.30 2097.03 -0.80 0.4252
UN(3,3) ID 2617.28 636.48 4.11 <.0001
session ID 17673 1435.84 12.31 <.0001

Fit Statistics
-2 Res Log Likelihood 8275.4
AIC (Smaller is Better) 8289.4
AICC (Smaller is Better) 8289.6
BIC (Smaller is Better) 8307.7

Null Model Likelihood Ratio Test
DF Chi-Square Pr > ChiSq
6 913.11 <.0001

Information Criteria
Neg2LogLike Parms AIC AICC HQIC BIC CAIC
8275.4 7 8289.4 8289.6 8296.8 8307.7 8314.7

Solution for Fixed Effects
Effect Estimate Standard Error DF t Value Pr > |t|
Intercept 1961.89 54.6805 100 35.88 <.0001
Slope12 -163.64 30.2188 100 -5.42 <.0001
Slope26 -32.8932 6.5888 100 -4.99 <.0001

Type 3 Tests of Fixed Effects
Effect Num DF Den DF F Value Pr > F
Slope12 1 100 29.33 <.0001
Slope26 1 100 24.92 <.0001



The CORR Procedure

1 With Variables: nm3rt
1 Variables: Pred

Simple Statistics
Variable N Mean Std Dev Sum Minimum Maximum Label
nm3rt 606 1771 494.08763 1073045 917.66667 4159 Number Match 3 RT
Pred 606 1771 95.54686 1073045 1667 1962 Predicted Mean

Pearson Correlation Coefficients, N = 606
Prob > |r| under H0: Rho=0
  Pred
nm3rt
Number Match 3 RT
0.19338
<.0001



1b: Add Fixed Effects for Age on Intercept, Slope12, and Slope26

The Mixed Procedure

Model Information
Data Set WORK.TRIMMED
Dependent Variable nm3rt
Covariance Structures Unstructured, Variance Components
Subject Effects ID, ID
Estimation Method REML
Residual Variance Method Parameter
Fixed Effects SE Method Model-Based
Degrees of Freedom Method Satterthwaite

Dimensions
Covariance Parameters 7
Columns in X 6
Columns in Z per Subject 3
Subjects 101
Max Obs per Subject 6

Number of Observations
Number of Observations Read 606
Number of Observations Used 606
Number of Observations Not Used 0

Iteration History
Iteration Evaluations -2 Res Log Like Criterion
0 1 9129.00024109  
1 1 8251.03576285 0.00000000

Convergence criteria met.

Estimated G Correlation Matrix
Row Effect Participant ID Col1 Col2 Col3
1 Intercept 101 1.0000 -0.3687 -0.3603
2 Slope12 101 -0.3687 1.0000 -0.1651
3 Slope26 101 -0.3603 -0.1651 1.0000

Covariance Parameter Estimates
Cov Parm Subject Estimate Standard Error Z Value Pr Z
UN(1,1) ID 254286 38681 6.57 <.0001
UN(2,1) ID -46576 17140 -2.72 0.0066
UN(2,2) ID 62742 13139 4.78 <.0001
UN(3,1) ID -9251.94 3583.89 -2.58 0.0098
UN(3,2) ID -2106.57 2100.78 -1.00 0.3160
UN(3,3) ID 2593.60 636.25 4.08 <.0001
session ID 17673 1435.84 12.31 <.0001

Fit Statistics
-2 Res Log Likelihood 8251.0
AIC (Smaller is Better) 8265.0
AICC (Smaller is Better) 8265.2
BIC (Smaller is Better) 8283.3

Null Model Likelihood Ratio Test
DF Chi-Square Pr > ChiSq
6 877.96 <.0001

Information Criteria
Neg2LogLike Parms AIC AICC HQIC BIC CAIC
8251.0 7 8265.0 8265.2 8272.4 8283.3 8290.3

Solution for Fixed Effects
Effect Estimate Standard Error DF t Value Pr > |t|
Intercept 1966.86 51.9106 99 37.89 <.0001
Slope12 -164.91 30.0311 99 -5.49 <.0001
Slope26 -33.1182 6.5734 99 -5.04 <.0001
Age80 29.7804 8.5822 99 3.47 0.0008
Slope12*Age80 -7.5810 4.9650 99 -1.53 0.1300
Slope26*Age80 -1.3499 1.0868 99 -1.24 0.2171

Type 3 Tests of Fixed Effects
Effect Num DF Den DF F Value Pr > F
Slope12 1 99 30.15 <.0001
Slope26 1 99 25.38 <.0001
Age80 1 99 12.04 0.0008
Slope12*Age80 1 99 2.33 0.1300
Slope26*Age80 1 99 1.54 0.2171

Estimates
Label Estimate Standard Error DF t Value Pr > |t|
Slope 1-2 for Age 80 -164.91 30.0311 99 -5.49 <.0001
Slope 1-2 for Age 90 -240.72 58.7292 99 -4.10 <.0001
Slope 2-6 for Age 80 -33.1182 6.5734 99 -5.04 <.0001
Slope 2-6 for Age 90 -46.6173 12.8551 99 -3.63 0.0005
Age Effect at Session 1 29.7804 8.5822 99 3.47 0.0008
Age Effect at Session 2 22.1993 7.9689 99 2.79 0.0064
Age Effect at Session 3 20.8494 7.5245 99 2.77 0.0067
Age Effect at Session 4 19.4995 7.2176 99 2.70 0.0081
Age Effect at Session 5 18.1496 7.0663 99 2.57 0.0117
Age Effect at Session 6 16.7997 7.0805 99 2.37 0.0196

Contrasts
Label Num DF Den DF Chi-Square F Value Pr > ChiSq Pr > F
DF=3 Wald Test for Effects of Age 3 99 12.23 4.08 0.0066 0.0089



Total R2 (% Reduction) for PredPUnc vs. PredPAge

Name PredCorr TotalR2 TotalR2Diff
PredPUnc 0.19338 0.03740 .
PredPAge 0.32795 0.10755 0.070156



PsuedoR2 (% Reduction) for CovPUnc vs. CovPAge

Name CovParm Subject Estimate StdErr ZValue ProbZ PseudoR2
CovPUnc UN(1,1) ID 284312 42731 6.65 <.0001 .
CovPUnc UN(2,2) ID 63954 13244 4.83 <.0001 .
CovPUnc UN(3,3) ID 2617.28 636.48 4.11 <.0001 .
CovPUnc session ID 17673 1435.84 12.31 <.0001 .
CovPAge UN(1,1) ID 254286 38681 6.57 <.0001 0.10561
CovPAge UN(2,2) ID 62742 13139 4.78 <.0001 0.01895
CovPAge UN(3,3) ID 2593.60 636.25 4.08 <.0001 0.00905
CovPAge session ID 17673 1435.84 12.31 <.0001 0.00000



1c: Keep Age, Add Fixed Effects for Reasoning on Intercept, Slope12, and Slope26

The Mixed Procedure

Model Information
Data Set WORK.TRIMMED
Dependent Variable nm3rt
Covariance Structures Unstructured, Variance Components
Subject Effects ID, ID
Estimation Method REML
Residual Variance Method Parameter
Fixed Effects SE Method Model-Based
Degrees of Freedom Method Satterthwaite

Dimensions
Covariance Parameters 7
Columns in X 9
Columns in Z per Subject 3
Subjects 101
Max Obs per Subject 6

Number of Observations
Number of Observations Read 606
Number of Observations Used 606
Number of Observations Not Used 0

Iteration History
Iteration Evaluations -2 Res Log Like Criterion
0 1 9076.17781375  
1 1 8226.46708705 0.00000000

Convergence criteria met.

Estimated G Correlation Matrix
Row Effect Participant ID Col1 Col2 Col3
1 Intercept 101 1.0000 -0.4026 -0.3108
2 Slope12 101 -0.4026 1.0000 -0.1494
3 Slope26 101 -0.3108 -0.1494 1.0000

Covariance Parameter Estimates
Cov Parm Subject Estimate Standard Error Z Value Pr Z
UN(1,1) ID 242192 37151 6.52 <.0001
UN(2,1) ID -49817 17064 -2.92 0.0035
UN(2,2) ID 63222 13272 4.76 <.0001
UN(3,1) ID -7510.98 3414.18 -2.20 0.0278
UN(3,2) ID -1845.11 2068.67 -0.89 0.3724
UN(3,3) ID 2411.55 614.00 3.93 <.0001
session ID 17673 1435.84 12.31 <.0001

Fit Statistics
-2 Res Log Likelihood 8226.5
AIC (Smaller is Better) 8240.5
AICC (Smaller is Better) 8240.7
BIC (Smaller is Better) 8258.8

Null Model Likelihood Ratio Test
DF Chi-Square Pr > ChiSq
6 849.71 <.0001

Information Criteria
Neg2LogLike Parms AIC AICC HQIC BIC CAIC
8226.5 7 8240.5 8240.7 8247.9 8258.8 8265.8

Solution for Fixed Effects
Effect Estimate Standard Error DF t Value Pr > |t|
Intercept 1982.64 51.1793 98 38.74 <.0001
Slope12 -162.16 30.3688 98 -5.34 <.0001
Slope26 -35.0669 6.4901 98 -5.40 <.0001
Age80 23.0041 8.8639 98 2.60 0.0109
Slope12*Age80 -8.7589 5.2597 98 -1.67 0.0990
Slope26*Age80 -0.5135 1.1240 98 -0.46 0.6488
Reas22 -27.1200 11.4528 98 -2.37 0.0198
Slope12*Reas22 -4.7141 6.7959 98 -0.69 0.4895
Slope26*Reas22 3.3476 1.4523 98 2.30 0.0233

Type 3 Tests of Fixed Effects
Effect Num DF Den DF F Value Pr > F
Slope12 1 98 28.51 <.0001
Slope26 1 98 29.19 <.0001
Age80 1 98 6.74 0.0109
Slope12*Age80 1 98 2.77 0.0990
Slope26*Age80 1 98 0.21 0.6488
Reas22 1 98 5.61 0.0198
Slope12*Reas22 1 98 0.48 0.4895
Slope26*Reas22 1 98 5.31 0.0233

Estimates
Label Estimate Standard Error DF t Value Pr > |t|
Reasoning Effect at Session 1 -27.1200 11.4528 98 -2.37 0.0198
Reasoning Effect at Session 2 -31.8340 10.4508 98 -3.05 0.0030
Reasoning Effect at Session 3 -28.4864 9.9154 98 -2.87 0.0050
Reasoning Effect at Session 4 -25.1388 9.5724 98 -2.63 0.0100
Reasoning Effect at Session 5 -21.7912 9.4427 98 -2.31 0.0231
Reasoning Effect at Session 6 -18.4436 9.5350 98 -1.93 0.0560

Contrasts
Label Num DF Den DF Chi-Square F Value Pr > ChiSq Pr > F
DF=3 Wald Test for Effects of Age 3 98 7.10 2.37 0.0688 0.0756
DF=3 Wald Test for Effects of Reasoning 3 98 10.51 3.50 0.0147 0.0183



Total R2 (% Reduction) for PredPAge vs. PredPReas

Name PredCorr TotalR2 TotalR2Diff
PredPAge 0.32795 0.10755 .
PredPReas 0.40163 0.16131 0.053755



PsuedoR2 (% Reduction) for CovPAge vs. CovPReas

Name CovParm Subject Estimate StdErr ZValue ProbZ PseudoR2
CovPAge UN(1,1) ID 254286 38681 6.57 <.0001 .
CovPAge UN(2,2) ID 62742 13139 4.78 <.0001 .
CovPAge UN(3,3) ID 2593.60 636.25 4.08 <.0001 .
CovPAge session ID 17673 1435.84 12.31 <.0001 .
CovPReas UN(1,1) ID 242192 37151 6.52 <.0001 0.047560
CovPReas UN(2,2) ID 63222 13272 4.76 <.0001 -0.007643
CovPReas UN(3,3) ID 2411.55 614.00 3.93 <.0001 0.070193
CovPReas session ID 17673 1435.84 12.31 <.0001 -0.000000



1d: Keep Age & Reasoning, Add Effects of Education Group on Intercept, Slope12, and Slope26

The Mixed Procedure

Model Information
Data Set WORK.TRIMMED
Dependent Variable nm3rt
Covariance Structures Unstructured, Variance Components
Subject Effects ID, ID
Estimation Method REML
Residual Variance Method Parameter
Fixed Effects SE Method Model-Based
Degrees of Freedom Method Satterthwaite

Dimensions
Covariance Parameters 7
Columns in X 18
Columns in Z per Subject 3
Subjects 101
Max Obs per Subject 6

Number of Observations
Number of Observations Read 606
Number of Observations Used 606
Number of Observations Not Used 0

Iteration History
Iteration Evaluations -2 Res Log Like Criterion
0 1 9007.41365628  
1 1 8164.20937127 0.00000000

Convergence criteria met.

Estimated G Correlation Matrix
Row Effect Participant ID Col1 Col2 Col3
1 Intercept 101 1.0000 -0.4173 -0.3069
2 Slope12 101 -0.4173 1.0000 -0.1461
3 Slope26 101 -0.3069 -0.1461 1.0000

Covariance Parameter Estimates
Cov Parm Subject Estimate Standard Error Z Value Pr Z
UN(1,1) ID 246920 38218 6.46 <.0001
UN(2,1) ID -52254 17491 -2.99 0.0028
UN(2,2) ID 63495 13444 4.72 <.0001
UN(3,1) ID -7543.48 3493.64 -2.16 0.0308
UN(3,2) ID -1820.21 2099.75 -0.87 0.3860
UN(3,3) ID 2446.05 624.87 3.91 <.0001
session ID 17673 1435.84 12.31 <.0001

Fit Statistics
-2 Res Log Likelihood 8164.2
AIC (Smaller is Better) 8178.2
AICC (Smaller is Better) 8178.4
BIC (Smaller is Better) 8196.5

Null Model Likelihood Ratio Test
DF Chi-Square Pr > ChiSq
6 843.20 <.0001

Information Criteria
Neg2LogLike Parms AIC AICC HQIC BIC CAIC
8164.2 7 8178.2 8178.4 8185.6 8196.5 8203.5

Solution for Fixed Effects
Effect EducGrp Estimate Standard Error DF t Value Pr > |t|
Intercept   1978.15 105.83 96 18.69 <.0001
Slope12   -153.14 62.3250 96 -2.46 0.0158
Slope26   -24.6403 13.3543 96 -1.85 0.0681
Age80   22.9367 8.9490 96 2.56 0.0119
Slope12*Age80   -8.9054 5.2704 96 -1.69 0.0943
Slope26*Age80   -0.5289 1.1293 96 -0.47 0.6406
Reas22   -28.5673 11.9710 96 -2.39 0.0190
Slope12*Reas22   -7.0891 7.0501 96 -1.01 0.3172
Slope26*Reas22   3.4883 1.5106 96 2.31 0.0231
EducGrp 1 -41.9718 157.35 96 -0.27 0.7902
EducGrp 2 25.4470 125.54 96 0.20 0.8398
EducGrp 3 0 . . . .
Slope12*EducGrp 1 -85.9455 92.6714 96 -0.93 0.3560
Slope12*EducGrp 2 18.5834 73.9371 96 0.25 0.8021
Slope12*EducGrp 3 0 . . . .
Slope26*EducGrp 1 -6.3237 19.8566 96 -0.32 0.7508
Slope26*EducGrp 2 -16.5965 15.8424 96 -1.05 0.2975
Slope26*EducGrp 3 0 . . . .

Type 3 Tests of Fixed Effects
Effect Num DF Den DF F Value Pr > F
Slope12 1 96 28.16 <.0001
Slope26 1 96 20.73 <.0001
Age80 1 96 6.57 0.0119
Slope12*Age80 1 96 2.86 0.0943
Slope26*Age80 1 96 0.22 0.6406
Reas22 1 96 5.69 0.0190
Slope12*Reas22 1 96 1.01 0.3172
Slope26*Reas22 1 96 5.33 0.0231
EducGrp 2 96 0.12 0.8831
Slope12*EducGrp 2 96 0.85 0.4289
Slope26*EducGrp 2 96 0.60 0.5516

Estimates
Label Estimate Standard Error DF t Value Pr > |t|
L vs. H Educ for Intercept 41.9718 157.35 96 0.27 0.7902
M vs. H Educ for Intercept -25.4470 125.54 96 -0.20 0.8398
L vs. M Educ for Intercept 67.4187 136.36 96 0.49 0.6221
L vs. H Educ for Slope12 85.9455 92.6714 96 0.93 0.3560
M vs. H Educ for Slope12 -18.5834 73.9371 96 -0.25 0.8021
L vs. M Educ for Slope12 104.53 80.3066 96 1.30 0.1962
L vs. H Educ for Slope26 6.3237 19.8566 96 0.32 0.7508
M vs. H Educ for Slope26 16.5965 15.8424 96 1.05 0.2975
L vs. M Educ for Slope26 -10.2728 17.2072 96 -0.60 0.5519

Contrasts
Label Num DF Den DF Chi-Square F Value Pr > ChiSq Pr > F
DF=3 Wald Test for Effects of Age 3 96 6.96 2.32 0.0732 0.0802
DF=3 Wald Test for Effects of Reasoning 3 96 11.92 3.97 0.0077 0.0103
DF=6 Wald Test for Effects of Education 6 96 4.38 0.73 0.6252 0.6264

Least Squares Means
Effect EducGrp Slope12 Slope26 Age80 Reas22 Estimate Standard Error DF t Value Pr > |t|
EducGrp 1 0.00 0.00 0.00 0.00 1936.18 114.13 96 16.97 <.0001
EducGrp 2 0.00 0.00 0.00 0.00 2003.60 70.3593 96 28.48 <.0001
EducGrp 3 0.00 0.00 0.00 0.00 1978.15 105.83 96 18.69 <.0001
EducGrp 1 1.00 4.00 0.00 0.00 1573.24 94.3228 96 16.68 <.0001
EducGrp 2 1.00 4.00 0.00 0.00 1704.10 58.1509 96 29.30 <.0001
EducGrp 3 1.00 4.00 0.00 0.00 1726.45 87.4643 96 19.74 <.0001

Differences of Least Squares Means
Effect EducGrp _EducGrp Slope12 Slope26 Age80 Reas22 Estimate Standard Error DF t Value Pr > |t|
EducGrp 1 2 0.00 0.00 0.00 0.00 -67.4187 136.36 96 -0.49 0.6221
EducGrp 1 3 0.00 0.00 0.00 0.00 -41.9718 157.35 96 -0.27 0.7902
EducGrp 2 3 0.00 0.00 0.00 0.00 25.4470 125.54 96 0.20 0.8398
EducGrp 1 2 1.00 4.00 0.00 0.00 -130.86 112.70 96 -1.16 0.2485
EducGrp 1 3 1.00 4.00 0.00 0.00 -153.21 130.05 96 -1.18 0.2417
EducGrp 2 3 1.00 4.00 0.00 0.00 -22.3558 103.76 96 -0.22 0.8299



Total R2 (% Reduction) for PredPReas vs. PredPEduc

Name PredCorr TotalR2 TotalR2Diff
PredPReas 0.40163 0.16131 .
PredPEduc 0.41669 0.17363 0.012322



PsuedoR2 (% Reduction) for CovPReas vs. CovPEduc

Name CovParm Subject Estimate StdErr ZValue ProbZ PseudoR2
CovPReas UN(1,1) ID 242192 37151 6.52 <.0001 .
CovPReas UN(2,2) ID 63222 13272 4.76 <.0001 .
CovPReas UN(3,3) ID 2411.55 614.00 3.93 <.0001 .
CovPReas session ID 17673 1435.84 12.31 <.0001 .
CovPEduc UN(1,1) ID 246920 38218 6.46 <.0001 -0.019521
CovPEduc UN(2,2) ID 63495 13444 4.72 <.0001 -0.004322
CovPEduc UN(3,3) ID 2446.05 624.87 3.91 <.0001 -0.014309
CovPEduc session ID 17673 1435.84 12.31 <.0001 0.000000



1e: Drop EducGrp, Add Fixed Effects for Age*Reasoning on Intercept, Slope12, and Slope26

The Mixed Procedure

Model Information
Data Set WORK.TRIMMED
Dependent Variable nm3rt
Covariance Structures Unstructured, Variance Components
Subject Effects ID, ID
Estimation Method REML
Residual Variance Method Parameter
Fixed Effects SE Method Model-Based
Degrees of Freedom Method Satterthwaite

Dimensions
Covariance Parameters 7
Columns in X 12
Columns in Z per Subject 3
Subjects 101
Max Obs per Subject 6

Number of Observations
Number of Observations Read 606
Number of Observations Used 606
Number of Observations Not Used 0

Iteration History
Iteration Evaluations -2 Res Log Like Criterion
0 1 9070.73631363  
1 1 8220.92912545 0.00000000

Convergence criteria met.

Estimated G Correlation Matrix
Row Effect Participant ID Col1 Col2 Col3
1 Intercept 101 1.0000 -0.4001 -0.3074
2 Slope12 101 -0.4001 1.0000 -0.1611
3 Slope26 101 -0.3074 -0.1611 1.0000

Covariance Parameter Estimates
Cov Parm Subject Estimate Standard Error Z Value Pr Z
UN(1,1) ID 244192 37629 6.49 <.0001
UN(2,1) ID -49617 17179 -2.89 0.0039
UN(2,2) ID 62984 13304 4.73 <.0001
UN(3,1) ID -7513.67 3457.96 -2.17 0.0298
UN(3,2) ID -1999.16 2088.67 -0.96 0.3385
UN(3,3) ID 2446.40 621.86 3.93 <.0001
session ID 17673 1435.84 12.31 <.0001

Fit Statistics
-2 Res Log Likelihood 8220.9
AIC (Smaller is Better) 8234.9
AICC (Smaller is Better) 8235.1
BIC (Smaller is Better) 8253.2

Null Model Likelihood Ratio Test
DF Chi-Square Pr > ChiSq
6 849.81 <.0001

Information Criteria
Neg2LogLike Parms AIC AICC HQIC BIC CAIC
8220.9 7 8234.9 8235.1 8242.3 8253.2 8260.2

Solution for Fixed Effects
Effect Estimate Standard Error DF t Value Pr > |t|
Intercept 1974.57 53.8381 97 36.68 <.0001
Slope12 -151.52 31.7828 97 -4.77 <.0001
Slope26 -34.1783 6.8294 97 -5.00 <.0001
Age80 22.7598 8.9112 97 2.55 0.0122
Slope12*Age80 -8.4366 5.2607 97 -1.60 0.1120
Slope26*Age80 -0.4866 1.1304 97 -0.43 0.6678
Reas22 -28.0448 11.6437 97 -2.41 0.0179
Slope12*Reas22 -3.4941 6.8738 97 -0.51 0.6124
Slope26*Reas22 3.4494 1.4770 97 2.34 0.0216
Age80*Reas22 -0.9317 1.8579 97 -0.50 0.6172
Slope12*Age80*Reas22 1.2290 1.0968 97 1.12 0.2652
Slope26*Age80*Reas22 0.1026 0.2357 97 0.44 0.6644

Type 3 Tests of Fixed Effects
Effect Num DF Den DF F Value Pr > F
Slope12 1 97 22.73 <.0001
Slope26 1 97 25.05 <.0001
Age80 1 97 6.52 0.0122
Slope12*Age80 1 97 2.57 0.1120
Slope26*Age80 1 97 0.19 0.6678
Reas22 1 97 5.80 0.0179
Slope12*Reas22 1 97 0.26 0.6124
Slope26*Reas22 1 97 5.45 0.0216
Age80*Reas22 1 97 0.25 0.6172
Slope12*Age80*Reas22 1 97 1.26 0.2652
Slope26*Age80*Reas22 1 97 0.19 0.6644

Estimates
Label Estimate Standard Error DF t Value Pr > |t|
Age Effect on Intercept, Reas 17 27.4184 12.5162 97 2.19 0.0309
Age Effect on Intercept, Reas 22 22.7598 8.9112 97 2.55 0.0122
Age Effect on Intercept, Reas 27 18.1011 13.2197 97 1.37 0.1741
Age Effect on Slope12, Reas 17 -14.5818 7.3888 97 -1.97 0.0513
Age Effect on Slope12, Reas 22 -8.4366 5.2607 97 -1.60 0.1120
Age Effect on Slope12, Reas 27 -2.2914 7.8042 97 -0.29 0.7697
Age Effect on Slope26, Reas 17 -0.9994 1.5877 97 -0.63 0.5305
Age Effect on Slope26, Reas 22 -0.4866 1.1304 97 -0.43 0.6678
Age Effect on Slope26, Reas 27 0.02627 1.6769 97 0.02 0.9875
Reasoning Effect on Intercept, Age 70 -18.7275 20.3038 97 -0.92 0.3586
Reasoning Effect on Intercept, Age 80 -28.0448 11.6437 97 -2.41 0.0179
Reasoning Effect on Intercept, Age 90 -37.3622 23.4371 97 -1.59 0.1142
Reasoning Effect on Slope12, Age 70 -15.7845 11.9862 97 -1.32 0.1910
Reasoning Effect on Slope12, Age 80 -3.4941 6.8738 97 -0.51 0.6124
Reasoning Effect on Slope12, Age 90 8.7963 13.8359 97 0.64 0.5264
Reasoning Effect on Slope26, Age 70 2.4237 2.5756 97 0.94 0.3490
Reasoning Effect on Slope26, Age 80 3.4494 1.4770 97 2.34 0.0216
Reasoning Effect on Slope26, Age 90 4.4751 2.9730 97 1.51 0.1355

Contrasts
Label Num DF Den DF Chi-Square F Value Pr > ChiSq Pr > F
DF=3 Wald Test for Effects of Age*Reasoning 3 97 1.98 0.66 0.5771 0.5791



Total R2 (% Reduction) for PredPReas vs. PredPAgeReas

Name PredCorr TotalR2 TotalR2Diff
PredPReas 0.40163 0.16131 .
PredPAgeReas 0.40306 0.16246 .001148258



PsuedoR2 (% Reduction) for CovPReas vs. CovPAgeReas

Name CovParm Subject Estimate StdErr ZValue ProbZ PseudoR2
CovPReas UN(1,1) ID 242192 37151 6.52 <.0001 .
CovPReas UN(2,2) ID 63222 13272 4.76 <.0001 .
CovPReas UN(3,3) ID 2411.55 614.00 3.93 <.0001 .
CovPReas session ID 17673 1435.84 12.31 <.0001 .
CovPAgeReas UN(1,1) ID 244192 37629 6.49 <.0001 -0.008258
CovPAgeReas UN(2,2) ID 62984 13304 4.73 <.0001 0.003765
CovPAgeReas UN(3,3) ID 2446.40 621.86 3.93 <.0001 -0.014453
CovPAgeReas session ID 17673 1435.84 12.31 <.0001 0.000000



2a: Random Quadratic Unconditional Model

The Mixed Procedure

Model Information
Data Set WORK.TRIMMED
Dependent Variable nm3rt
Covariance Structures Unstructured, Variance Components
Subject Effects ID, ID
Estimation Method REML
Residual Variance Method Parameter
Fixed Effects SE Method Model-Based
Degrees of Freedom Method Satterthwaite

Dimensions
Covariance Parameters 7
Columns in X 3
Columns in Z per Subject 3
Subjects 101
Max Obs per Subject 6

Number of Observations
Number of Observations Read 606
Number of Observations Used 606
Number of Observations Not Used 0

Iteration History
Iteration Evaluations -2 Res Log Like Criterion
0 1 9193.25780414  
1 1 8302.74566856 0.00000000

Convergence criteria met.

Estimated G Correlation Matrix
Row Effect Participant ID Col1 Col2 Col3
1 Intercept 101 1.0000 -0.4230 0.2948
2 time1 101 -0.4230 1.0000 -0.9640
3 time1*time1 101 0.2948 -0.9640 1.0000

Covariance Parameter Estimates
Cov Parm Subject Estimate Standard Error Z Value Pr Z
UN(1,1) ID 276206 41442 6.66 <.0001
UN(2,1) ID -35734 11941 -2.99 0.0028
UN(2,2) ID 25840 5864.41 4.41 <.0001
UN(3,1) ID 3901.96 1949.06 2.00 0.0453
UN(3,2) ID -3903.32 982.61 -3.97 <.0001
UN(3,3) ID 634.47 172.37 3.68 0.0001
session ID 20298 1649.11 12.31 <.0001

Fit Statistics
-2 Res Log Likelihood 8302.7
AIC (Smaller is Better) 8316.7
AICC (Smaller is Better) 8316.9
BIC (Smaller is Better) 8335.1

Null Model Likelihood Ratio Test
DF Chi-Square Pr > ChiSq
6 890.51 <.0001

Information Criteria
Neg2LogLike Parms AIC AICC HQIC BIC CAIC
8302.7 7 8316.7 8316.9 8324.2 8335.1 8342.1

Solution for Fixed Effects
Effect Estimate Standard Error DF t Value Pr > |t|
Intercept 1945.85 53.8497 100 36.13 <.0001
time1 -120.90 20.0476 100 -6.03 <.0001
time1*time1 13.8656 3.4154 100 4.06 <.0001

Type 3 Tests of Fixed Effects
Effect Num DF Den DF F Value Pr > F
time1 1 100 36.37 <.0001
time1*time1 1 100 16.48 <.0001



The CORR Procedure

1 With Variables: nm3rt
1 Variables: Pred

Simple Statistics
Variable N Mean Std Dev Sum Minimum Maximum Label
nm3rt 606 1771 494.08763 1073045 917.66667 4159 Number Match 3 RT
Pred 606 1771 94.70156 1073045 1684 1946 Predicted Mean

Pearson Correlation Coefficients, N = 606
Prob > |r| under H0: Rho=0
  Pred
nm3rt
Number Match 3 RT
0.19167
<.0001



2b: Add Fixed Effects for Age on Intercept, Linear, and Quadratic Time Slopes

The Mixed Procedure

Model Information
Data Set WORK.TRIMMED
Dependent Variable nm3rt
Covariance Structures Unstructured, Variance Components
Subject Effects ID, ID
Estimation Method REML
Residual Variance Method Parameter
Fixed Effects SE Method Model-Based
Degrees of Freedom Method Satterthwaite

Dimensions
Covariance Parameters 7
Columns in X 6
Columns in Z per Subject 3
Subjects 101
Max Obs per Subject 6

Number of Observations
Number of Observations Read 606
Number of Observations Used 606
Number of Observations Not Used 0

Iteration History
Iteration Evaluations -2 Res Log Like Criterion
0 1 9138.19490348  
1 1 8283.15568212 0.00000000

Convergence criteria met.

Estimated G Correlation Matrix
Row Effect Participant ID Col1 Col2 Col3
1 Intercept 101 1.0000 -0.3826 0.2589
2 time1 101 -0.3826 1.0000 -0.9639
3 time1*time1 101 0.2589 -0.9639 1.0000

Covariance Parameter Estimates
Cov Parm Subject Estimate Standard Error Z Value Pr Z
UN(1,1) ID 247691 37599 6.59 <.0001
UN(2,1) ID -30154 11191 -2.69 0.0070
UN(2,2) ID 25083 5787.37 4.33 <.0001
UN(3,1) ID 3232.78 1847.12 1.75 0.0801
UN(3,2) ID -3830.21 976.76 -3.92 <.0001
UN(3,3) ID 629.58 172.51 3.65 0.0001
session ID 20298 1649.11 12.31 <.0001

Fit Statistics
-2 Res Log Likelihood 8283.2
AIC (Smaller is Better) 8297.2
AICC (Smaller is Better) 8297.3
BIC (Smaller is Better) 8315.5

Null Model Likelihood Ratio Test
DF Chi-Square Pr > ChiSq
6 855.04 <.0001

Information Criteria
Neg2LogLike Parms AIC AICC HQIC BIC CAIC
8283.2 7 8297.2 8297.3 8304.6 8315.5 8322.5

Solution for Fixed Effects
Effect Estimate Standard Error DF t Value Pr > |t|
Intercept 1950.69 51.1806 99 38.11 <.0001
time1 -121.83 19.8672 99 -6.13 <.0001
time1*time1 13.9774 3.4096 99 4.10 <.0001
Age80 29.0495 8.4616 99 3.43 0.0009
time1*Age80 -5.5946 3.2846 99 -1.70 0.0916
time1*time1*Age80 0.6709 0.5637 99 1.19 0.2368

Type 3 Tests of Fixed Effects
Effect Num DF Den DF F Value Pr > F
time1 1 99 37.61 <.0001
time1*time1 1 99 16.81 <.0001
Age80 1 99 11.79 0.0009
time1*Age80 1 99 2.90 0.0916
time1*time1*Age80 1 99 1.42 0.2368

Estimates
Label Estimate Standard Error DF t Value Pr > |t|
Age Effect at Session 1 29.0495 8.4616 99 3.43 0.0009
Age Effect at Session 2 24.1258 7.6862 99 3.14 0.0022
Age Effect at Session 3 20.5439 7.5343 99 2.73 0.0076
Age Effect at Session 4 18.3038 7.4038 99 2.47 0.0151
Age Effect at Session 5 17.4056 7.1425 99 2.44 0.0166
Age Effect at Session 6 17.8492 7.1254 99 2.51 0.0139
Age*Linear Time Slope at Session 1 -5.5946 3.2846 99 -1.70 0.0916
Age*Linear Time Slope at Session 2 -4.2528 2.2283 99 -1.91 0.0592
Age*Linear Time Slope at Session 3 -2.9110 1.2977 99 -2.24 0.0271
Age*Linear Time Slope at Session 4 -1.5692 0.9720 99 -1.61 0.1096
Age*Linear Time Slope at Session 5 -0.2273 1.6576 99 -0.14 0.8912
Age*Linear Time Slope at Session 6 1.1145 2.6632 99 0.42 0.6765

Contrasts
Label Num DF Den DF Chi-Square F Value Pr > ChiSq Pr > F
DF=3 Wald Test for Effects of Age 3 99 12.01 4.00 0.0073 0.0098



Total R2 (% Reduction) for PredQUnc vs. PredQAge

Name PredCorr TotalR2 TotalR2Diff
PredQUnc 0.19167 0.03674 .
PredQAge 0.32688 0.10685 0.070114



PsuedoR2 (% Reduction) for CovQUnc vs. CovQAge

Name CovParm Subject Estimate StdErr ZValue ProbZ PseudoR2
CovQUnc UN(1,1) ID 276206 41442 6.66 <.0001 .
CovQUnc UN(2,2) ID 25840 5864.41 4.41 <.0001 .
CovQUnc UN(3,3) ID 634.47 172.37 3.68 0.0001 .
CovQUnc session ID 20298 1649.11 12.31 <.0001 .
CovQAge UN(1,1) ID 247691 37599 6.59 <.0001 0.10324
CovQAge UN(2,2) ID 25083 5787.37 4.33 <.0001 0.02931
CovQAge UN(3,3) ID 629.58 172.51 3.65 0.0001 0.00770
CovQAge session ID 20298 1649.11 12.31 <.0001 0.00000



2c: Keep Age, Add Fixed Effects for Reasoning on Intercept, Linear, and Quadratic Time Slopes

The Mixed Procedure

Model Information
Data Set WORK.TRIMMED
Dependent Variable nm3rt
Covariance Structures Unstructured, Variance Components
Subject Effects ID, ID
Estimation Method REML
Residual Variance Method Parameter
Fixed Effects SE Method Model-Based
Degrees of Freedom Method Satterthwaite

Dimensions
Covariance Parameters 7
Columns in X 9
Columns in Z per Subject 3
Subjects 101
Max Obs per Subject 6

Number of Observations
Number of Observations Read 606
Number of Observations Used 606
Number of Observations Not Used 0

Iteration History
Iteration Evaluations -2 Res Log Like Criterion
0 1 9089.57995991  
1 1 8261.01548560 0.00000000

Convergence criteria met.

Estimated G Correlation Matrix
Row Effect Participant ID Col1 Col2 Col3
1 Intercept 101 1.0000 -0.4223 0.3257
2 time1 101 -0.4223 1.0000 -0.9684
3 time1*time1 101 0.3257 -0.9684 1.0000

Covariance Parameter Estimates
Cov Parm Subject Estimate Standard Error Z Value Pr Z
UN(1,1) ID 235541 36056 6.53 <.0001
UN(2,1) ID -32552 11138 -2.92 0.0035
UN(2,2) ID 25228 5835.93 4.32 <.0001
UN(3,1) ID 3918.44 1826.88 2.14 0.0320
UN(3,2) ID -3812.99 978.05 -3.90 <.0001
UN(3,3) ID 614.47 171.25 3.59 0.0002
session ID 20298 1649.11 12.31 <.0001

Fit Statistics
-2 Res Log Likelihood 8261.0
AIC (Smaller is Better) 8275.0
AICC (Smaller is Better) 8275.2
BIC (Smaller is Better) 8293.3

Null Model Likelihood Ratio Test
DF Chi-Square Pr > ChiSq
6 828.56 <.0001

Information Criteria
Neg2LogLike Parms AIC AICC HQIC BIC CAIC
8261.0 7 8275.0 8275.2 8282.4 8293.3 8300.3

Solution for Fixed Effects
Effect Estimate Standard Error DF t Value Pr > |t|
Intercept 1966.47 50.4203 98 39.00 <.0001
time1 -119.74 20.0746 98 -5.96 <.0001
time1*time1 13.3036 3.4167 98 3.89 0.0002
Age80 22.2782 8.7324 98 2.55 0.0123
time1*Age80 -6.4921 3.4768 98 -1.87 0.0649
time1*time1*Age80 0.9601 0.5917 98 1.62 0.1079
Reas22 -27.1004 11.2829 98 -2.40 0.0182
time1*Reas22 -3.5917 4.4922 98 -0.80 0.4259
time1*time1*Reas22 1.1575 0.7646 98 1.51 0.1333

Type 3 Tests of Fixed Effects
Effect Num DF Den DF F Value Pr > F
time1 1 98 35.58 <.0001
time1*time1 1 98 15.16 0.0002
Age80 1 98 6.51 0.0123
time1*Age80 1 98 3.49 0.0649
time1*time1*Age80 1 98 2.63 0.1079
Reas22 1 98 5.77 0.0182
time1*Reas22 1 98 0.64 0.4259
time1*time1*Reas22 1 98 2.29 0.1333

Estimates
Label Estimate Standard Error DF t Value Pr > |t|
Reasoning Effect at Session 1 -27.1004 11.2829 98 -2.40 0.0182
Reasoning Effect at Session 2 -29.5346 10.1156 98 -2.92 0.0043
Reasoning Effect at Session 3 -29.6537 9.8944 98 -3.00 0.0035
Reasoning Effect at Session 4 -27.4578 9.7730 98 -2.81 0.0060
Reasoning Effect at Session 5 -22.9468 9.5224 98 -2.41 0.0178
Reasoning Effect at Session 6 -16.1207 9.6403 98 -1.67 0.0977
Reasoning*Linear Time Slope at Session 1 -3.5917 4.4922 98 -0.80 0.4259
Reasoning*Linear Time Slope at Session 2 -1.2767 3.0547 98 -0.42 0.6769
Reasoning*Linear Time Slope at Session 3 1.0384 1.7775 98 0.58 0.5604
Reasoning*Linear Time Slope at Session 4 3.3535 1.2900 98 2.60 0.0108
Reasoning*Linear Time Slope at Session 5 5.6686 2.2012 98 2.58 0.0115
Reasoning*Linear Time Slope at Session 6 7.9836 3.5642 98 2.24 0.0274

Contrasts
Label Num DF Den DF Chi-Square F Value Pr > ChiSq Pr > F
DF=3 Wald Test for Effects of Age 3 98 7.19 2.40 0.0660 0.0727
DF=3 Wald Test for Effects of Reasoning 3 98 12.88 4.29 0.0049 0.0068



Total R2 (% Reduction) for PredQAge vs. PredQReas

Name PredCorr TotalR2 TotalR2Diff
PredQAge 0.32688 0.10685 .
PredQReas 0.40108 0.16086 0.054011



PsuedoR2 (% Reduction) for CovQAge vs. CovQReas

Name CovParm Subject Estimate StdErr ZValue ProbZ PseudoR2
CovQAge UN(1,1) ID 247691 37599 6.59 <.0001 .
CovQAge UN(2,2) ID 25083 5787.37 4.33 <.0001 .
CovQAge UN(3,3) ID 629.58 172.51 3.65 0.0001 .
CovQAge session ID 20298 1649.11 12.31 <.0001 .
CovQReas UN(1,1) ID 235541 36056 6.53 <.0001 0.049052
CovQReas UN(2,2) ID 25228 5835.93 4.32 <.0001 -0.005808
CovQReas UN(3,3) ID 614.47 171.25 3.59 0.0002 0.024008
CovQReas session ID 20298 1649.11 12.31 <.0001 -0.000000



2d: Remove Reasoning Effect on Quadratic Time Slope

The Mixed Procedure

Model Information
Data Set WORK.TRIMMED
Dependent Variable nm3rt
Covariance Structures Unstructured, Variance Components
Subject Effects ID, ID
Estimation Method REML
Residual Variance Method Parameter
Fixed Effects SE Method Model-Based
Degrees of Freedom Method Satterthwaite

Dimensions
Covariance Parameters 7
Columns in X 8
Columns in Z per Subject 3
Subjects 101
Max Obs per Subject 6

Number of Observations
Number of Observations Read 606
Number of Observations Used 606
Number of Observations Not Used 0

Iteration History
Iteration Evaluations -2 Res Log Like Criterion
0 1 9092.92701329  
1 2 8264.60019032 0.00000000

Convergence criteria met.

Estimated G Correlation Matrix
Row Effect Participant ID Col1 Col2 Col3
1 Intercept 101 1.0000 -0.4234 0.3277
2 time1 101 -0.4234 1.0000 -0.9689
3 time1*time1 101 0.3277 -0.9689 1.0000

Covariance Parameter Estimates
Cov Parm Subject Estimate Standard Error Z Value Pr Z
UN(1,1) ID 235909 36153 6.53 <.0001
UN(2,1) ID -32972 11262 -2.93 0.0034
UN(2,2) ID 25707 5883.65 4.37 <.0001
UN(3,1) ID 3993.04 1848.58 2.16 0.0308
UN(3,2) ID -3897.93 985.52 -3.96 <.0001
UN(3,3) ID 629.52 172.50 3.65 0.0001
session ID 20298 1649.11 12.31 <.0001

Fit Statistics
-2 Res Log Likelihood 8264.6
AIC (Smaller is Better) 8278.6
AICC (Smaller is Better) 8278.8
BIC (Smaller is Better) 8296.9

Null Model Likelihood Ratio Test
DF Chi-Square Pr > ChiSq
6 828.33 <.0001

Information Criteria
Neg2LogLike Parms AIC AICC HQIC BIC CAIC
8264.6 7 8278.6 8278.8 8286.0 8296.9 8303.9

Solution for Fixed Effects
Effect Estimate Standard Error DF t Value Pr > |t|
Intercept 1969.80 50.4084 98.1 39.08 <.0001
time1 -123.54 20.0358 98.9 -6.17 <.0001
time1*time1 13.9774 3.4095 99 4.10 <.0001
Age80 20.8470 8.6868 99.7 2.40 0.0183
time1*Age80 -4.8610 3.3252 100 -1.46 0.1469
time1*time1*Age80 0.6709 0.5637 99 1.19 0.2368
Reas22 -32.8281 10.6297 98 -3.09 0.0026
time1*Reas22 2.9362 1.2602 98 2.33 0.0219

Type 3 Tests of Fixed Effects
Effect Num DF Den DF F Value Pr > F
time1 1 98.9 38.02 <.0001
time1*time1 1 99 16.81 <.0001
Age80 1 99.7 5.76 0.0183
time1*Age80 1 100 2.14 0.1469
time1*time1*Age80 1 99 1.42 0.2368
Reas22 1 98 9.54 0.0026
time1*Reas22 1 98 5.43 0.0219

Estimates
Label Estimate Standard Error DF t Value Pr > |t|
Reasoning Effect at Session 1 -32.8281 10.6297 98 -3.09 0.0026
Reasoning Effect at Session 2 -29.8919 10.1128 98 -2.96 0.0039
Reasoning Effect at Session 3 -26.9557 9.7326 98 -2.77 0.0067
Reasoning Effect at Session 4 -24.0195 9.5055 98 -2.53 0.0131
Reasoning Effect at Session 5 -21.0833 9.4425 98 -2.23 0.0278
Reasoning Effect at Session 6 -18.1471 9.5469 98 -1.90 0.0603

Contrasts
Label Num DF Den DF Chi-Square F Value Pr > ChiSq Pr > F
DF=3 Wald Test for Effects of Age 3 100 5.98 1.99 0.1128 0.1200
DF=2 Wald Test for Effects of Reasoning 2 98 10.59 5.29 0.0050 0.0066



Total R2 (% Reduction) for PredQAge vs. PredLReas

Name PredCorr TotalR2 TotalR2Diff
PredQAge 0.32688 0.10685 .
PredLReas 0.40008 0.16006 0.053213



PsuedoR2 (% Reduction) for CovQAge vs. CovLReas

Name CovParm Subject Estimate StdErr ZValue ProbZ PseudoR2
CovQAge UN(1,1) ID 247691 37599 6.59 <.0001 .
CovQAge UN(2,2) ID 25083 5787.37 4.33 <.0001 .
CovQAge UN(3,3) ID 629.58 172.51 3.65 0.0001 .
CovQAge session ID 20298 1649.11 12.31 <.0001 .
CovLReas UN(1,1) ID 235909 36153 6.53 <.0001 0.047565
CovLReas UN(2,2) ID 25707 5883.65 4.37 <.0001 -0.024908
CovLReas UN(3,3) ID 629.52 172.50 3.65 0.0001 0.000095
CovLReas session ID 20298 1649.11 12.31 <.0001 -0.000000



2e: Keep Age & Reas, Add Effect of Education Group on Intercept, Linear, and Quadratic

The Mixed Procedure

Model Information
Data Set WORK.TRIMMED
Dependent Variable nm3rt
Covariance Structures Unstructured, Variance Components
Subject Effects ID, ID
Estimation Method REML
Residual Variance Method Parameter
Fixed Effects SE Method Model-Based
Degrees of Freedom Method Satterthwaite

Dimensions
Covariance Parameters 7
Columns in X 17
Columns in Z per Subject 3
Subjects 101
Max Obs per Subject 6

Number of Observations
Number of Observations Read 606
Number of Observations Used 606
Number of Observations Not Used 0

Iteration History
Iteration Evaluations -2 Res Log Like Criterion
0 1 9032.94560623  
1 2 8211.36197595 0.00000000

Convergence criteria met.

Estimated G Correlation Matrix
Row Effect Participant ID Col1 Col2 Col3
1 Intercept 101 1.0000 -0.4475 0.3553
2 time1 101 -0.4475 1.0000 -0.9684
3 time1*time1 101 0.3553 -0.9684 1.0000

Covariance Parameter Estimates
Cov Parm Subject Estimate Standard Error Z Value Pr Z
UN(1,1) ID 241027 37339 6.46 <.0001
UN(2,1) ID -35271 11645 -3.03 0.0025
UN(2,2) ID 25772 5956.96 4.33 <.0001
UN(3,1) ID 4371.57 1907.59 2.29 0.0219
UN(3,2) ID -3896.53 995.30 -3.91 <.0001
UN(3,3) ID 628.15 173.93 3.61 0.0002
session ID 20298 1649.11 12.31 <.0001

Fit Statistics
-2 Res Log Likelihood 8211.4
AIC (Smaller is Better) 8225.4
AICC (Smaller is Better) 8225.6
BIC (Smaller is Better) 8243.7

Null Model Likelihood Ratio Test
DF Chi-Square Pr > ChiSq
6 821.58 <.0001

Information Criteria
Neg2LogLike Parms AIC AICC HQIC BIC CAIC
8211.4 7 8225.4 8225.6 8232.8 8243.7 8250.7

Solution for Fixed Effects
Effect EducGrp Estimate Standard Error DF t Value Pr > |t|
Intercept   1961.89 104.34 95.7 18.80 <.0001
time1   -106.50 41.1184 96.7 -2.59 0.0111
time1*time1   12.4797 6.9879 97 1.79 0.0772
Age80   20.2894 8.7750 97.5 2.31 0.0229
time1*Age80   -4.5759 3.3351 98 -1.37 0.1732
time1*time1*Age80   0.6177 0.5646 97 1.09 0.2767
Reas22   -36.6221 11.0407 96 -3.32 0.0013
time1*Reas22   2.9786 1.3130 96.1 2.27 0.0255
EducGrp 1 -51.3792 154.85 96.3 -0.33 0.7408
EducGrp 2 37.6426 123.90 95.4 0.30 0.7619
EducGrp 3 0 . . . .
time1*EducGrp 1 -70.2451 60.3032 97.1 -1.16 0.2469
time1*EducGrp 2 -4.3577 49.1299 96.5 -0.09 0.9295
time1*EducGrp 3 0 . . . .
time1*time1*EducGrp 1 11.0653 10.2358 97 1.08 0.2824
time1*time1*EducGrp 2 -1.4641 8.3545 97 -0.18 0.8612
time1*time1*EducGrp 3 0 . . . .

Type 3 Tests of Fixed Effects
Effect Num DF Den DF F Value Pr > F
time1 1 96.5 35.77 <.0001
time1*time1 1 97 17.62 <.0001
Age80 1 97.5 5.35 0.0229
time1*Age80 1 98 1.88 0.1732
time1*time1*Age80 1 97 1.20 0.2767
Reas22 1 96 11.00 0.0013
time1*Reas22 1 96.1 5.15 0.0255
EducGrp 2 96.1 0.23 0.7965
time1*EducGrp 2 97 0.92 0.4012
time1*time1*EducGrp 2 97 1.05 0.3545

Estimates
Label Estimate Standard Error DF t Value Pr > |t|
L vs. H Educ for Intercept 51.3792 154.85 96.3 0.33 0.7408
M vs. H Educ for Intercept -37.6426 123.90 95.4 -0.30 0.7619
L vs. M Educ for Intercept 89.0218 134.02 96.8 0.66 0.5081
L vs. H Educ for Linear Time 70.2451 60.3032 97.1 1.16 0.2469
M vs. H Educ for Linear Time 4.3577 49.1299 96.5 0.09 0.9295
L vs. M Educ for Linear Time 65.8874 51.7661 97.4 1.27 0.2061
L vs. H Educ for Quadratic Time -11.0653 10.2358 97 -1.08 0.2824
M vs. H Educ for Quadratic Time 1.4641 8.3545 97 0.18 0.8612
L vs. M Educ for Quadratic Time -12.5294 8.7793 97 -1.43 0.1567

Contrasts
Label Num DF Den DF Chi-Square F Value Pr > ChiSq Pr > F
DF=3 Wald Test for Effects of Age 3 98.1 5.49 1.83 0.1395 0.1469
DF=2 Wald Test for Effects of Reasoning 2 96 11.70 5.85 0.0029 0.0040
DF=6 Wald Test for Effects of Education 6 96.4 4.59 0.76 0.5976 0.5994

Least Squares Means
Effect EducGrp time1 Age80 Reas22 Estimate Standard Error DF t Value Pr > |t|
EducGrp 1 0.00 0.00 0.00 1910.51 112.41 96.1 17.00 <.0001
EducGrp 2 0.00 0.00 0.00 1999.53 69.2506 96.3 28.87 <.0001
EducGrp 3 0.00 0.00 0.00 1961.89 104.34 95.7 18.80 <.0001
EducGrp 1 5.00 0.00 0.00 1615.41 95.7317 96 16.87 <.0001
EducGrp 2 5.00 0.00 0.00 1720.63 59.0105 96.1 29.16 <.0001
EducGrp 3 5.00 0.00 0.00 1741.38 88.7887 95.9 19.61 <.0001

Differences of Least Squares Means
Effect EducGrp _EducGrp time1 Age80 Reas22 Estimate Standard Error DF t Value Pr > |t|
EducGrp 1 2 0.00 0.00 0.00 -89.0218 134.02 96.8 -0.66 0.5081
EducGrp 1 3 0.00 0.00 0.00 -51.3792 154.85 96.3 -0.33 0.7408
EducGrp 2 3 0.00 0.00 0.00 37.6426 123.90 95.4 0.30 0.7619
EducGrp 1 2 5.00 0.00 0.00 -105.22 114.33 96.2 -0.92 0.3597
EducGrp 1 3 5.00 0.00 0.00 -125.97 131.97 96.1 -0.95 0.3422
EducGrp 2 3 5.00 0.00 0.00 -20.7486 105.35 95.9 -0.20 0.8443



Total R2 (% Reduction) for PredLReas vs. PredQEduc

Name PredCorr TotalR2 TotalR2Diff
PredLReas 0.40008 0.16006 .
PredQEduc 0.41510 0.17231 0.012242



PsuedoR2 (% Reduction) for CovLReas vs. CovQEduc

Name CovParm Subject Estimate StdErr ZValue ProbZ PseudoR2
CovLReas UN(1,1) ID 235909 36153 6.53 <.0001 .
CovLReas UN(2,2) ID 25707 5883.65 4.37 <.0001 .
CovLReas UN(3,3) ID 629.52 172.50 3.65 0.0001 .
CovLReas session ID 20298 1649.11 12.31 <.0001 .
CovQEduc UN(1,1) ID 241027 37339 6.46 <.0001 -0.021693
CovQEduc UN(2,2) ID 25772 5956.96 4.33 <.0001 -0.002519
CovQEduc UN(3,3) ID 628.15 173.93 3.61 0.0002 0.002185
CovQEduc session ID 20298 1649.11 12.31 <.0001 -0.000000