This page is still under construction!!
The examples on this page is done using Mplus 4.2.
Page 16 mplus16.dat The data are shown on page 16, and the probabilities are shown at the bottom of page 15. Also, we have included all of the output for this analysis only. For all other analyses, we will limit the output presented only to relevant parts.
data: file is "D:\work\mplus_examples\mplus16.dat"; variable: names are a b c wt; usevar a b c; freqweight is wt ; classes = grp(2); categorical = a b c; analysis: type = mixture;RESULTS IN PROBABILITY SCALE Latent Class 1 A Category 1 0.333 0.038 8.660 Category 2 0.667 0.038 17.320 B Category 1 0.200 0.033 6.124 Category 2 0.800 0.033 24.495 C Category 1 0.000 0.000 0.000 Category 2 1.000 0.000 0.000 Latent Class 2 A Category 1 0.667 0.038 17.320 Category 2 0.333 0.038 8.660 B Category 1 0.700 0.037 18.708 Category 2 0.300 0.037 8.018 C Category 1 1.000 0.000 0.000 Category 2 0.000 0.000 0.000
Page 33, Table 3.2 data file
Model: Complete Independence
data: file is "D:\work\mplus_examples\mplus31.raw"; variable: names are coorp uds acc purpose wt; freqweight is wt ; classes = grp(1); categorical = coorp uds acc purpose; analysis: type = mixture;TESTS OF MODEL FIT Loglikelihood H0 Value -2872.230 H0 Scaling Correction Factor 1.000 for MLR Information Criteria Number of Free Parameters 6 Akaike (AIC) 5756.459 Bayesian (BIC) 5787.010 Sample-Size Adjusted BIC 5767.951 (n* = (n + 2) / 24) Chi-Square Test of Model Fit for the Binary and Ordered Categorical (Ordinal) Outcomes Pearson Chi-Square Value 368.666 Degrees of Freedom 29 P-Value 0.0000 Likelihood Ratio Chi-Square Value 257.260 Degrees of Freedom 29 P-Value 0.0000
Model: Two-class Model
data: file is "D:\work\mplus_examples\mplus31.raw"; variable: names are coorp uds acc purpose wt; freqweight is wt ; classes = grp(2); categorical = coorp uds acc purpose; analysis: type = mixture;TESTS OF MODEL FIT Loglikelihood H0 Value -2783.268 H0 Scaling Correction Factor 1.033 for MLR Information Criteria Number of Free Parameters 13 Akaike (AIC) 5592.536 Bayesian (BIC) 5658.729 Sample-Size Adjusted BIC 5617.436 (n* = (n + 2) / 24) Entropy 0.703 Chi-Square Test of Model Fit for the Binary and Ordered Categorical (Ordinal) Outcomes Pearson Chi-Square Value 93.253 Degrees of Freedom 22 P-Value 0.0000 Likelihood Ratio Chi-Square Value 79.337 Degrees of Freedom 22 P-Value 0.0000
Model: Three-class Model
This three-class model is not quite stable, probably because of the small number of categorical manifest variables. At the bottom of page 32, it is pointed out that one of the parameters is actually found to be zero. It is pointed then that “in such circumstances it is customary to reclaim this degree of freedom for testing the model.” To this end, we run the model fixing the parameter, which is prob(P=3|grp=3).
data: file is "D:\work\mplus_examples\mplus31.raw"; variable: names are coorp uds acc purpose wt; freqweight is wt ; classes = grp(3); categorical = coorp uds acc purpose; analysis: type = mixture; model: %grp#3% [coorp$1] (c1); [coorp$2] (c2); model constraint: c1 + c2 = 15;TESTS OF MODEL FIT Loglikelihood H0 Value -2754.545 H0 Scaling Correction Factor 1.026 for MLR Information Criteria Number of Free Parameters 19 Akaike (AIC) 5547.091 Bayesian (BIC) 5643.834 Sample-Size Adjusted BIC 5583.483 (n* = (n + 2) / 24) Entropy 0.667 Chi-Square Test of Model Fit for the Binary and Ordered Categorical (Ordinal) Outcomes Pearson Chi-Square Value 23.532 Degrees of Freedom 16 P-Value 0.1002 Likelihood Ratio Chi-Square Value 21.892 Degrees of Freedom 16 P-Value 0.1467
Table 3.3 on page 35 based on the three-class model in previous example. Notice that the classes defined below are in different order from the book. The Class I labeled as “Ideal” is Latent Class 3, Class II labeled as “Believers” is Latent Class 1 and the Class III labeled as “Skeptics” is Latent Class 2.
data: file is "D:\work\mplus_examples\mplus31.raw"; variable: names are coorp uds acc purpose wt; freqweight is wt ; classes = grp(3); categorical = coorp uds acc purpose; analysis: type = mixture; model: %grp#3% [coorp$1] (c1); [coorp$2] (c2); model constraint: c1 + c2 = 15;RESULTS IN PROBABILITY SCALE Latent Class 1 COORP Category 1 0.690 0.040 17.266 Category 2 0.255 0.037 6.823 Category 3 0.055 0.020 2.787 UDS Category 1 0.313 0.159 1.974 Category 2 0.687 0.159 4.330 ACC Category 1 0.648 0.051 12.690 Category 2 0.352 0.051 6.901 PURPOSE Category 1 0.912 0.045 20.264 Category 2 0.072 0.027 2.677 Category 3 0.017 0.028 0.607 Latent Class 2 COORP Category 1 0.641 0.045 14.357 Category 2 0.256 0.040 6.469 Category 3 0.103 0.026 3.978 UDS Category 1 0.753 0.041 18.504 Category 2 0.247 0.041 6.065 ACC Category 1 0.031 0.071 0.441 Category 2 0.969 0.071 13.654 PURPOSE Category 1 0.143 0.105 1.365 Category 2 0.225 0.044 5.064 Category 3 0.633 0.093 6.808 Latent Class 3 COORP Category 1 0.943 0.019 48.772 Category 2 0.057 0.019 2.942 Category 3 0.000 0.000 0.000 UDS Category 1 1.000 0.000 0.000 Category 2 0.000 0.000 0.000 ACC Category 1 0.613 0.025 24.429 Category 2 0.387 0.025 15.424 PURPOSE Category 1 0.888 0.022 40.039 Category 2 0.053 0.012 4.276 Category 3 0.059 0.016 3.581
Page 40, Table 3.4.
Unrestricted three-class model – (almost unrestricted, see the analysis in previous example)
data: file is "D:\work\mplus_examples\mplus31.raw"; variable: names are coorp uds acc purpose wt; freqweight is wt ; classes = grp(3); categorical = coorp uds acc purpose; analysis: type = mixture; model: %grp#3% [coorp$1] (c1); [coorp$2] (c2); model constraint: c1 + c2 = 15;TESTS OF MODEL FIT Loglikelihood H0 Value -2754.545 H0 Scaling Correction Factor 1.026 for MLR Information Criteria Number of Free Parameters 19 Akaike (AIC) 5547.091 Bayesian (BIC) 5643.834 Sample-Size Adjusted BIC 5583.483 (n* = (n + 2) / 24) Entropy 0.667 Chi-Square Test of Model Fit for the Binary and Ordered Categorical (Ordinal) Outcomes Pearson Chi-Square Value 23.532 Degrees of Freedom 16 P-Value 0.1002 Likelihood Ratio Chi-Square Value 21.892 Degrees of Freedom 16 P-Value 0.1467
Specific value and equality restrictions
title: page 40 - specific value restrictions and equality restrictions data: file is "d:\test\mplus31.dat"; variable: names are p a u c wt; weight is wt (frequency); classes = grp(3); categorical = p a u c; analysis: type = mixture; model: %overall% %grp#1% [a$1] (a11) [p$1] (p11) [p$2] (p21) ! the next two lines deal with the restriction in group 2 of the latent ! variable. You cannot specify probabilities, so you have to work with ! thresholds. A threshold of -15 equals a probability of 0. (15 is ! a default value in MPlus). %grp#2% [a$1@-15]; ! The code from here to the output command gives the restrictions ! for group 3 of the latent variable. %grp#3% [u$1@15]; [a$1] (a13) [p$1] (p13) [p$2] (p23) ! the next four lines set the probability of ideal responding ! "impatient or hostile" to 0. The last category is the reference ! category and you can't make any specifications for that category ! (which is the one that we want to set to 0), so we label the ! probability for category one and two and then add them together ! to sum to 1 (meaning the probability in category three would be ! 0). [c$1] (c1); [c$2] (c2); model constraint: c1 + c2 = 15; a11 = a13; p11 = p13; p21 = p23; ! this output also gives the results shown in Table 3.5 on page 43
RANDOM STARTS RESULTS RANKED FROM THE BEST TO THE WORST LOGLIKELIHOOD VALUES
Initial stage loglikelihood values, seeds, and initial stage start numbers:
-2757.056 127215 9
-2757.930 285380 1
-2757.941 939021 8
-2758.235 253358 2
-2763.091 608496 4
-2763.557 93468 3
-2767.090 903420 5
-2769.596 unperturbed 0
-2773.270 462953 7
-2774.439 415931 10
-2784.372 195873 6
Loglikelihood values at local maxima, seeds, and initial stage start numbers:
-2756.394 127215 9
THE MODEL ESTIMATION TERMINATED NORMALLY
TESTS OF MODEL FIT
Loglikelihood
H0 Value -2756.394
Information Criteria
Number of Free Parameters 14
Akaike (AIC) 5540.788
Bayesian (BIC) 5612.072
Sample-Size Adjusted BIC 5567.603
(n* = (n + 2) / 24)
Entropy 0.673
Chi-Square Test of Model Fit for the Binary and Ordered Categorical
(Ordinal) Outcomes
Pearson Chi-Square
Value 27.028
Degrees of Freedom 21
P-Value 0.1699
Likelihood Ratio Chi-Square
Value 25.589
Degrees of Freedom 21
P-Value 0.2225
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASSES
BASED ON THE ESTIMATED MODEL
Latent
Classes
1 267.50607 0.22255
2 189.70657 0.15783
3 744.78736 0.61962
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASS PATTERNS
BASED ON ESTIMATED POSTERIOR PROBABILITIES
Latent
Classes
1 267.50596 0.22255
2 189.70674 0.15783
3 744.78730 0.61962
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY LATENT CLASS MEMBERSHIP
Class Counts and Proportions
Latent
Classes
1 182 0.15141
2 215 0.17887
3 805 0.66972
Average Latent Class Probabilities for Most Likely Latent Class Membership (Row)
by Latent Class (Column)
1 2 3
1 0.963 0.037 0.000
2 0.052 0.785 0.163
3 0.101 0.018 0.882
MODEL RESULTS
Estimates S.E. Est./S.E.
Latent Class 1
Thresholds
P$1 2.062 0.126 16.358
P$2 2.886 0.175 16.494
A$1 0.479 0.095 5.017
U$1 -0.678 0.632 -1.074
C$1 0.766 0.168 4.552
C$2 2.793 0.352 7.946
Latent Class 2
Thresholds
P$1 -2.088 1.064 -1.963
P$2 -0.670 0.409 -1.640
A$1 -15.000 0.000 0.000
U$1 1.181 0.207 5.698
C$1 0.614 0.188 3.256
C$2 2.163 0.269 8.028
Latent Class 3
Thresholds
P$1 2.062 0.126 16.358
P$2 2.886 0.175 16.494
A$1 0.479 0.095 5.017
U$1 15.000 0.000 0.000
C$1 2.806 0.357 7.869
C$2 15.000 0.000 0.000
Categorical Latent Variables
Means
GRP#1 -1.024 0.293 -3.496
GRP#2 -1.368 0.182 -7.509
RESULTS IN PROBABILITY SCALE
Latent Class 1
P
Category 1 0.887 0.013 70.310
Category 2 0.060 0.009 6.336
Category 3 0.053 0.009 6.035
A
Category 1 0.617 0.023 27.399
Category 2 0.383 0.023 16.978
U
Category 1 0.337 0.141 2.387
Category 2 0.663 0.141 4.704
C
Category 1 0.683 0.036 18.731
Category 2 0.260 0.034 7.617
Category 3 0.058 0.019 3.019
Latent Class 2
P
Category 1 0.110 0.104 1.056
Category 2 0.228 0.045 5.060
Category 3 0.661 0.091 7.230
A
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
U
Category 1 0.765 0.037 20.541
Category 2 0.235 0.037 6.304
C
Category 1 0.649 0.043 15.109
Category 2 0.248 0.037 6.673
Category 3 0.103 0.025 4.138
Latent Class 3
P
Category 1 0.887 0.013 70.310
Category 2 0.060 0.009 6.336
Category 3 0.053 0.009 6.035
A
Category 1 0.617 0.023 27.399
Category 2 0.383 0.023 16.978
U
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
C
Category 1 0.943 0.019 49.201
Category 2 0.057 0.019 2.974
Category 3 0.000 0.000 0.000
ODDS RATIO RESULTS
Latent Class 1 Compared to Latent Class 2
P
Category > 1 0.016 0.017 0.938
Category > 2 0.029 0.013 2.219
A
Category > 1 0.000 0.000 10.482
U
Category > 1 6.420 4.356 1.474
C
Category > 1 0.859 0.224 3.833
Category > 2 0.533 0.248 2.145
Latent Class 1 Compared to Latent Class 3
P
Category > 1 1.000 0.000 999.000
Category > 2 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
U
Category > 1 ********* ******* 1.583
C
Category > 1 7.693 3.070 2.506
Category > 2 ********* 0.000 999.000
Latent Class 2 Compared to Latent Class 3
P
Category > 1 63.475 67.677 0.938
Category > 2 35.005 15.778 2.219
A
Category > 1 ********* ******* 10.482
U
Category > 1 ********* ******* 4.824
C
Category > 1 8.957 3.676 2.437
Category > 2 ********* 0.000 999.000
QUALITY OF NUMERICAL RESULTS
Condition Number for the Information Matrix 0.303E-02
(ratio of smallest to largest eigenvalue)
Page 47 data file
title: page 48 - Table 4.2 data: file is "d:\test\mplus47.txt"; variable: names are w a i v wt; weight is wt (frequency); classes = grp(1); categorical = w a i v; analysis: type = mixture;
THE MODEL ESTIMATION TERMINATED NORMALLY
TESTS OF MODEL FIT
Loglikelihood
H0 Value -2346.675
Information Criteria
Number of Free Parameters 4
Akaike (AIC) 4701.349
Bayesian (BIC) 4722.332
Sample-Size Adjusted BIC 4709.625
(n* = (n + 2) / 24)
Chi-Square Test of Model Fit for the Binary and Ordered Categorical
(Ordinal) Outcomes
Pearson Chi-Square
Value 798.963
Degrees of Freedom 11
P-Value 0.0000
Likelihood Ratio Chi-Square
Value 262.266
Degrees of Freedom 11
P-Value 0.0000
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASSES
BASED ON THE ESTIMATED MODEL
Latent
Classes
1 1402.00000 1.00000
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASS PATTERNS
BASED ON ESTIMATED POSTERIOR PROBABILITIES
Latent
Classes
1 1402.00000 1.00000
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY LATENT CLASS MEMBERSHIP
Class Counts and Proportions
Latent
Classes
1 1402 1.00000
Average Latent Class Probabilities for Most Likely Latent Class Membership (Row)
by Latent Class (Column)
1
1 1.000
MODEL RESULTS
Estimates S.E. Est./S.E.
Latent Class 1
Thresholds
W$1 3.297 0.144 22.896
A$1 2.504 0.101 24.783
I$1 0.565 0.056 10.173
V$1 -0.922 0.059 -15.574
RESULTS IN PROBABILITY SCALE
Latent Class 1
W
Category 1 0.964 0.005 194.705
Category 2 0.036 0.005 7.201
A
Category 1 0.924 0.007 130.925
Category 2 0.076 0.007 10.708
I
Category 1 0.638 0.013 49.672
Category 2 0.362 0.013 28.225
V
Category 1 0.285 0.012 23.616
Category 2 0.715 0.012 59.366
QUALITY OF NUMERICAL RESULTS
Condition Number for the Information Matrix 0.992E-01
(ratio of smallest to largest eigenvalue)
Page 51
Proctor’s model
title: page 51 - Table 4.4 - Proctor's Model data: file is "d:\test\mplus47.txt"; variable: names are w a i v wt; weight is wt (frequency); classes = grp(5); categorical = w a i v; analysis: type = mixture; model: %overall% [w$1 a$1 i$1 v$1] (p5); %grp#1% [w$1 a$1 i$1 v$1] (p1); %grp#2% [w$1 a$1 i$1] (p1); [v$1] (p5); %grp#3% [w$1 a$1] (p1); [i$1 v$1] (p5); %grp#4% [w$1] (p1); [a$1 i$1 v$1] (p5); model constraint: p1 = -p5; ! The group in the overall statement is ! group 5. It is specified here because in ! MPlus you cannot specify anything regarding ! the last category. Because there is only one ! label, each of the values in the square brackets ! is set equal to one another. The rest of the ! groups are set up according to the subscripts ! on the letters on pages 49-50 (4.1), which match ! the response patterns in Table 4.3 on page 48. ! In the model constraint statement, p1 is set ! equal to -p5 because p5 is really for group 2, ! but since you can't specify group 2 (b/c it is ! the reference category), it is coded as 1, so the ! minus sign indicates that it should be group 2.
RANDOM STARTS RESULTS RANKED FROM THE BEST TO THE WORST LOGLIKELIHOOD VALUES
Initial stage loglikelihood values, seeds, and initial stage start numbers:
-2284.691 903420 5
-2284.694 608496 4
-2284.705 462953 7
-2284.716 93468 3
-2284.804 285380 1
-2285.281 195873 6
-2289.136 939021 8
-2295.433 127215 9
-3411.273 415931 10
-3411.297 253358 2
-3411.402 unperturbed 0
Loglikelihood values at local maxima, seeds, and initial stage start numbers:
-2284.639 903420 5
THE MODEL ESTIMATION TERMINATED NORMALLY
TESTS OF MODEL FIT
Loglikelihood
H0 Value -2284.639
Information Criteria
Number of Free Parameters 5
Akaike (AIC) 4579.277
Bayesian (BIC) 4605.506
Sample-Size Adjusted BIC 4589.623
(n* = (n + 2) / 24)
Entropy 0.794
Chi-Square Test of Model Fit for the Binary and Ordered Categorical
(Ordinal) Outcomes
Pearson Chi-Square
Value 137.461
Degrees of Freedom 10
P-Value 0.0000
Likelihood Ratio Chi-Square
Value 138.194
Degrees of Freedom 10
P-Value 0.0000
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASSES
BASED ON THE ESTIMATED MODEL
Latent
Classes
1 340.63236 0.24296
2 597.15714 0.42593
3 409.25098 0.29191
4 26.77394 0.01910
5 28.18558 0.02010
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASS PATTERNS
BASED ON ESTIMATED POSTERIOR PROBABILITIES
Latent
Classes
1 340.63249 0.24296
2 597.15699 0.42593
3 409.25052 0.29190
4 26.77447 0.01910
5 28.18554 0.02010
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY LATENT CLASS MEMBERSHIP
Class Counts and Proportions
Latent
Classes
1 313 0.22325
2 581 0.41441
3 438 0.31241
4 43 0.03067
5 27 0.01926
Average Latent Class Probabilities for Most Likely Latent Class Membership (Row)
by Latent Class (Column)
1 2 3 4 5
1 0.921 0.077 0.003 0.000 0.000
2 0.026 0.940 0.031 0.002 0.002
3 0.084 0.059 0.852 0.003 0.002
4 0.018 0.027 0.391 0.537 0.027
5 0.000 0.002 0.031 0.042 0.925
MODEL RESULTS
Estimates S.E. Est./S.E.
Latent Class 1
Thresholds
W$1 3.043 0.083 36.451
A$1 3.043 0.083 36.451
I$1 3.043 0.083 36.451
V$1 3.043 0.083 36.451
Latent Class 2
Thresholds
W$1 3.043 0.083 36.451
A$1 3.043 0.083 36.451
I$1 3.043 0.083 36.451
V$1 -3.043 0.083 -36.451
Latent Class 3
Thresholds
W$1 3.043 0.083 36.451
A$1 3.043 0.083 36.451
I$1 -3.043 0.083 -36.451
V$1 -3.043 0.083 -36.451
Latent Class 4
Thresholds
W$1 3.043 0.083 36.451
A$1 -3.043 0.083 -36.451
I$1 -3.043 0.083 -36.451
V$1 -3.043 0.083 -36.451
Latent Class 5
Thresholds
W$1 -3.043 0.083 -36.451
A$1 -3.043 0.083 -36.451
I$1 -3.043 0.083 -36.451
V$1 -3.043 0.083 -36.451
Categorical Latent Variables
Means
GRP#1 2.492 0.215 11.584
GRP#2 3.053 0.212 14.421
GRP#3 2.676 0.215 12.444
GRP#4 -0.051 0.373 -0.138
RESULTS IN PROBABILITY SCALE
Latent Class 1
W
Category 1 0.955 0.004 263.231
Category 2 0.045 0.004 12.548
A
Category 1 0.955 0.004 263.231
Category 2 0.045 0.004 12.548
I
Category 1 0.955 0.004 263.231
Category 2 0.045 0.004 12.548
V
Category 1 0.955 0.004 263.231
Category 2 0.045 0.004 12.548
Latent Class 2
W
Category 1 0.955 0.004 263.231
Category 2 0.045 0.004 12.548
A
Category 1 0.955 0.004 263.231
Category 2 0.045 0.004 12.548
I
Category 1 0.955 0.004 263.231
Category 2 0.045 0.004 12.548
V
Category 1 0.045 0.004 12.548
Category 2 0.955 0.004 263.231
Latent Class 3
W
Category 1 0.955 0.004 263.231
Category 2 0.045 0.004 12.548
A
Category 1 0.955 0.004 263.231
Category 2 0.045 0.004 12.548
I
Category 1 0.045 0.004 12.548
Category 2 0.955 0.004 263.231
V
Category 1 0.045 0.004 12.548
Category 2 0.955 0.004 263.231
Latent Class 4
W
Category 1 0.955 0.004 263.231
Category 2 0.045 0.004 12.548
A
Category 1 0.045 0.004 12.548
Category 2 0.955 0.004 263.231
I
Category 1 0.045 0.004 12.548
Category 2 0.955 0.004 263.231
V
Category 1 0.045 0.004 12.548
Category 2 0.955 0.004 263.231
Latent Class 5
W
Category 1 0.045 0.004 12.548
Category 2 0.955 0.004 263.231
A
Category 1 0.045 0.004 12.548
Category 2 0.955 0.004 263.231
I
Category 1 0.045 0.004 12.548
Category 2 0.955 0.004 263.231
V
Category 1 0.045 0.004 12.548
Category 2 0.955 0.004 263.231
ODDS RATIO RESULTS
Latent Class 1 Compared to Latent Class 2
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 0.002 0.000 5.988
Latent Class 1 Compared to Latent Class 3
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.002 0.000 5.988
V
Category > 1 0.002 0.000 5.988
Latent Class 1 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.002 0.000 5.988
I
Category > 1 0.002 0.000 5.988
V
Category > 1 0.002 0.000 5.988
Latent Class 1 Compared to Latent Class 5
W
Category > 1 0.002 0.000 5.988
A
Category > 1 0.002 0.000 5.988
I
Category > 1 0.002 0.000 5.988
V
Category > 1 0.002 0.000 5.988
Latent Class 2 Compared to Latent Class 3
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.002 0.000 5.988
V
Category > 1 1.000 0.000 999.000
Latent Class 2 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.002 0.000 5.988
I
Category > 1 0.002 0.000 5.988
V
Category > 1 1.000 0.000 999.000
Latent Class 2 Compared to Latent Class 5
W
Category > 1 0.002 0.000 5.988
A
Category > 1 0.002 0.000 5.988
I
Category > 1 0.002 0.000 5.988
V
Category > 1 1.000 0.000 999.000
Latent Class 3 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.002 0.000 5.988
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 3 Compared to Latent Class 5
W
Category > 1 0.002 0.000 5.988
A
Category > 1 0.002 0.000 5.988
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 4 Compared to Latent Class 5
W
Category > 1 0.002 0.000 5.988
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
QUALITY OF NUMERICAL RESULTS
Condition Number for the Information Matrix 0.256E-02
(ratio of smallest to largest eigenvalue)
Item-specific error rates
title: page 51 - Table 4.4 - Item Specific Error Rates data: file is "d:\test\mplus47.txt"; variable: names are w a i v wt; weight is wt (frequency); classes = grp(5); categorical = w a i v; analysis: type = mixture; model: %overall% [w$1] (q1); [a$1] (q2); [i$1] (q3); [v$1] (q4); %grp#1% [w$1] (p1); [a$1] (p2); [i$1] (p3); [v$1] (p4); %grp#2% [w$1] (p1); [a$1] (p2); [i$1] (p3); [v$1] (q4); %grp#3% [w$1] (p1); [a$1] (p2); [i$1] (q3); [v$1] (q4); %grp#4% [w$1] (p1); [a$1] (q2); [i$1] (q3); [v$1] (q4); model constraint: p1 = -q1; p2 = -q2; p3 = -q3; p4 = -q4;
RANDOM STARTS RESULTS RANKED FROM THE BEST TO THE WORST LOGLIKELIHOOD VALUES
Initial stage loglikelihood values, seeds, and initial stage start numbers:
-2239.056 93468 3
-2239.783 285380 1
-2240.521 253358 2
-2244.311 608496 4
-2284.936 415931 10
-2286.090 195873 6
-2286.373 unperturbed 0
-2286.493 903420 5
-2309.477 939021 8
-2346.675 462953 7
-2346.675 127215 9
Loglikelihood values at local maxima, seeds, and initial stage start numbers:
-2233.791 93468 3
THE MODEL ESTIMATION TERMINATED NORMALLY
TESTS OF MODEL FIT
Loglikelihood
H0 Value -2233.791
Information Criteria
Number of Free Parameters 8
Akaike (AIC) 4483.582
Bayesian (BIC) 4525.547
Sample-Size Adjusted BIC 4500.134
(n* = (n + 2) / 24)
Entropy 0.781
Chi-Square Test of Model Fit for the Binary and Ordered Categorical
(Ordinal) Outcomes
Pearson Chi-Square
Value 36.672
Degrees of Freedom 7
P-Value 0.0000
Likelihood Ratio Chi-Square
Value 36.499
Degrees of Freedom 7
P-Value 0.0000
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASSES
BASED ON THE ESTIMATED MODEL
Latent
Classes
1 398.99988 0.28459
2 641.78414 0.45776
3 284.58376 0.20298
4 46.33341 0.03305
5 30.29881 0.02161
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASS PATTERNS
BASED ON ESTIMATED POSTERIOR PROBABILITIES
Latent
Classes
1 398.99988 0.28459
2 641.78402 0.45776
3 284.58389 0.20298
4 46.33340 0.03305
5 30.29881 0.02161
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY LATENT CLASS MEMBERSHIP
Class Counts and Proportions
Latent
Classes
1 399 0.28459
2 579 0.41298
3 355 0.25321
4 40 0.02853
5 29 0.02068
Average Latent Class Probabilities for Most Likely Latent Class Membership (Row)
by Latent Class (Column)
1 2 3 4 5
1 1.000 0.000 0.000 0.000 0.000
2 0.000 0.872 0.105 0.023 0.000
3 0.000 0.376 0.617 0.003 0.005
4 0.000 0.076 0.125 0.791 0.008
5 0.000 0.004 0.003 0.022 0.970
MODEL RESULTS
Estimates S.E. Est./S.E.
Latent Class 1
Thresholds
W$1 4.206 0.230 18.322
A$1 3.661 0.287 12.758
I$1 1.307 0.119 10.951
V$1 15.384 0.000 0.000
Latent Class 2
Thresholds
W$1 4.206 0.230 18.322
A$1 3.661 0.287 12.758
I$1 1.307 0.119 10.951
V$1 -15.384 0.000 0.000
Latent Class 3
Thresholds
W$1 4.206 0.230 18.322
A$1 3.661 0.287 12.758
I$1 -1.307 0.119 -10.951
V$1 -15.384 0.000 0.000
Latent Class 4
Thresholds
W$1 4.206 0.230 18.322
A$1 -3.661 0.287 -12.758
I$1 -1.307 0.119 -10.951
V$1 -15.384 0.000 0.000
Latent Class 5
Thresholds
W$1 -4.206 0.230 -18.322
A$1 -3.661 0.287 -12.758
I$1 -1.307 0.119 -10.951
V$1 -15.384 0.000 0.000
Categorical Latent Variables
Means
GRP#1 2.578 0.201 12.851
GRP#2 3.053 0.202 15.080
GRP#3 2.240 0.218 10.276
GRP#4 0.425 0.319 1.330
RESULTS IN PROBABILITY SCALE
Latent Class 1
W
Category 1 0.985 0.003 296.605
Category 2 0.015 0.003 4.421
A
Category 1 0.975 0.007 139.052
Category 2 0.025 0.007 3.574
I
Category 1 0.787 0.020 39.338
Category 2 0.213 0.020 10.641
V
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
Latent Class 2
W
Category 1 0.985 0.003 296.605
Category 2 0.015 0.003 4.421
A
Category 1 0.975 0.007 139.052
Category 2 0.025 0.007 3.574
I
Category 1 0.787 0.020 39.338
Category 2 0.213 0.020 10.641
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 3
W
Category 1 0.985 0.003 296.605
Category 2 0.015 0.003 4.421
A
Category 1 0.975 0.007 139.052
Category 2 0.025 0.007 3.574
I
Category 1 0.213 0.020 10.641
Category 2 0.787 0.020 39.338
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 4
W
Category 1 0.985 0.003 296.605
Category 2 0.015 0.003 4.421
A
Category 1 0.025 0.007 3.574
Category 2 0.975 0.007 139.052
I
Category 1 0.213 0.020 10.641
Category 2 0.787 0.020 39.338
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 5
W
Category 1 0.015 0.003 4.421
Category 2 0.985 0.003 296.605
A
Category 1 0.025 0.007 3.574
Category 2 0.975 0.007 139.052
I
Category 1 0.213 0.020 10.641
Category 2 0.787 0.020 39.338
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
ODDS RATIO RESULTS
Latent Class 1 Compared to Latent Class 2
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 3
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.073 0.017 4.188
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.001 0.000 1.742
I
Category > 1 0.073 0.017 4.188
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 5
W
Category > 1 0.000 0.000 2.178
A
Category > 1 0.001 0.000 1.742
I
Category > 1 0.073 0.017 4.188
V
Category > 1 0.000 0.000 999.000
Latent Class 2 Compared to Latent Class 3
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.073 0.017 4.188
V
Category > 1 1.000 0.000 999.000
Latent Class 2 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.001 0.000 1.742
I
Category > 1 0.073 0.017 4.188
V
Category > 1 1.000 0.000 999.000
Latent Class 2 Compared to Latent Class 5
W
Category > 1 0.000 0.000 2.178
A
Category > 1 0.001 0.000 1.742
I
Category > 1 0.073 0.017 4.188
V
Category > 1 1.000 0.000 999.000
Latent Class 3 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.001 0.000 1.742
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 3 Compared to Latent Class 5
W
Category > 1 0.000 0.000 2.178
A
Category > 1 0.001 0.000 1.742
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 4 Compared to Latent Class 5
W
Category > 1 0.000 0.000 2.178
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
QUALITY OF NUMERICAL RESULTS
Condition Number for the Information Matrix 0.225E-02
(ratio of smallest to largest eigenvalue)
True-type-specific error rates
title: page 51 - Table 4.4 - True-type-specific error rates data: file is "d:\test\mplus47.txt"; variable: names are w a i v wt; weight is wt (frequency); classes = grp(5); categorical = w a i v; analysis: type = mixture; model: %overall% [w$1 a$1 i$1 v$1] (5); ! last line on page 53 %grp#1% [w$1 a$1 i$1 v$1] (p1); %grp#2% [w$1 a$1 i$1] (p2); [v$1] (q2); %grp#3% [w$1 a$1] (p3); [i$1 v$1] (q3); %grp#4% [w$1] (p4); [a$1 i$1 v$1] (q4); model constraint: p2 = -q2; p3 = -q3; p4 = -q4; p1 = -15; ! This last constraint is set to -15 instead of -q5 because when you run ! the model with p1 = - q5;, you find that the parameter is 0. Hence, you ! see the note in the text indicating that this parameter is set to 0 and the ! number of degrees of freedom increases from 6 to 7. ! In the labeling above, the numbers indicate the number of the latent class, ! p indicates the probability of responding "yes" to the item (a 1 in the subscripts ! in the text), and q indicates the probability of responding "no" to the item ! (a 2 in the subscripts).
RANDOM STARTS RESULTS RANKED FROM THE BEST TO THE WORST LOGLIKELIHOOD VALUES
Initial stage loglikelihood values, seeds, and initial stage start numbers:
-2261.052 285380 1
-2261.103 unperturbed 0
-2262.554 93468 3
-2262.895 127215 9
-2267.259 608496 4
-2269.834 903420 5
-2271.668 253358 2
-2279.663 415931 10
-2280.679 939021 8
-2292.788 462953 7
-2351.977 195873 6
Loglikelihood values at local maxima, seeds, and initial stage start numbers:
-2260.050 285380 1
THE MODEL ESTIMATION TERMINATED NORMALLY
TESTS OF MODEL FIT
Loglikelihood
H0 Value -2260.050
Information Criteria
Number of Free Parameters 8
Akaike (AIC) 4536.101
Bayesian (BIC) 4578.066
Sample-Size Adjusted BIC 4552.653
(n* = (n + 2) / 24)
Entropy 0.836
Chi-Square Test of Model Fit for the Binary and Ordered Categorical
(Ordinal) Outcomes
Pearson Chi-Square
Value 86.381
Degrees of Freedom 7
P-Value 0.0000
Likelihood Ratio Chi-Square
Value 89.018
Degrees of Freedom 7
P-Value 0.0000
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASSES
BASED ON THE ESTIMATED MODEL
Latent
Classes
1 21.86222 0.01559
2 555.51624 0.39623
3 517.52214 0.36913
4 6.37973 0.00455
5 300.71967 0.21449
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASS PATTERNS
BASED ON ESTIMATED POSTERIOR PROBABILITIES
Latent
Classes
1 21.86229 0.01559
2 555.51638 0.39623
3 517.52211 0.36913
4 6.37967 0.00455
5 300.71956 0.21449
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY LATENT CLASS MEMBERSHIP
Class Counts and Proportions
Latent
Classes
1 27 0.01926
2 579 0.41298
3 483 0.34451
4 0 0.00000
5 313 0.22325
Average Latent Class Probabilities for Most Likely Latent Class Membership (Row)
by Latent Class (Column)
1 2 3 4 5
1 0.810 0.000 0.167 0.023 0.000
2 0.000 0.912 0.083 0.002 0.004
3 0.000 0.030 0.953 0.009 0.008
4 0.000 0.000 0.000 0.000 0.000
5 0.000 0.042 0.016 0.000 0.941
MODEL RESULTS
Estimates S.E. Est./S.E.
Latent Class 1
Thresholds
W$1 -15.000 0.000 0.000
A$1 -15.000 0.000 0.000
I$1 -15.000 0.000 0.000
V$1 -15.000 0.000 0.000
Latent Class 2
Thresholds
W$1 3.664 0.238 15.390
A$1 3.664 0.238 15.390
I$1 3.664 0.238 15.390
V$1 -3.664 0.238 -15.390
Latent Class 3
Thresholds
W$1 2.152 0.099 21.746
A$1 2.152 0.099 21.746
I$1 -2.152 0.099 -21.746
V$1 -2.152 0.099 -21.746
Latent Class 4
Thresholds
W$1 1.564 0.221 7.070
A$1 -1.564 0.221 -7.070
I$1 -1.564 0.221 -7.070
V$1 -1.564 0.221 -7.070
Latent Class 5
Thresholds
W$1 4.985 0.796 6.259
A$1 4.985 0.796 6.259
I$1 4.985 0.796 6.259
V$1 4.985 0.796 6.259
Categorical Latent Variables
Means
GRP#1 -2.621 0.253 -10.377
GRP#2 0.614 0.084 7.331
GRP#3 0.543 0.089 6.105
GRP#4 -3.853 2.289 -1.683
RESULTS IN PROBABILITY SCALE
Latent Class 1
W
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
A
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
I
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 2
W
Category 1 0.975 0.006 168.031
Category 2 0.025 0.006 4.309
A
Category 1 0.975 0.006 168.031
Category 2 0.025 0.006 4.309
I
Category 1 0.975 0.006 168.031
Category 2 0.025 0.006 4.309
V
Category 1 0.025 0.006 4.309
Category 2 0.975 0.006 168.031
Latent Class 3
W
Category 1 0.896 0.009 97.029
Category 2 0.104 0.009 11.280
A
Category 1 0.896 0.009 97.029
Category 2 0.104 0.009 11.280
I
Category 1 0.104 0.009 11.280
Category 2 0.896 0.009 97.029
V
Category 1 0.104 0.009 11.280
Category 2 0.896 0.009 97.029
Latent Class 4
W
Category 1 0.827 0.032 26.120
Category 2 0.173 0.032 5.465
A
Category 1 0.173 0.032 5.465
Category 2 0.827 0.032 26.120
I
Category 1 0.173 0.032 5.465
Category 2 0.827 0.032 26.120
V
Category 1 0.173 0.032 5.465
Category 2 0.827 0.032 26.120
Latent Class 5
W
Category 1 0.993 0.005 184.802
Category 2 0.007 0.005 1.264
A
Category 1 0.993 0.005 184.802
Category 2 0.007 0.005 1.264
I
Category 1 0.993 0.005 184.802
Category 2 0.007 0.005 1.264
V
Category 1 0.993 0.005 184.802
Category 2 0.007 0.005 1.264
ODDS RATIO RESULTS
Latent Class 1 Compared to Latent Class 2
W
Category > 1 ********* 0.000 999.000
A
Category > 1 ********* 0.000 999.000
I
Category > 1 ********* 0.000 999.000
V
Category > 1 83825.578 0.000 999.000
Latent Class 1 Compared to Latent Class 3
W
Category > 1 ********* 0.000 999.000
A
Category > 1 ********* 0.000 999.000
I
Category > 1 ********* 0.000 999.000
V
Category > 1 ********* 0.000 999.000
Latent Class 1 Compared to Latent Class 4
W
Category > 1 ********* 0.000 999.000
A
Category > 1 ********* 0.000 999.000
I
Category > 1 ********* 0.000 999.000
V
Category > 1 ********* 0.000 999.000
Latent Class 1 Compared to Latent Class 5
W
Category > 1 ********* 0.000 999.000
A
Category > 1 ********* 0.000 999.000
I
Category > 1 ********* 0.000 999.000
V
Category > 1 ********* 0.000 999.000
Latent Class 2 Compared to Latent Class 3
W
Category > 1 0.221 0.057 3.887
A
Category > 1 0.221 0.057 3.887
I
Category > 1 0.003 0.001 3.872
V
Category > 1 4.534 1.166 3.887
Latent Class 2 Compared to Latent Class 4
W
Category > 1 0.123 0.040 3.057
A
Category > 1 0.005 0.002 3.097
I
Category > 1 0.005 0.002 3.097
V
Category > 1 8.159 2.669 3.057
Latent Class 2 Compared to Latent Class 5
W
Category > 1 3.749 3.137 1.195
A
Category > 1 3.749 3.137 1.195
I
Category > 1 3.749 3.137 1.195
V
Category > 1 5701.205 ******* 1.211
Latent Class 3 Compared to Latent Class 4
W
Category > 1 0.556 0.134 4.137
A
Category > 1 0.024 0.006 4.114
I
Category > 1 1.800 0.435 4.137
V
Category > 1 1.800 0.435 4.137
Latent Class 3 Compared to Latent Class 5
W
Category > 1 16.995 14.027 1.212
A
Category > 1 16.995 14.027 1.212
I
Category > 1 1257.531 979.903 1.283
V
Category > 1 1257.531 979.903 1.283
Latent Class 4 Compared to Latent Class 5
W
Category > 1 30.586 25.360 1.206
A
Category > 1 698.762 575.895 1.213
I
Category > 1 698.762 575.895 1.213
V
Category > 1 698.762 575.895 1.213
QUALITY OF NUMERICAL RESULTS
Condition Number for the Information Matrix 0.226E-03
(ratio of smallest to largest eigenvalue)
Lazarsfeld’s latent distance model
Page 56
title: page 56 - Table 4.5 - Error Rates and Scale-Type Proportions Estimated Under Lazarsfeld's Latent Distance Model data: file is "d:\test\mplus47.txt"; variable: names are w a i v wt; weight is wt (frequency); classes = grp(5); categorical = w a i v; analysis: type = mixture; model: %overall% ! group 5 (second subscript) [w$1] (q1); [a$1] (2); [i$1] (3); [v$1] (q4); %grp#1% [w$1] (p1); [a$1] (1); [i$1] (4); [v$1] (p4); %grp#2% [w$1] (p1); [a$1] (1); [i$1] (4); [v$1] (q4); %grp#3% [w$1] (p1); [a$1] (1); [i$1] (3); [v$1] (q4); %grp#4% [w$1] (p1); [a$1] (2); [i$1] (3); [v$1] (q4); model constraint: p1 = -q1; p4 = -q4;
RANDOM STARTS RESULTS RANKED FROM THE BEST TO THE WORST LOGLIKELIHOOD VALUES
Initial stage loglikelihood values, seeds, and initial stage start numbers:
-2241.097 253358 2
-2247.678 285380 1
-2267.091 93468 3
-2271.942 unperturbed 0
-2273.677 195873 6
-2273.847 903420 5
-2273.870 415931 10
-2274.709 127215 9
-2275.289 608496 4
-2301.882 939021 8
-2346.675 462953 7
Loglikelihood values at local maxima, seeds, and initial stage start numbers:
-2222.920 253358 2
THE MODEL ESTIMATION TERMINATED NORMALLY
TESTS OF MODEL FIT
Loglikelihood
H0 Value -2222.920
Information Criteria
Number of Free Parameters 10
Akaike (AIC) 4465.841
Bayesian (BIC) 4518.298
Sample-Size Adjusted BIC 4486.531
(n* = (n + 2) / 24)
Entropy 0.774
Chi-Square Test of Model Fit for the Binary and Ordered Categorical
(Ordinal) Outcomes
Pearson Chi-Square
Value 12.339
Degrees of Freedom 5
P-Value 0.0304
Likelihood Ratio Chi-Square
Value 14.758
Degrees of Freedom 5
P-Value 0.0114
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASSES
BASED ON THE ESTIMATED MODEL
Latent
Classes
1 398.99995 0.28459
2 670.75157 0.47842
3 217.39038 0.15506
4 67.86897 0.04841
5 46.98914 0.03352
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASS PATTERNS
BASED ON ESTIMATED POSTERIOR PROBABILITIES
Latent
Classes
1 398.99992 0.28459
2 670.75160 0.47842
3 217.39038 0.15506
4 67.86894 0.04841
5 46.98917 0.03352
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY LATENT CLASS MEMBERSHIP
Class Counts and Proportions
Latent
Classes
1 399 0.28459
2 575 0.41013
3 339 0.24180
4 40 0.02853
5 49 0.03495
Average Latent Class Probabilities for Most Likely Latent Class Membership (Row)
by Latent Class (Column)
1 2 3 4 5
1 1.000 0.000 0.000 0.000 0.000
2 0.000 0.913 0.062 0.025 0.000
3 0.000 0.415 0.521 0.064 0.000
4 0.000 0.090 0.113 0.795 0.001
5 0.000 0.030 0.010 0.003 0.957
MODEL RESULTS
Estimates S.E. Est./S.E.
Latent Class 1
Thresholds
W$1 6.072 0.997 6.087
A$1 3.569 0.504 7.084
I$1 1.292 0.122 10.611
V$1 15.771 0.000 0.000
Latent Class 2
Thresholds
W$1 6.072 0.997 6.087
A$1 3.569 0.504 7.084
I$1 1.292 0.122 10.611
V$1 -15.771 0.000 0.000
Latent Class 3
Thresholds
W$1 6.072 0.997 6.087
A$1 3.569 0.504 7.084
I$1 -1.621 0.887 -1.829
V$1 -15.771 0.000 0.000
Latent Class 4
Thresholds
W$1 6.072 0.997 6.087
A$1 -0.471 0.337 -1.399
I$1 -1.621 0.887 -1.829
V$1 -15.771 0.000 0.000
Latent Class 5
Thresholds
W$1 -6.072 0.997 -6.087
A$1 -0.471 0.337 -1.399
I$1 -1.621 0.887 -1.829
V$1 -15.771 0.000 0.000
Categorical Latent Variables
Means
GRP#1 2.139 0.158 13.581
GRP#2 2.658 0.189 14.049
GRP#3 1.532 0.276 5.559
GRP#4 0.368 0.528 0.696
RESULTS IN PROBABILITY SCALE
Latent Class 1
W
Category 1 0.998 0.002 435.546
Category 2 0.002 0.002 1.005
A
Category 1 0.973 0.013 72.411
Category 2 0.027 0.013 2.041
I
Category 1 0.784 0.021 38.107
Category 2 0.216 0.021 10.470
V
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
Latent Class 2
W
Category 1 0.998 0.002 435.546
Category 2 0.002 0.002 1.005
A
Category 1 0.973 0.013 72.411
Category 2 0.027 0.013 2.041
I
Category 1 0.784 0.021 38.107
Category 2 0.216 0.021 10.470
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 3
W
Category 1 0.998 0.002 435.546
Category 2 0.002 0.002 1.005
A
Category 1 0.973 0.013 72.411
Category 2 0.027 0.013 2.041
I
Category 1 0.165 0.122 1.351
Category 2 0.835 0.122 6.836
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 4
W
Category 1 0.998 0.002 435.546
Category 2 0.002 0.002 1.005
A
Category 1 0.384 0.080 4.820
Category 2 0.616 0.080 7.724
I
Category 1 0.165 0.122 1.351
Category 2 0.835 0.122 6.836
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 5
W
Category 1 0.002 0.002 1.005
Category 2 0.998 0.002 435.546
A
Category 1 0.384 0.080 4.820
Category 2 0.616 0.080 7.724
I
Category 1 0.165 0.122 1.351
Category 2 0.835 0.122 6.836
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
ODDS RATIO RESULTS
Latent Class 1 Compared to Latent Class 2
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 3
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.054 0.048 1.124
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.018 0.009 1.910
I
Category > 1 0.054 0.048 1.124
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 5
W
Category > 1 0.000 0.000 0.501
A
Category > 1 0.018 0.009 1.910
I
Category > 1 0.054 0.048 1.124
V
Category > 1 0.000 0.000 999.000
Latent Class 2 Compared to Latent Class 3
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.054 0.048 1.124
V
Category > 1 1.000 0.000 999.000
Latent Class 2 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.018 0.009 1.910
I
Category > 1 0.054 0.048 1.124
V
Category > 1 1.000 0.000 999.000
Latent Class 2 Compared to Latent Class 5
W
Category > 1 0.000 0.000 0.501
A
Category > 1 0.018 0.009 1.910
I
Category > 1 0.054 0.048 1.124
V
Category > 1 1.000 0.000 999.000
Latent Class 3 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.018 0.009 1.910
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 3 Compared to Latent Class 5
W
Category > 1 0.000 0.000 0.501
A
Category > 1 0.018 0.009 1.910
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 4 Compared to Latent Class 5
W
Category > 1 0.000 0.000 0.501
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
QUALITY OF NUMERICAL RESULTS
Condition Number for the Information Matrix 0.209E-02
(ratio of smallest to largest eigenvalue)
Page 58
Intrinsically unscalable
title: page 58 - Table 4.6 - Intrinsically Unscalable data: file is "d:\test\mplus47.txt"; variable: names are w a i v wt; weight is wt (frequency); classes = grp(6); categorical = w a i v; analysis: type = mixture; model: %overall% %grp#1% [w$1@15 a$1@15 i$1@15 v$1@15]; %grp#2% [w$1@15 a$1@15 i$1@15]; [v$1@-15]; %grp#3% [w$1@15 a$1@15]; [i$1@-15 v$1@-15]; %grp#4% [w$1@15]; [a$1@-15 i$1@-15 v$1@-15]; %grp#5% [w$1@-15 a$1@-15 i$1@-15 v$1@-15]; ! group 6, which is coded in %overall%, is the ! unscalable portion, which has no restrictions. ! The @ symbols are used to fix the thresholds. ! You fix the thresholds and not the probabilities ! in MPlus.
RANDOM STARTS RESULTS RANKED FROM THE BEST TO THE WORST LOGLIKELIHOOD VALUES
Initial stage loglikelihood values, seeds, and initial stage start numbers:
-2234.693 195873 6
-2235.407 939021 8
-2236.547 608496 4
-2237.463 253358 2
-2238.227 903420 5
-2239.604 93468 3
-2240.706 unperturbed 0
-2242.429 285380 1
-2260.803 415931 10
-2264.477 462953 7
-2267.448 127215 9
Loglikelihood values at local maxima, seeds, and initial stage start numbers:
-2225.804 195873 6
THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES MAY NOT BE
TRUSTWORTHY FOR SOME PARAMETERS DUE TO A NON-POSITIVE DEFINITE
FIRST-ORDER DERIVATIVE PRODUCT MATRIX. THIS MAY BE DUE TO THE STARTING
VALUES BUT MAY ALSO BE AN INDICATION OF MODEL NONIDENTIFICATION. THE
CONDITION NUMBER IS 0.238D-17. PROBLEM INVOLVING PARAMETER 7.
THE MODEL ESTIMATION TERMINATED NORMALLY
TESTS OF MODEL FIT
Loglikelihood
H0 Value -2225.804
Information Criteria
Number of Free Parameters 9
Akaike (AIC) 4469.608
Bayesian (BIC) 4516.818
Sample-Size Adjusted BIC 4488.229
(n* = (n + 2) / 24)
Entropy 0.785
Chi-Square Test of Model Fit for the Binary and Ordered Categorical
(Ordinal) Outcomes
Pearson Chi-Square
Value 17.526
Degrees of Freedom 6
P-Value 0.0075
Likelihood Ratio Chi-Square
Value 20.525
Degrees of Freedom 6
P-Value 0.0022
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASSES
BASED ON THE ESTIMATED MODEL
Latent
Classes
1 256.11841 0.18268
2 283.72303 0.20237
3 0.00000 0.00000
4 5.75672 0.00411
5 25.98006 0.01853
6 830.42177 0.59231
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASS PATTERNS
BASED ON ESTIMATED POSTERIOR PROBABILITIES
Latent
Classes
1 256.11832 0.18268
2 283.72303 0.20237
3 0.00000 0.00000
4 5.75674 0.00411
5 25.98006 0.01853
6 830.42185 0.59231
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY LATENT CLASS MEMBERSHIP
Class Counts and Proportions
Latent
Classes
1 310 0.22111
2 543 0.38730
3 0 0.00000
4 0 0.00000
5 27 0.01926
6 522 0.37233
Average Latent Class Probabilities for Most Likely Latent Class Membership (Row)
by Latent Class (Column)
1 2 3 4 5 6
1 0.826 0.000 0.000 0.000 0.000 0.174
2 0.000 0.523 0.000 0.000 0.000 0.477
3 0.000 0.000 0.000 0.000 0.000 0.000
4 0.000 0.000 0.000 0.000 0.000 0.000
5 0.000 0.000 0.000 0.000 0.962 0.038
6 0.000 0.000 0.000 0.011 0.000 0.989
MODEL RESULTS
Estimates S.E. Est./S.E.
Latent Class 1
Thresholds
W$1 15.000 0.000 0.000
A$1 15.000 0.000 0.000
I$1 15.000 0.000 0.000
V$1 15.000 0.000 0.000
Latent Class 2
Thresholds
W$1 15.000 0.000 0.000
A$1 15.000 0.000 0.000
I$1 15.000 0.000 0.000
V$1 -15.000 0.000 0.000
Latent Class 3
Thresholds
W$1 15.000 0.000 0.000
A$1 15.000 0.000 0.000
I$1 -15.000 0.000 0.000
V$1 -15.000 0.000 0.000
Latent Class 4
Thresholds
W$1 15.000 0.000 0.000
A$1 -15.000 0.000 0.000
I$1 -15.000 0.000 0.000
V$1 -15.000 0.000 0.000
Latent Class 5
Thresholds
W$1 -15.000 0.000 0.000
A$1 -15.000 0.000 0.000
I$1 -15.000 0.000 0.000
V$1 -15.000 0.000 0.000
Latent Class 6
Thresholds
W$1 3.514 0.264 13.310
A$1 2.321 0.241 9.627
I$1 -0.296 0.251 -1.181
V$1 -1.571 0.123 -12.732
Categorical Latent Variables
Means
GRP#1 -1.176 0.171 -6.870
GRP#2 -1.074 0.335 -3.210
GRP#3 -19.453 0.289 -67.392
GRP#4 -4.972 1.782 -2.790
GRP#5 -3.465 0.218 -15.861
RESULTS IN PROBABILITY SCALE
Latent Class 1
W
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
A
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
I
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
V
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
Latent Class 2
W
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
A
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
I
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 3
W
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
A
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
I
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 4
W
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
A
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
I
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 5
W
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
A
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
I
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 6
W
Category 1 0.971 0.007 130.965
Category 2 0.029 0.007 3.901
A
Category 1 0.911 0.020 46.389
Category 2 0.089 0.020 4.556
I
Category 1 0.426 0.061 6.951
Category 2 0.574 0.061 9.348
V
Category 1 0.172 0.018 9.788
Category 2 0.828 0.018 47.098
ODDS RATIO RESULTS
Latent Class 1 Compared to Latent Class 2
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 3
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 5
W
Category > 1 0.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 6
W
Category > 1 0.000 0.000 3.788
A
Category > 1 0.000 0.000 4.148
I
Category > 1 0.000 0.000 3.987
V
Category > 1 0.000 0.000 8.104
Latent Class 2 Compared to Latent Class 3
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 2 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 2 Compared to Latent Class 5
W
Category > 1 0.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 2 Compared to Latent Class 6
W
Category > 1 0.000 0.000 3.788
A
Category > 1 0.000 0.000 4.148
I
Category > 1 0.000 0.000 3.987
V
Category > 1 ********* ******* 8.104
Latent Class 3 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 3 Compared to Latent Class 5
W
Category > 1 0.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 3 Compared to Latent Class 6
W
Category > 1 0.000 0.000 3.788
A
Category > 1 0.000 0.000 4.148
I
Category > 1 ********* ******* 3.987
V
Category > 1 ********* ******* 8.104
Latent Class 4 Compared to Latent Class 5
W
Category > 1 0.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 4 Compared to Latent Class 6
W
Category > 1 0.000 0.000 3.788
A
Category > 1 ********* ******* 4.148
I
Category > 1 ********* ******* 3.987
V
Category > 1 ********* ******* 8.104
Latent Class 5 Compared to Latent Class 6
W
Category > 1 ********* ******* 3.788
A
Category > 1 ********* ******* 4.148
I
Category > 1 ********* ******* 3.987
V
Category > 1 ********* ******* 8.104
QUALITY OF NUMERICAL RESULTS
Condition Number for the Information Matrix 0.238E-17
(ratio of smallest to largest eigenvalue)
Procotor-Goodman
title: page 58 - Table 4.6 - Proctor Goodman data: file is "d:\test\mplus47.txt"; variable: names are w a i v wt; weight is wt (frequency); classes = grp(6); categorical = w a i v; analysis: type = mixture; model: %overall% %grp#1% [w$1 a$1 i$1 v$1] (p5); %grp#2% [w$1 a$1 i$1 v$1] (p1); %grp#3% [w$1 a$1 i$1] (p1); [v$1] (p5); %grp#4% [w$1 a$1] (p1); [i$1 v$1] (p5); %grp#5% [w$1] (p1); [a$1 i$1 v$1] (p5); model constraint: p1 = -p5;
RANDOM STARTS RESULTS RANKED FROM THE BEST TO THE WORST LOGLIKELIHOOD VALUES
Initial stage loglikelihood values, seeds, and initial stage start numbers:
-2238.899 285380 1
-2240.730 93468 3
-2242.227 903420 5
-2243.101 253358 2
-2243.170 unperturbed 0
-2246.006 127215 9
-2254.780 939021 8
-2257.568 608496 4
-2273.301 195873 6
-2284.021 415931 10
-2285.583 462953 7
Loglikelihood values at local maxima, seeds, and initial stage start numbers:
-2225.858 285380 1
THE MODEL ESTIMATION TERMINATED NORMALLY
TESTS OF MODEL FIT
Loglikelihood
H0 Value -2225.858
Information Criteria
Number of Free Parameters 10
Akaike (AIC) 4471.716
Bayesian (BIC) 4524.173
Sample-Size Adjusted BIC 4492.406
(n* = (n + 2) / 24)
Entropy 0.707
Chi-Square Test of Model Fit for the Binary and Ordered Categorical
(Ordinal) Outcomes
Pearson Chi-Square
Value 16.251
Degrees of Freedom 5
P-Value 0.0062
Likelihood Ratio Chi-Square
Value 20.633
Degrees of Freedom 5
P-Value 0.0010
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASSES
BASED ON THE ESTIMATED MODEL
Latent
Classes
1 29.57421 0.02109
2 8.15422 0.00582
3 106.27651 0.07580
4 281.73036 0.20095
5 28.66118 0.02044
6 947.60353 0.67589
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASS PATTERNS
BASED ON ESTIMATED POSTERIOR PROBABILITIES
Latent
Classes
1 29.57421 0.02109
2 8.15422 0.00582
3 106.27650 0.07580
4 281.73046 0.20095
5 28.66115 0.02044
6 947.60346 0.67589
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY LATENT CLASS MEMBERSHIP
Class Counts and Proportions
Latent
Classes
1 29 0.02068
2 1 0.00071
3 4 0.00285
4 355 0.25321
5 40 0.02853
6 973 0.69401
Average Latent Class Probabilities for Most Likely Latent Class Membership (Row)
by Latent Class (Column)
1 2 3 4 5 6
1 0.913 0.000 0.011 0.027 0.049 0.000
2 0.006 0.544 0.392 0.058 0.000 0.000
3 0.013 0.004 0.857 0.126 0.001 0.000
4 0.004 0.000 0.014 0.678 0.004 0.300
5 0.033 0.000 0.006 0.310 0.570 0.080
6 0.000 0.008 0.100 0.028 0.003 0.861
MODEL RESULTS
Estimates S.E. Est./S.E.
Latent Class 1
Thresholds
W$1 -2.894 0.182 -15.863
A$1 -2.894 0.182 -15.863
I$1 -2.894 0.182 -15.863
V$1 -2.894 0.182 -15.863
Latent Class 2
Thresholds
W$1 2.894 0.182 15.863
A$1 2.894 0.182 15.863
I$1 2.894 0.182 15.863
V$1 2.894 0.182 15.863
Latent Class 3
Thresholds
W$1 2.894 0.182 15.863
A$1 2.894 0.182 15.863
I$1 2.894 0.182 15.863
V$1 -2.894 0.182 -15.863
Latent Class 4
Thresholds
W$1 2.894 0.182 15.863
A$1 2.894 0.182 15.863
I$1 -2.894 0.182 -15.863
V$1 -2.894 0.182 -15.863
Latent Class 5
Thresholds
W$1 2.894 0.182 15.863
A$1 -2.894 0.182 -15.863
I$1 -2.894 0.182 -15.863
V$1 -2.894 0.182 -15.863
Latent Class 6
Thresholds
W$1 26.329 0.000 0.000
A$1 3.493 0.330 10.572
I$1 1.445 0.156 9.277
V$1 -0.443 0.166 -2.674
Categorical Latent Variables
Means
GRP#1 -3.467 0.221 -15.685
GRP#2 -4.755 2.033 -2.339
GRP#3 -2.188 0.764 -2.864
GRP#4 -1.213 0.185 -6.573
GRP#5 -3.498 0.318 -10.988
RESULTS IN PROBABILITY SCALE
Latent Class 1
W
Category 1 0.052 0.009 5.785
Category 2 0.948 0.009 104.508
A
Category 1 0.052 0.009 5.785
Category 2 0.948 0.009 104.508
I
Category 1 0.052 0.009 5.785
Category 2 0.948 0.009 104.508
V
Category 1 0.052 0.009 5.785
Category 2 0.948 0.009 104.508
Latent Class 2
W
Category 1 0.948 0.009 104.508
Category 2 0.052 0.009 5.785
A
Category 1 0.948 0.009 104.508
Category 2 0.052 0.009 5.785
I
Category 1 0.948 0.009 104.508
Category 2 0.052 0.009 5.785
V
Category 1 0.948 0.009 104.508
Category 2 0.052 0.009 5.785
Latent Class 3
W
Category 1 0.948 0.009 104.508
Category 2 0.052 0.009 5.785
A
Category 1 0.948 0.009 104.508
Category 2 0.052 0.009 5.785
I
Category 1 0.948 0.009 104.508
Category 2 0.052 0.009 5.785
V
Category 1 0.052 0.009 5.785
Category 2 0.948 0.009 104.508
Latent Class 4
W
Category 1 0.948 0.009 104.508
Category 2 0.052 0.009 5.785
A
Category 1 0.948 0.009 104.508
Category 2 0.052 0.009 5.785
I
Category 1 0.052 0.009 5.785
Category 2 0.948 0.009 104.508
V
Category 1 0.052 0.009 5.785
Category 2 0.948 0.009 104.508
Latent Class 5
W
Category 1 0.948 0.009 104.508
Category 2 0.052 0.009 5.785
A
Category 1 0.052 0.009 5.785
Category 2 0.948 0.009 104.508
I
Category 1 0.052 0.009 5.785
Category 2 0.948 0.009 104.508
V
Category 1 0.052 0.009 5.785
Category 2 0.948 0.009 104.508
Latent Class 6
W
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
A
Category 1 0.970 0.009 102.538
Category 2 0.030 0.009 3.119
I
Category 1 0.809 0.024 33.653
Category 2 0.191 0.024 7.931
V
Category 1 0.391 0.039 9.920
Category 2 0.609 0.039 15.444
ODDS RATIO RESULTS
Latent Class 1 Compared to Latent Class 2
W
Category > 1 326.407 119.101 2.741
A
Category > 1 326.407 119.101 2.741
I
Category > 1 326.407 119.101 2.741
V
Category > 1 326.407 119.101 2.741
Latent Class 1 Compared to Latent Class 3
W
Category > 1 326.407 119.101 2.741
A
Category > 1 326.407 119.101 2.741
I
Category > 1 326.407 119.101 2.741
V
Category > 1 1.000 0.000 999.000
Latent Class 1 Compared to Latent Class 4
W
Category > 1 326.407 119.101 2.741
A
Category > 1 326.407 119.101 2.741
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 1 Compared to Latent Class 5
W
Category > 1 326.407 119.101 2.741
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 1 Compared to Latent Class 6
W
Category > 1 ********* 0.000 999.000
A
Category > 1 593.933 210.537 2.821
I
Category > 1 76.661 17.166 4.466
V
Category > 1 11.604 2.873 4.039
Latent Class 2 Compared to Latent Class 3
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 0.003 0.001 2.741
Latent Class 2 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.003 0.001 2.741
V
Category > 1 0.003 0.001 2.741
Latent Class 2 Compared to Latent Class 5
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.003 0.001 2.741
I
Category > 1 0.003 0.001 2.741
V
Category > 1 0.003 0.001 2.741
Latent Class 2 Compared to Latent Class 6
W
Category > 1 ********* 0.000 999.000
A
Category > 1 1.820 0.726 2.506
I
Category > 1 0.235 0.060 3.923
V
Category > 1 0.036 0.009 4.079
Latent Class 3 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.003 0.001 2.741
V
Category > 1 1.000 0.000 999.000
Latent Class 3 Compared to Latent Class 5
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.003 0.001 2.741
I
Category > 1 0.003 0.001 2.741
V
Category > 1 1.000 0.000 999.000
Latent Class 3 Compared to Latent Class 6
W
Category > 1 ********* 0.000 999.000
A
Category > 1 1.820 0.726 2.506
I
Category > 1 0.235 0.060 3.923
V
Category > 1 11.604 2.873 4.039
Latent Class 4 Compared to Latent Class 5
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.003 0.001 2.741
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 4 Compared to Latent Class 6
W
Category > 1 ********* 0.000 999.000
A
Category > 1 1.820 0.726 2.506
I
Category > 1 76.661 17.166 4.466
V
Category > 1 11.604 2.873 4.039
Latent Class 5 Compared to Latent Class 6
W
Category > 1 ********* 0.000 999.000
A
Category > 1 593.933 210.537 2.821
I
Category > 1 76.661 17.166 4.466
V
Category > 1 11.604 2.873 4.039
QUALITY OF NUMERICAL RESULTS
Condition Number for the Information Matrix 0.128E-02
(ratio of smallest to largest eigenvalue)
Biform scale
title: page 58 - Table 4.6 - Biform Scale data: file is mplus47.txt; variable: names are w a i v wt; weight is wt (frequency); classes = grp(7); categorical = w a i v; analysis: type = mixture; starts = 50 2; model: %overall% [w$1 a$1 i$1 v$1]; %grp#1% [w$1@15 a$1@15 i$1@15 v$1@15]; %grp#2% [w$1@15 a$1@15 i$1@15]; [v$1@-15]; %grp#3% [w$1@15 a$1@15]; [i$1@-15 v$1@-15]; %grp#4% [w$1@15]; [a$1@-15 i$1@-15 v$1@-15]; %grp#5% [w$1@-15 a$1@-15 i$1@-15 v$1@-15]; !%grp#6% ![a$1@15]; ![w$1@-15 i$1@-15 v$1@-15]; %grp#6% [w$1@15 a$1@15 i$1@-15 v$1@15]; ! this model has one additional possible ! response pattern; hence, the one additional ! group. output: tech1;
RANDOM STARTS RESULTS RANKED FROM THE BEST TO THE WORST LOGLIKELIHOOD VALUES
Initial stage loglikelihood values, seeds, and initial stage start numbers:
-2220.174 195873 6
-2220.596 848163 47
-2222.426 260601 36
-2222.833 364676 27
-2223.137 93468 3
-2223.394 967237 48
-2224.032 851945 18
-2224.116 761633 50
-2224.212 107446 12
-2224.503 902278 21
-2224.696 372176 23
-2224.743 318230 46
-2224.810 153942 31
-2224.865 626891 32
-2224.865 68985 17
-2225.092 392418 28
-2225.112 407168 44
-2226.045 120506 45
-2226.062 253358 2
-2226.727 352277 42
-2226.779 533739 11
-2227.682 903420 5
-2229.145 unperturbed 0
-2229.795 650371 14
-2230.227 27071 15
-2230.651 285380 1
-2231.614 608496 4
-2232.092 76974 16
-2232.132 887676 22
-2232.176 370466 41
-2232.226 645664 39
-2233.017 915642 40
-2233.109 939021 8
-2233.691 207896 25
-2235.228 963053 43
-2236.956 569131 26
-2237.014 246261 38
-2237.520 966014 37
-2237.959 568859 49
-2240.819 341041 34
-2242.111 366706 29
-2247.901 573096 20
-2252.745 347515 24
-2259.445 830392 35
-2260.975 637345 19
-2261.545 749453 33
-2261.924 432148 30
-2264.532 415931 10
-2264.802 462953 7
-2265.040 399671 13
-2267.751 127215 9
Loglikelihood values at local maxima, seeds, and initial stage start numbers:
-2218.922 848163 47
-2218.922 195873 6
ONE OR MORE MULTINOMIAL LOGIT PARAMETERS WERE FIXED TO AVOID SINGULARITY
OF THE INFORMATION MATRIX. THE SINGULARITY IS MOST LIKELY BECAUSE THE
MODEL IS NOT IDENTIFIED, OR BECAUSE OF EMPTY CELLS IN THE JOINT
DISTRIBUTION OF THE CATEGORICAL LATENT VARIABLES AND ANY INDEPENDENT
VARIABLES. THE FOLLOWING PARAMETERS WERE FIXED:
8
THE MODEL ESTIMATION TERMINATED NORMALLY
TESTS OF MODEL FIT
Loglikelihood
H0 Value -2218.922
Information Criteria
Number of Free Parameters 10
Akaike (AIC) 4457.844
Bayesian (BIC) 4510.300
Sample-Size Adjusted BIC 4478.534
(n* = (n + 2) / 24)
Entropy 0.775
Chi-Square Test of Model Fit for the Binary and Ordered Categorical
(Ordinal) Outcomes
Pearson Chi-Square
Value 5.596
Degrees of Freedom 5
P-Value 0.3475
Likelihood Ratio Chi-Square
Value 6.761
Degrees of Freedom 5
P-Value 0.2390
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASSES
BASED ON THE ESTIMATED MODEL
Latent
Classes
1 301.78714 0.21525
2 410.76166 0.29298
3 140.31783 0.10008
4 0.00000 0.00000
5 24.39940 0.01740
6 70.65999 0.05040
7 454.07397 0.32388
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASS PATTERNS
BASED ON ESTIMATED POSTERIOR PROBABILITIES
Latent
Classes
1 301.78716 0.21525
2 410.76182 0.29298
3 140.31775 0.10008
4 0.00000 0.00000
5 24.39941 0.01740
6 70.66001 0.05040
7 454.07385 0.32388
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY LATENT CLASS MEMBERSHIP
Class Counts and Proportions
Latent
Classes
1 310 0.22111
2 543 0.38730
3 0 0.00000
4 0 0.00000
5 27 0.01926
6 83 0.05920
7 439 0.31312
Average Latent Class Probabilities for Most Likely Latent Class Membership (Row)
by Latent Class (Column)
1 2 3 4 5 6 7
1 0.974 0.000 0.000 0.000 0.000 0.000 0.026
2 0.000 0.756 0.000 0.000 0.000 0.000 0.244
3 0.000 0.000 0.000 0.000 0.000 0.000 0.000
4 0.000 0.000 0.000 0.000 0.000 0.000 0.000
5 0.000 0.000 0.000 0.000 0.904 0.000 0.096
6 0.000 0.000 0.000 0.000 0.000 0.851 0.149
7 0.000 0.000 0.320 0.000 0.000 0.000 0.680
MODEL RESULTS
Estimates S.E. Est./S.E.
Latent Class 1
Thresholds
W$1 15.000 0.000 0.000
A$1 15.000 0.000 0.000
I$1 15.000 0.000 0.000
V$1 15.000 0.000 0.000
Latent Class 2
Thresholds
W$1 15.000 0.000 0.000
A$1 15.000 0.000 0.000
I$1 15.000 0.000 0.000
V$1 -15.000 0.000 0.000
Latent Class 3
Thresholds
W$1 15.000 0.000 0.000
A$1 15.000 0.000 0.000
I$1 -15.000 0.000 0.000
V$1 -15.000 0.000 0.000
Latent Class 4
Thresholds
W$1 15.000 0.000 0.000
A$1 -15.000 0.000 0.000
I$1 -15.000 0.000 0.000
V$1 -15.000 0.000 0.000
Latent Class 5
Thresholds
W$1 -15.000 0.000 0.000
A$1 -15.000 0.000 0.000
I$1 -15.000 0.000 0.000
V$1 -15.000 0.000 0.000
Latent Class 6
Thresholds
W$1 15.000 0.000 0.000
A$1 15.000 0.000 0.000
I$1 -15.000 0.000 0.000
V$1 15.000 0.000 0.000
Latent Class 7
Thresholds
W$1 2.818 0.727 3.877
A$1 1.518 0.744 2.041
I$1 -0.407 0.205 -1.986
V$1 -2.779 0.423 -6.563
Categorical Latent Variables
Means
GRP#1 -0.409 0.608 -0.672
GRP#2 -0.100 0.831 -0.121
GRP#3 -1.174 1.645 -0.714
GRP#4 -21.046 0.000 0.000
GRP#5 -2.924 0.554 -5.277
GRP#6 -1.860 0.737 -2.526
RESULTS IN PROBABILITY SCALE
Latent Class 1
W
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
A
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
I
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
V
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
Latent Class 2
W
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
A
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
I
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 3
W
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
A
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
I
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 4
W
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
A
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
I
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 5
W
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
A
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
I
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 6
W
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
A
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
I
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
V
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
Latent Class 7
W
Category 1 0.944 0.039 24.408
Category 2 0.056 0.039 1.458
A
Category 1 0.820 0.110 7.481
Category 2 0.180 0.110 1.639
I
Category 1 0.400 0.049 8.124
Category 2 0.600 0.049 12.205
V
Category 1 0.058 0.023 2.509
Category 2 0.942 0.023 40.389
ODDS RATIO RESULTS
Latent Class 1 Compared to Latent Class 2
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 3
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 5
W
Category > 1 0.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 6
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 1 Compared to Latent Class 7
W
Category > 1 0.000 0.000 1.376
A
Category > 1 0.000 0.000 1.344
I
Category > 1 0.000 0.000 4.877
V
Category > 1 0.000 0.000 2.362
Latent Class 2 Compared to Latent Class 3
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 2 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 2 Compared to Latent Class 5
W
Category > 1 0.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 2 Compared to Latent Class 6
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 ********* 0.000 999.000
Latent Class 2 Compared to Latent Class 7
W
Category > 1 0.000 0.000 1.376
A
Category > 1 0.000 0.000 1.344
I
Category > 1 0.000 0.000 4.877
V
Category > 1 ********* ******* 2.362
Latent Class 3 Compared to Latent Class 4
W
Category > 1 1.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 3 Compared to Latent Class 5
W
Category > 1 0.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 3 Compared to Latent Class 6
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 ********* 0.000 999.000
Latent Class 3 Compared to Latent Class 7
W
Category > 1 0.000 0.000 1.376
A
Category > 1 0.000 0.000 1.344
I
Category > 1 ********* ******* 4.877
V
Category > 1 ********* ******* 2.362
Latent Class 4 Compared to Latent Class 5
W
Category > 1 0.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 4 Compared to Latent Class 6
W
Category > 1 1.000 0.000 999.000
A
Category > 1 ********* 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 ********* 0.000 999.000
Latent Class 4 Compared to Latent Class 7
W
Category > 1 0.000 0.000 1.376
A
Category > 1 ********* ******* 1.344
I
Category > 1 ********* ******* 4.877
V
Category > 1 ********* ******* 2.362
Latent Class 5 Compared to Latent Class 6
W
Category > 1 ********* 0.000 999.000
A
Category > 1 ********* 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 ********* 0.000 999.000
Latent Class 5 Compared to Latent Class 7
W
Category > 1 ********* ******* 1.376
A
Category > 1 ********* ******* 1.344
I
Category > 1 ********* ******* 4.877
V
Category > 1 ********* ******* 2.362
Latent Class 6 Compared to Latent Class 7
W
Category > 1 0.000 0.000 1.376
A
Category > 1 0.000 0.000 1.344
I
Category > 1 ********* ******* 4.877
V
Category > 1 0.000 0.000 2.362
QUALITY OF NUMERICAL RESULTS
Condition Number for the Information Matrix 0.586E-03
(ratio of smallest to largest eigenvalue)
TECHNICAL 1 OUTPUT
PARAMETER SPECIFICATION FOR LATENT CLASS 1
PARAMETER SPECIFICATION FOR LATENT CLASS 2
PARAMETER SPECIFICATION FOR LATENT CLASS 3
PARAMETER SPECIFICATION FOR LATENT CLASS 4
PARAMETER SPECIFICATION FOR LATENT CLASS 5
PARAMETER SPECIFICATION FOR LATENT CLASS 6
PARAMETER SPECIFICATION FOR LATENT CLASS 7
PARAMETER SPECIFICATION FOR LATENT CLASS INDICATOR MODEL PART
LAMBDA(U)
GRP#1 GRP#2 GRP#3 GRP#4 GRP#5
________ ________ ________ ________ ________
W 0 0 0 0 0
A 0 0 0 0 0
I 0 0 0 0 0
V 0 0 0 0 0
LAMBDA(U)
GRP#6 GRP#7
________ ________
W 0 1
A 0 2
I 0 3
V 0 4
PARAMETER SPECIFICATION FOR LATENT CLASS REGRESSION MODEL PART
ALPHA(C)
GRP#1 GRP#2 GRP#3 GRP#4 GRP#5
________ ________ ________ ________ ________
1 5 6 7 8 9
ALPHA(C)
GRP#6 GRP#7
________ ________
1 10 0
STARTING VALUES FOR LATENT CLASS 1
STARTING VALUES FOR LATENT CLASS 2
STARTING VALUES FOR LATENT CLASS 3
STARTING VALUES FOR LATENT CLASS 4
STARTING VALUES FOR LATENT CLASS 5
STARTING VALUES FOR LATENT CLASS 6
STARTING VALUES FOR LATENT CLASS 7
STARTING VALUES FOR LATENT CLASS INDICATOR MODEL PART
LAMBDA(U)
GRP#1 GRP#2 GRP#3 GRP#4 GRP#5
________ ________ ________ ________ ________
W -15.000 -15.000 -15.000 -15.000 15.000
A -15.000 -15.000 -15.000 15.000 15.000
I -15.000 -15.000 15.000 15.000 15.000
V -15.000 15.000 15.000 15.000 15.000
LAMBDA(U)
GRP#6 GRP#7
________ ________
W -15.000 -5.297
A -15.000 -4.504
I 15.000 -2.565
V -15.000 -1.078
STARTING VALUES FOR LATENT CLASS REGRESSION MODEL PART
ALPHA(C)
GRP#1 GRP#2 GRP#3 GRP#4 GRP#5
________ ________ ________ ________ ________
1 0.000 0.000 0.000 0.000 0.000
ALPHA(C)
GRP#6 GRP#7
________ ________
1 0.000 0.000
Biform scale with type 2 excluded
title: page 58 - Table 4.6 - Biform Scale with type 2 excluded data: file is mplus47.txt; variable: names are w a i v wt; weight is wt (frequency); classes = grp(6); categorical = w a i v; analysis: type = mixture; starts = 50 2; model: %overall% [w$1 a$1 i$1 v$1]; %grp#1% [w$1@15 a$1@15 i$1@15 v$1@15]; %grp#2% [w$1@15 a$1@15 i$1@15]; [v$1@-15]; %grp#3% [w$1@15 a$1@15]; [i$1@-15 v$1@-15]; !%grp#4% ![w$1@15]; ![a$1@-15 i$1@-15 v$1@-15]; %grp#4% [w$1@-15 a$1@-15 i$1@-15 v$1@-15]; !%grp#6% ![a$1@15]; ![w$1@-15 i$1@-15 v$1@-15]; %grp#5% [w$1@15 a$1@15 i$1@-15 v$1@15]; ! this model has one additional possible ! response pattern; hence, the one additional ! group. output: tech1;
RANDOM STARTS RESULTS RANKED FROM THE BEST TO THE WORST LOGLIKELIHOOD VALUES
Initial stage loglikelihood values, seeds, and initial stage start numbers:
-2219.937 93468 3
-2219.963 107446 12
-2220.054 761633 50
-2220.185 903420 5
-2220.190 195873 6
-2220.203 851945 18
-2220.340 370466 41
-2220.496 260601 36
-2220.520 253358 2
-2220.591 352277 42
-2220.596 120506 45
-2220.738 unperturbed 0
-2220.839 848163 47
-2221.117 364676 27
-2221.465 27071 15
-2221.568 76974 16
-2222.647 902278 21
-2222.723 153942 31
-2223.341 626891 32
-2223.449 318230 46
-2223.533 372176 23
-2223.577 967237 48
-2223.843 68985 17
-2223.923 392418 28
-2224.094 407168 44
-2224.666 533739 11
-2224.675 966014 37
-2226.286 650371 14
-2226.453 341041 34
-2226.575 887676 22
-2226.900 608496 4
-2228.345 645664 39
-2229.351 285380 1
-2231.249 915642 40
-2232.057 939021 8
-2232.885 207896 25
-2235.420 366706 29
-2236.068 569131 26
-2236.725 246261 38
-2237.559 963053 43
-2237.902 568859 49
-2239.182 347515 24
-2245.539 573096 20
-2253.338 830392 35
-2257.663 749453 33
-2265.918 432148 30
-2270.350 637345 19
-2270.585 399671 13
-2272.419 415931 10
-2280.789 462953 7
-2283.710 127215 9
Loglikelihood values at local maxima, seeds, and initial stage start numbers:
-2218.922 93468 3
-2218.922 107446 12
THE MODEL ESTIMATION TERMINATED NORMALLY
TESTS OF MODEL FIT
Loglikelihood
H0 Value -2218.922
Information Criteria
Number of Free Parameters 9
Akaike (AIC) 4455.844
Bayesian (BIC) 4503.055
Sample-Size Adjusted BIC 4474.465
(n* = (n + 2) / 24)
Entropy 0.756
Chi-Square Test of Model Fit for the Binary and Ordered Categorical
(Ordinal) Outcomes
Pearson Chi-Square
Value 5.596
Degrees of Freedom 6
P-Value 0.4699
Likelihood Ratio Chi-Square
Value 6.761
Degrees of Freedom 6
P-Value 0.3435
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASSES
BASED ON THE ESTIMATED MODEL
Latent
Classes
1 301.78742 0.21525
2 410.76745 0.29299
3 140.32280 0.10009
4 24.39932 0.01740
5 70.66020 0.05040
6 454.06282 0.32387
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASS PATTERNS
BASED ON ESTIMATED POSTERIOR PROBABILITIES
Latent
Classes
1 301.78745 0.21525
2 410.76734 0.29299
3 140.32283 0.10009
4 24.39932 0.01740
5 70.66024 0.05040
6 454.06282 0.32387
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY LATENT CLASS MEMBERSHIP
Class Counts and Proportions
Latent
Classes
1 310 0.22111
2 543 0.38730
3 0 0.00000
4 27 0.01926
5 83 0.05920
6 439 0.31312
Average Latent Class Probabilities for Most Likely Latent Class Membership (Row)
by Latent Class (Column)
1 2 3 4 5 6
1 0.974 0.000 0.000 0.000 0.000 0.026
2 0.000 0.756 0.000 0.000 0.000 0.244
3 0.000 0.000 0.000 0.000 0.000 0.000
4 0.000 0.000 0.000 0.904 0.000 0.096
5 0.000 0.000 0.000 0.000 0.851 0.149
6 0.000 0.000 0.320 0.000 0.000 0.680
MODEL RESULTS
Estimates S.E. Est./S.E.
Latent Class 1
Thresholds
W$1 15.000 0.000 0.000
A$1 15.000 0.000 0.000
I$1 15.000 0.000 0.000
V$1 15.000 0.000 0.000
Latent Class 2
Thresholds
W$1 15.000 0.000 0.000
A$1 15.000 0.000 0.000
I$1 15.000 0.000 0.000
V$1 -15.000 0.000 0.000
Latent Class 3
Thresholds
W$1 15.000 0.000 0.000
A$1 15.000 0.000 0.000
I$1 -15.000 0.000 0.000
V$1 -15.000 0.000 0.000
Latent Class 4
Thresholds
W$1 -15.000 0.000 0.000
A$1 -15.000 0.000 0.000
I$1 -15.000 0.000 0.000
V$1 -15.000 0.000 0.000
Latent Class 5
Thresholds
W$1 15.000 0.000 0.000
A$1 15.000 0.000 0.000
I$1 -15.000 0.000 0.000
V$1 15.000 0.000 0.000
Latent Class 6
Thresholds
W$1 2.818 0.727 3.877
A$1 1.518 0.744 2.041
I$1 -0.407 0.205 -1.986
V$1 -2.779 0.423 -6.563
Categorical Latent Variables
Means
GRP#1 -0.409 0.608 -0.672
GRP#2 -0.100 0.831 -0.121
GRP#3 -1.174 1.645 -0.714
GRP#4 -2.924 0.554 -5.277
GRP#5 -1.860 0.737 -2.525
RESULTS IN PROBABILITY SCALE
Latent Class 1
W
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
A
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
I
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
V
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
Latent Class 2
W
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
A
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
I
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 3
W
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
A
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
I
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 4
W
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
A
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
I
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
V
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
Latent Class 5
W
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
A
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
I
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
V
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
Latent Class 6
W
Category 1 0.944 0.039 24.405
Category 2 0.056 0.039 1.458
A
Category 1 0.820 0.110 7.480
Category 2 0.180 0.110 1.639
I
Category 1 0.400 0.049 8.124
Category 2 0.600 0.049 12.206
V
Category 1 0.058 0.023 2.509
Category 2 0.942 0.023 40.389
ODDS RATIO RESULTS
Latent Class 1 Compared to Latent Class 2
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 3
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 4
W
Category > 1 0.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 0.000 0.000 999.000
Latent Class 1 Compared to Latent Class 5
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 1 Compared to Latent Class 6
W
Category > 1 0.000 0.000 1.376
A
Category > 1 0.000 0.000 1.344
I
Category > 1 0.000 0.000 4.877
V
Category > 1 0.000 0.000 2.362
Latent Class 2 Compared to Latent Class 3
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 2 Compared to Latent Class 4
W
Category > 1 0.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 2 Compared to Latent Class 5
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 0.000 0.000 999.000
V
Category > 1 ********* 0.000 999.000
Latent Class 2 Compared to Latent Class 6
W
Category > 1 0.000 0.000 1.376
A
Category > 1 0.000 0.000 1.344
I
Category > 1 0.000 0.000 4.877
V
Category > 1 ********* ******* 2.362
Latent Class 3 Compared to Latent Class 4
W
Category > 1 0.000 0.000 999.000
A
Category > 1 0.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 1.000 0.000 999.000
Latent Class 3 Compared to Latent Class 5
W
Category > 1 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 ********* 0.000 999.000
Latent Class 3 Compared to Latent Class 6
W
Category > 1 0.000 0.000 1.376
A
Category > 1 0.000 0.000 1.344
I
Category > 1 ********* ******* 4.877
V
Category > 1 ********* ******* 2.362
Latent Class 4 Compared to Latent Class 5
W
Category > 1 ********* 0.000 999.000
A
Category > 1 ********* 0.000 999.000
I
Category > 1 1.000 0.000 999.000
V
Category > 1 ********* 0.000 999.000
Latent Class 4 Compared to Latent Class 6
W
Category > 1 ********* ******* 1.376
A
Category > 1 ********* ******* 1.344
I
Category > 1 ********* ******* 4.877
V
Category > 1 ********* ******* 2.362
Latent Class 5 Compared to Latent Class 6
W
Category > 1 0.000 0.000 1.376
A
Category > 1 0.000 0.000 1.344
I
Category > 1 ********* ******* 4.877
V
Category > 1 0.000 0.000 2.362
QUALITY OF NUMERICAL RESULTS
Condition Number for the Information Matrix 0.577E-03
(ratio of smallest to largest eigenvalue)
TECHNICAL 1 OUTPUT
PARAMETER SPECIFICATION FOR LATENT CLASS 1
PARAMETER SPECIFICATION FOR LATENT CLASS 2
PARAMETER SPECIFICATION FOR LATENT CLASS 3
PARAMETER SPECIFICATION FOR LATENT CLASS 4
PARAMETER SPECIFICATION FOR LATENT CLASS 5
PARAMETER SPECIFICATION FOR LATENT CLASS 6
PARAMETER SPECIFICATION FOR LATENT CLASS INDICATOR MODEL PART
LAMBDA(U)
GRP#1 GRP#2 GRP#3 GRP#4 GRP#5
________ ________ ________ ________ ________
W 0 0 0 0 0
A 0 0 0 0 0
I 0 0 0 0 0
V 0 0 0 0 0
LAMBDA(U)
GRP#6
________
W 1
A 2
I 3
V 4
PARAMETER SPECIFICATION FOR LATENT CLASS REGRESSION MODEL PART
ALPHA(C)
GRP#1 GRP#2 GRP#3 GRP#4 GRP#5
________ ________ ________ ________ ________
1 5 6 7 8 9
ALPHA(C)
GRP#6
________
1 0
STARTING VALUES FOR LATENT CLASS 1
STARTING VALUES FOR LATENT CLASS 2
STARTING VALUES FOR LATENT CLASS 3
STARTING VALUES FOR LATENT CLASS 4
STARTING VALUES FOR LATENT CLASS 5
STARTING VALUES FOR LATENT CLASS 6
STARTING VALUES FOR LATENT CLASS INDICATOR MODEL PART
LAMBDA(U)
GRP#1 GRP#2 GRP#3 GRP#4 GRP#5
________ ________ ________ ________ ________
W -15.000 -15.000 -15.000 15.000 -15.000
A -15.000 -15.000 -15.000 15.000 -15.000
I -15.000 -15.000 15.000 15.000 15.000
V -15.000 15.000 15.000 15.000 -15.000
LAMBDA(U)
GRP#6
________
W -5.297
A -4.504
I -2.565
V -1.078
STARTING VALUES FOR LATENT CLASS REGRESSION MODEL PART
ALPHA(C)
GRP#1 GRP#2 GRP#3 GRP#4 GRP#5
________ ________ ________ ________ ________
1 0.000 0.000 0.000 0.000 0.000
ALPHA(C)
GRP#6
________
1 0.000
Page 70 – this may not be the correct code. data file
title: page 70 - unrestricted, heterogeneous three-class model
data: file is "d:\test\mplus69.txt";
variable: names are race p a u c wt;
!usevariables are race p a u c wt;
weight is wt (frequency);
classes = r(2) grp(3);
categorical = race p a u c;
analysis: type = mixture;
model:
%overall%
grp#1 on r#1;
grp#2 on r#1;
[grp#1 grp#2 r#1];
model r:
%r#1%
[race$1@15];
%r#2%
[race$1@-15];
model grp:
%grp#1%
[p$1 p$2 a$1 u$1 c$1 c$2];
%grp#2%
[p$1 p$2 a$1 u$1 c$1 c$2];
%grp#3%
[p$1 p$2 a$1 u$1 c$1 c$2];
! You can't use grouping with type = mixture ! See example on page 378 of manual and pages 328-340 ! (white =1 black =2) ! You need the usevariables so that Mplus knows not to use ! race in the analysis. ! See also page 151 of the manual, example 7.21
RANDOM STARTS RESULTS RANKED FROM THE BEST TO THE WORST LOGLIKELIHOOD VALUES
Initial stage loglikelihood values, seeds, and initial stage start numbers:
-4867.705 127215 9
-4899.600 939021 8
-4901.769 93468 3
-4906.455 285380 1
-4909.705 195873 6
-4927.386 unperturbed 0
-4929.380 903420 5
-4933.149 253358 2
-4936.477 415931 10
-4940.304 462953 7
-4980.604 608496 4
Loglikelihood values at local maxima, seeds, and initial stage start numbers:
-4858.404 127215 9
THE MODEL ESTIMATION TERMINATED NORMALLY
TESTS OF MODEL FIT
Loglikelihood
H0 Value -4858.404
Information Criteria
Number of Free Parameters 23
Akaike (AIC) 9762.807
Bayesian (BIC) 9887.190
Sample-Size Adjusted BIC 9814.122
(n* = (n + 2) / 24)
Entropy 0.781
Chi-Square Test of Model Fit for the Binary and Ordered Categorical
(Ordinal) Outcomes
Pearson Chi-Square
Value 47.978
Degrees of Freedom 48
P-Value 0.4737
Likelihood Ratio Chi-Square
Value 49.077
Degrees of Freedom 48
P-Value 0.4297
MODEL RESULTS USE THE LATENT CLASS VARIABLE ORDER
R GRP
Latent Class Variable Patterns
R GRP
Class Class
1 1
1 2
1 3
2 1
2 2
2 3
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASS PATTERNS
BASED ON THE ESTIMATED MODEL
Latent Class
Pattern
1 1 192.95286 0.11701
1 2 169.88252 0.10302
1 3 839.16486 0.50889
2 1 81.65096 0.04952
2 2 128.37271 0.07785
2 3 236.97610 0.14371
FINAL CLASS COUNTS AND PROPORTIONS FOR EACH LATENT CLASS VARIABLE
BASED ON THE ESTIMATED MODEL
Latent Class
Variable Class
R 1 1202.00024 0.72893
2 446.99976 0.27107
GRP 1 274.60382 0.16653
2 298.25522 0.18087
3 1076.14099 0.65260
LATENT TRANSITION PROBABILITIES BASED ON THE ESTIMATED MODEL
R Classes (Rows) by GRP Classes (Columns)
1 2 3
1 0.161 0.141 0.698
2 0.183 0.287 0.530
FINAL CLASS COUNTS AND PROPORTIONS FOR THE LATENT CLASSES
BASED ON ESTIMATED POSTERIOR PROBABILITIES
Latent Class
Pattern
1 1 192.95297 0.11701
1 2 169.88333 0.10302
1 3 839.16393 0.50889
2 1 81.65102 0.04952
2 2 128.37313 0.07785
2 3 236.97562 0.14371
FINAL CLASS COUNTS AND PROPORTIONS FOR EACH LATENT CLASS VARIABLE
BASED ON ESTIMATED POSTERIOR PROBABILITIES
Latent Class
Variable Class
R 1 1202.00024 0.72893
2 446.99979 0.27107
GRP 1 274.60397 0.16653
2 298.25647 0.18087
3 1076.13953 0.65260
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY LATENT CLASS PATTERN
Class Counts and Proportions
Latent Class
Pattern
1 1 219 0.13281
1 2 178 0.10794
1 3 805 0.48817
2 1 86 0.05215
2 2 136 0.08247
2 3 225 0.13645
Average Latent Class Probabilities for Most Likely Latent Class Pattern (Row)
by Latent Class Pattern (Column)
Latent Class Variable Patterns
Latent Class R GRP
Pattern No. Class Class
1 1 1
2 1 2
3 1 3
4 2 1
5 2 2
6 2 3
1 2 3 4 5 6
1 0.773 0.035 0.191 0.000 0.000 0.000
2 0.038 0.657 0.305 0.000 0.000 0.000
3 0.021 0.056 0.923 0.000 0.000 0.000
4 0.000 0.000 0.000 0.808 0.051 0.140
5 0.000 0.000 0.000 0.045 0.763 0.193
6 0.000 0.000 0.000 0.027 0.090 0.883
CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY LATENT CLASS MEMBERSHIP
FOR EACH LATENT CLASS VARIABLE
Latent Class
Variable Class
R 1 1202 0.72893
2 447 0.27107
GRP 1 305 0.18496
2 314 0.19042
3 1030 0.62462
MODEL RESULTS
Estimates S.E. Est./S.E.
Latent Class Pattern 1 1
Thresholds
RACE$1 15.000 0.000 0.000
P$1 -1.833 0.964 -1.901
P$2 -0.600 0.425 -1.411
A$1 -15.783 0.000 0.000
U$1 1.056 0.199 5.299
C$1 0.613 0.176 3.474
C$2 2.134 0.232 9.213
Latent Class Pattern 1 2
Thresholds
RACE$1 15.000 0.000 0.000
P$1 2.147 0.318 6.747
P$2 3.134 0.694 4.519
A$1 0.566 0.262 2.165
U$1 -0.853 1.052 -0.810
C$1 0.208 0.781 0.266
C$2 2.372 0.567 4.183
Latent Class Pattern 1 3
Thresholds
RACE$1 15.000 0.000 0.000
P$1 2.047 0.156 13.087
P$2 2.842 0.261 10.869
A$1 0.417 0.098 4.277
U$1 2.650 1.382 1.917
C$1 2.780 0.724 3.841
C$2 6.575 7.577 0.868
Latent Class Pattern 2 1
Thresholds
RACE$1 -15.000 0.000 0.000
P$1 -1.833 0.964 -1.901
P$2 -0.600 0.425 -1.411
A$1 -15.783 0.000 0.000
U$1 1.056 0.199 5.299
C$1 0.613 0.176 3.474
C$2 2.134 0.232 9.213
Latent Class Pattern 2 2
Thresholds
RACE$1 -15.000 0.000 0.000
P$1 2.147 0.318 6.747
P$2 3.134 0.694 4.519
A$1 0.566 0.262 2.165
U$1 -0.853 1.052 -0.810
C$1 0.208 0.781 0.266
C$2 2.372 0.567 4.183
Latent Class Pattern 2 3
Thresholds
RACE$1 -15.000 0.000 0.000
P$1 2.047 0.156 13.087
P$2 2.842 0.261 10.869
A$1 0.417 0.098 4.277
U$1 2.650 1.382 1.917
C$1 2.780 0.724 3.841
C$2 6.575 7.577 0.868
Categorical Latent Variables
GRP#1 ON
R#1 -0.404 0.235 -1.719
GRP#2 ON
R#1 -0.984 0.397 -2.478
Means
R#1 0.989 0.055 17.856
GRP#1 -1.066 0.457 -2.329
GRP#2 -0.613 0.807 -0.760
RESULTS IN PROBABILITY SCALE
Latent Class Pattern 1 1
RACE
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
P
Category 1 0.138 0.115 1.204
Category 2 0.216 0.040 5.346
Category 3 0.646 0.097 6.639
A
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
U
Category 1 0.742 0.038 19.444
Category 2 0.258 0.038 6.764
C
Category 1 0.649 0.040 16.133
Category 2 0.246 0.033 7.384
Category 3 0.106 0.022 4.828
Latent Class Pattern 1 2
RACE
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
P
Category 1 0.895 0.030 30.033
Category 2 0.063 0.028 2.233
Category 3 0.042 0.028 1.505
A
Category 1 0.638 0.060 10.559
Category 2 0.362 0.060 5.994
U
Category 1 0.299 0.220 1.356
Category 2 0.701 0.220 3.180
C
Category 1 0.552 0.193 2.856
Category 2 0.363 0.155 2.337
Category 3 0.085 0.044 1.928
Latent Class Pattern 1 3
RACE
Category 1 1.000 0.000 0.000
Category 2 0.000 0.000 0.000
P
Category 1 0.886 0.016 55.908
Category 2 0.059 0.012 4.797
Category 3 0.055 0.014 4.047
A
Category 1 0.603 0.023 25.807
Category 2 0.397 0.023 17.002
U
Category 1 0.934 0.085 10.962
Category 2 0.066 0.085 0.775
C
Category 1 0.942 0.040 23.657
Category 2 0.057 0.030 1.881
Category 3 0.001 0.011 0.132
Latent Class Pattern 2 1
RACE
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
P
Category 1 0.138 0.115 1.204
Category 2 0.216 0.040 5.346
Category 3 0.646 0.097 6.639
A
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
U
Category 1 0.742 0.038 19.444
Category 2 0.258 0.038 6.764
C
Category 1 0.649 0.040 16.133
Category 2 0.246 0.033 7.384
Category 3 0.106 0.022 4.828
Latent Class Pattern 2 2
RACE
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
P
Category 1 0.895 0.030 30.033
Category 2 0.063 0.028 2.233
Category 3 0.042 0.028 1.505
A
Category 1 0.638 0.060 10.559
Category 2 0.362 0.060 5.994
U
Category 1 0.299 0.220 1.356
Category 2 0.701 0.220 3.180
C
Category 1 0.552 0.193 2.856
Category 2 0.363 0.155 2.337
Category 3 0.085 0.044 1.928
Latent Class Pattern 2 3
RACE
Category 1 0.000 0.000 0.000
Category 2 1.000 0.000 0.000
P
Category 1 0.886 0.016 55.908
Category 2 0.059 0.012 4.797
Category 3 0.055 0.014 4.047
A
Category 1 0.603 0.023 25.807
Category 2 0.397 0.023 17.002
U
Category 1 0.934 0.085 10.962
Category 2 0.066 0.085 0.775
C
Category 1 0.942 0.040 23.657
Category 2 0.057 0.030 1.881
Category 3 0.001 0.011 0.132
ODDS RATIO RESULTS
Latent Class Pattern 1 1 Compared to Latent Class Pattern 1 2
RACE
Category > 1 1.000 0.000 999.000
P
Category > 1 53.465 55.195 0.969
Category > 2 41.833 29.868 1.401
A
Category > 1 ********* 0.000 999.000
U
Category > 1 0.148 0.165 0.901
C
Category > 1 0.667 0.558 1.197
Category > 2 1.269 0.821 1.546
Latent Class Pattern 1 1 Compared to Latent Class Pattern 1 3
RACE
Category > 1 1.000 0.000 999.000
P
Category > 1 48.402 47.459 1.020
Category > 2 31.246 17.368 1.799
A
Category > 1 ********* 0.000 999.000
U
Category > 1 4.923 7.186 0.685
C
Category > 1 8.738 6.678 1.308
Category > 2 84.816 644.358 0.132
Latent Class Pattern 1 1 Compared to Latent Class Pattern 2 1
RACE
Category > 1 0.000 0.000 999.000
P
Category > 1 1.000 0.000 999.000
Category > 2 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
U
Category > 1 1.000 0.000 999.000
C
Category > 1 1.000 0.000 999.000
Category > 2 1.000 0.000 999.000
Latent Class Pattern 1 1 Compared to Latent Class Pattern 2 2
RACE
Category > 1 0.000 0.000 999.000
P
Category > 1 53.465 55.195 0.969
Category > 2 41.833 29.868 1.401
A
Category > 1 ********* 0.000 999.000
U
Category > 1 0.148 0.165 0.901
C
Category > 1 0.667 0.558 1.197
Category > 2 1.269 0.821 1.546
Latent Class Pattern 1 1 Compared to Latent Class Pattern 2 3
RACE
Category > 1 0.000 0.000 999.000
P
Category > 1 48.402 47.459 1.020
Category > 2 31.246 17.368 1.799
A
Category > 1 ********* 0.000 999.000
U
Category > 1 4.923 7.186 0.685
C
Category > 1 8.738 6.678 1.308
Category > 2 84.816 644.358 0.132
Latent Class Pattern 1 2 Compared to Latent Class Pattern 1 3
RACE
Category > 1 1.000 0.000 999.000
P
Category > 1 0.905 0.363 2.492
Category > 2 0.747 0.657 1.138
A
Category > 1 0.862 0.241 3.573
U
Category > 1 33.199 31.118 1.067
C
Category > 1 13.097 7.172 1.826
Category > 2 66.848 488.469 0.137
Latent Class Pattern 1 2 Compared to Latent Class Pattern 2 1
RACE
Category > 1 0.000 0.000 999.000
P
Category > 1 0.019 0.019 0.969
Category > 2 0.024 0.017 1.401
A
Category > 1 0.000 0.000 999.000
U
Category > 1 6.744 7.482 0.901
C
Category > 1 1.499 1.253 1.197
Category > 2 0.788 0.510 1.546
Latent Class Pattern 1 2 Compared to Latent Class Pattern 2 2
RACE
Category > 1 0.000 0.000 999.000
P
Category > 1 1.000 0.000 999.000
Category > 2 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
U
Category > 1 1.000 0.000 999.000
C
Category > 1 1.000 0.000 999.000
Category > 2 1.000 0.000 999.000
Latent Class Pattern 1 2 Compared to Latent Class Pattern 2 3
RACE
Category > 1 0.000 0.000 999.000
P
Category > 1 0.905 0.363 2.492
Category > 2 0.747 0.657 1.138
A
Category > 1 0.862 0.241 3.573
U
Category > 1 33.199 31.118 1.067
C
Category > 1 13.097 7.172 1.826
Category > 2 66.848 488.469 0.137
Latent Class Pattern 1 3 Compared to Latent Class Pattern 2 1
RACE
Category > 1 0.000 0.000 999.000
P
Category > 1 0.021 0.020 1.020
Category > 2 0.032 0.018 1.799
A
Category > 1 0.000 0.000 999.000
U
Category > 1 0.203 0.297 0.685
C
Category > 1 0.114 0.087 1.308
Category > 2 0.012 0.090 0.132
Latent Class Pattern 1 3 Compared to Latent Class Pattern 2 2
RACE
Category > 1 0.000 0.000 999.000
P
Category > 1 1.105 0.443 2.492
Category > 2 1.339 1.177 1.138
A
Category > 1 1.161 0.325 3.573
U
Category > 1 0.030 0.028 1.067
C
Category > 1 0.076 0.042 1.826
Category > 2 0.015 0.109 0.137
Latent Class Pattern 1 3 Compared to Latent Class Pattern 2 3
RACE
Category > 1 0.000 0.000 999.000
P
Category > 1 1.000 0.000 999.000
Category > 2 1.000 0.000 999.000
A
Category > 1 1.000 0.000 999.000
U
Category > 1 1.000 0.000 999.000
C
Category > 1 1.000 0.000 999.000
Category > 2 1.000 0.000 999.000
Latent Class Pattern 2 1 Compared to Latent Class Pattern 2 2
RACE
Category > 1 1.000 0.000 999.000
P
Category > 1 53.465 55.195 0.969
Category > 2 41.833 29.868 1.401
A
Category > 1 ********* 0.000 999.000
U
Category > 1 0.148 0.165 0.901
C
Category > 1 0.667 0.558 1.197
Category > 2 1.269 0.821 1.546
Latent Class Pattern 2 1 Compared to Latent Class Pattern 2 3
RACE
Category > 1 1.000 0.000 999.000
P
Category > 1 48.402 47.459 1.020
Category > 2 31.246 17.368 1.799
A
Category > 1 ********* 0.000 999.000
U
Category > 1 4.923 7.186 0.685
C
Category > 1 8.738 6.678 1.308
Category > 2 84.816 644.358 0.132
Latent Class Pattern 2 2 Compared to Latent Class Pattern 2 3
RACE
Category > 1 1.000 0.000 999.000
P
Category > 1 0.905 0.363 2.492
Category > 2 0.747 0.657 1.138
A
Category > 1 0.862 0.241 3.573
U
Category > 1 33.199 31.118 1.067
C
Category > 1 13.097 7.172 1.826
Category > 2 66.848 488.469 0.137
QUALITY OF NUMERICAL RESULTS
Condition Number for the Information Matrix 0.972E-04
(ratio of smallest to largest eigenvalue)
Page 73
Unrestricted, heterogeneous T-class model
Partial homogeneity models
Restricted, complete homogeneity model