multi-group lavaan vs. LISREL

Salvador Ruiz

unread,

Jan 14, 2014, 6:37:38 PM1/14/14

to lav...@googlegroups.com

Hi,

I am running the same multi group analysis with lavaan and LISREL (SIMPLIS syntax), but obtain different result. These are the two scripts

SIMPLIS (Calculating the two covariance matrices)

Group 1: Testing Moderating Effects (grupo Bajo)

Observed variables

P9_3 P9_5 P9_6 P9_7 P9_12 P9_13

P9_16 P9_17 P9_18 P9_19 P9_20 P9_21

P11_3 P11_4 P11_5

Covariance Matrix From File base0_202

Sample Size: 202

Latent Variables: AMENPERC SUSCPERC BENCONSA AUTOEFIC PPFARM

Relationships:

P9_19 = 1*AMENPERC

P9_6 P9_7 P9_13= AMENPERC

P9_12 = 1*SUSCPERC

P9_21 = SUSCPERC

P9_16 = 1*BENCONSA

P9_17 P9_20 = BENCONSA

P9_18 = 1*AUTOEFIC

P9_3 P9_5 = AUTOEFIC

P11_3 = 1*PPFARM

P11_4 P11_5 = PPFARM

PPFARM = BENCONSA AUTOEFIC

AMENPERC = SUSCPERC

AUTOEFIC = AMENPERC

Group 2: Testing Moderating Effects (grupo Alto)

Covariance Matrix From File base1_149

Sample Size: 149

Number of decimals = 2

Path Diagram

Lisrel Output SC AD=OFF EF RS MI

End of Problem

In LAVAAN, I run

HBMStrmodelMG <- '

# measurement model

AMENPERC =~ P9.19 + P9.6 + P9.7 + P9.13

SUSCPERC =~ P9.12 + P9.21

BENCONSA =~ P9.16 + P9.17 + P9.20

AUTOEFIC =~ P9.18 + P9.3 + P9.5

PPFARM =~ P11.3 + P11.4 + P11.5

# regressions

PPFARM ~ BENCONSA + AUTOEFIC

AMENPERC ~ SUSCPERC

AUTOEFIC ~ AMENPERC

'

fitSR <- sem(HBMStrmodelMG, data = mydataCFAMarta, estimator = "MLM", group = "V1",

group.equal=c("loadings", "regressions", "residuals",

"residual.covariances", "lv.variances", "lv.covariances"))

summary(fitSR, fit.measures = TRUE)

Can anyone tell me if I am use equivalent scripts?

Thank you very much

Salvador

Jeremy Miles

unread,

Jan 14, 2014, 6:41:23 PM1/14/14

to lav...@googlegroups.com

Can we see the differences in the results?

J

--
You received this message because you are subscribed to the Google Groups "lavaan" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lavaan+un...@googlegroups.com.
To post to this group, send email to lav...@googlegroups.com.
Visit this group at http://groups.google.com/group/lavaan.
For more options, visit https://groups.google.com/groups/opt_out.

Salvador Ruiz

unread,

Jan 15, 2014, 9:22:04 AM1/15/14

to lav...@googlegroups.com

Hi Jeremy, thanks for you answer, here they go

LISREL results

Group 1: Testing Moderating Effects (grupo Bajo)

Number of Iterations = 14

LISREL Estimates (Maximum Likelihood)

LAMBDA-Y EQUALS LAMBDA-Y IN THE FOLLOWING GROUP

LAMBDA-X EQUALS LAMBDA-X IN THE FOLLOWING GROUP

BETA EQUALS BETA IN THE FOLLOWING GROUP

GAMMA EQUALS GAMMA IN THE FOLLOWING GROUP

Covariance Matrix of ETA and KSI

AMENPERC AUTOEFIC PPFARM SUSCPERC BENCONSA

-------- -------- -------- -------- --------

AMENPERC 10.86

AUTOEFIC -3.43 4.38

PPFARM -2.95 2.73 12.55

SUSCPERC 4.91 -1.55 -1.57 3.28

BENCONSA -2.36 0.74 2.58 -1.57 4.73

PHI EQUALS PHI IN THE FOLLOWING GROUP

PSI EQUALS PSI IN THE FOLLOWING GROUP

THETA-EPS EQUALS THETA-EPS IN THE FOLLOWING GROUP

Group Goodness of Fit Statistics

Contribution to Chi-Square = 1089.58

Percentage Contribution to Chi-Square = 66.06

Root Mean Square Residual (RMR) = 2.73

Standardized RMR = 0.22

Goodness of Fit Index (GFI) = 0.53

Group 1: Testing Moderating Effects (grupo Bajo)

Fitted Covariance Matrix

P9_3 P9_5 P9_6 P9_7 P9_13 P9_18

-------- -------- -------- -------- -------- --------

P9_3 19.18

P9_5 4.03 3.32

P9_6 -3.72 -1.27 8.46

P9_7 -4.37 -1.49 5.56 13.40

P9_13 -5.15 -1.75 6.56 7.70 12.84

P9_18 7.20 2.45 -2.26 -2.66 -3.13 5.56

P9_19 -5.63 -1.92 7.17 8.42 9.93 -3.43

P11_3 4.48 1.53 -1.95 -2.29 -2.70 2.73

P11_4 3.25 1.11 -1.41 -1.66 -1.96 1.98

P11_5 2.36 0.80 -1.02 -1.20 -1.42 1.43

P9_12 -2.55 -0.87 3.24 3.81 4.49 -1.55

P9_16 1.22 0.42 -1.56 -1.83 -2.15 0.74

P9_17 1.12 0.38 -1.43 -1.68 -1.98 0.68

P9_20 1.36 0.46 -1.73 -2.03 -2.40 0.83

P9_21 -3.42 -1.16 4.35 5.11 6.02 -2.08

Fitted Covariance Matrix

P9_19 P11_3 P11_4 P11_5 P9_12 P9_16

-------- -------- -------- -------- -------- --------

P9_19 15.72

P11_3 -2.95 23.26

P11_4 -2.14 9.10 7.31

P11_5 -1.55 6.60 4.79 9.71

P9_12 4.91 -1.57 -1.14 -0.82 5.20

P9_16 -2.36 2.58 1.87 1.36 -1.57 9.86

P9_17 -2.16 2.36 1.71 1.24 -1.44 4.34

P9_20 -2.62 2.87 2.08 1.51 -1.75 5.27

P9_21 6.59 -2.10 -1.52 -1.11 4.40 -2.11

Fitted Covariance Matrix

P9_17 P9_20 P9_21

-------- -------- --------

P9_17 6.42

P9_20 4.84 8.66

P9_21 -1.94 -2.35 9.59

Fitted Residuals

P9_3 P9_5 P9_6 P9_7 P9_13 P9_18

-------- -------- -------- -------- -------- --------

P9_3 -16.12

P9_5 -2.01 0.04

P9_6 1.86 0.80 0.11

P9_7 1.93 -0.35 1.13 0.99

P9_13 3.25 0.66 -1.79 -1.80 -5.44

P9_18 -5.83 -1.25 1.07 1.03 1.76 -4.10

P9_19 3.96 1.00 -2.57 -2.83 -5.37 1.93

P11_3 -4.14 -1.09 2.13 2.57 2.11 -1.77

P11_4 -3.16 -0.73 1.33 1.13 1.06 -1.68

P11_5 -2.03 -0.23 0.13 -0.04 0.52 -0.97

P9_12 1.10 -0.13 -1.17 -0.31 -2.16 0.46

P9_16 -0.77 0.47 0.50 0.06 0.38 0.02

P9_17 -0.25 0.60 -0.03 -0.15 -0.21 0.32

P9_20 -0.73 0.25 0.98 0.68 0.89 0.11

P9_21 1.08 -0.40 0.67 1.06 -1.89 0.27

Fitted Residuals

P9_19 P11_3 P11_4 P11_5 P9_12 P9_16

-------- -------- -------- -------- -------- --------

P9_19 -8.99

P11_3 3.13 -9.77

P11_4 1.74 -5.58 -5.15

P11_5 1.11 -1.94 -2.30 -3.62

P9_12 -2.51 1.23 0.70 -0.26 -1.42

P9_16 0.83 -1.43 -1.47 -0.69 0.60 -6.12

P9_17 0.36 1.17 -1.24 -0.57 0.39 -1.32

P9_20 0.61 -0.80 -1.76 -0.67 0.84 -2.87

P9_21 -2.26 0.78 0.52 -0.61 -0.49 0.29

Fitted Residuals

P9_17 P9_20 P9_21

-------- -------- --------

P9_17 -0.30

P9_20 -1.33 -4.63

P9_21 -0.42 1.15 1.14

Group 2: Testing Moderating Effects (grupo Alto)

Number of Iterations = 14

LISREL Estimates (Maximum Likelihood)

LAMBDA-Y

AMENPERC AUTOEFIC PPFARM

-------- -------- --------

P9_3 - - 1.64 - -

(0.11)

14.42

P9_5 - - 0.56 - -

(0.05)

11.99

P9_6 0.66 - - - -

(0.04)

15.32

P9_7 0.78 - - - -

(0.06)

14.00

P9_13 0.91 - - - -

(0.05)

17.80

P9_18 - - 1.00 - -

P9_19 1.00 - - - -

P11_3 - - - - 1.00

P11_4 - - - - 0.73

(0.06)

12.91

P11_5 - - - - 0.53

(0.05)

10.92

LAMBDA-X

SUSCPERC BENCONSA

-------- --------

P9_12 1.00 - -

P9_16 - - 1.00

P9_17 - - 0.92

(0.08)

12.04

P9_20 - - 1.11

(0.09)

12.14

P9_21 1.34 - -

(0.10)

13.34

BETA

AMENPERC AUTOEFIC PPFARM

-------- -------- --------

AMENPERC - - - - - -

AUTOEFIC -0.32 - - - -

(0.04)

-8.39

PPFARM - - 0.54 - -

(0.10)

5.25

GAMMA

SUSCPERC BENCONSA

-------- --------

AMENPERC 1.50 - -

(0.12)

12.11

AUTOEFIC - - - -

PPFARM - - 0.46

(0.10)

4.51

Covariance Matrix of ETA and KSI

AMENPERC AUTOEFIC PPFARM SUSCPERC BENCONSA

-------- -------- -------- -------- --------

AMENPERC 10.86

AUTOEFIC -3.43 4.38

PPFARM -2.95 2.73 12.55

SUSCPERC 4.91 -1.55 -1.57 3.28

BENCONSA -2.36 0.74 2.58 -1.57 4.73

PHI

SUSCPERC BENCONSA

-------- --------

SUSCPERC 3.28

(0.41)

8.01

BENCONSA -1.57 4.73

(0.29) (0.70)

-5.39 6.75

PSI

Note: This matrix is diagonal.

AMENPERC AUTOEFIC PPFARM

-------- -------- --------

3.51 3.30 9.88

(0.61) (0.38) (1.36)

5.72 8.65 7.27

Squared Multiple Correlations for Structural Equations

AMENPERC AUTOEFIC PPFARM

-------- -------- --------

0.68 0.25 0.21

Squared Multiple Correlations for Reduced Form

AMENPERC AUTOEFIC PPFARM

-------- -------- --------

0.68 0.17 0.13

Reduced Form

SUSCPERC BENCONSA

-------- --------

AMENPERC 1.50 - -

(0.12)

12.11

AUTOEFIC -0.47 - -

(0.06)

-7.47

PPFARM -0.26 0.46

(0.06) (0.10)

-4.41 4.51

THETA-EPS

P9_3 P9_5 P9_6 P9_7 P9_13 P9_18

-------- -------- -------- -------- -------- --------

7.35 1.95 3.72 6.87 3.76 1.18

(0.84) (0.17) (0.33) (0.59) (0.41) (0.24)

8.79 11.59 11.12 11.66 9.19 4.87

THETA-EPS

P9_19 P11_3 P11_4 P11_5

-------- -------- -------- --------

4.85 10.71 0.71 6.24

(0.51) (1.10) (0.39) (0.52)

9.48 9.71 1.82 12.00

Global Goodness of Fit Statistics

Degrees of Freedom = 205

Minimum Fit Function Chi-Square = 1649.36 (P = 0.0)

Normal Theory Weighted Least Squares Chi-Square = 1423.65 (P = 0.0)

Estimated Non-centrality Parameter (NCP) = 1218.65

90 Percent Confidence Interval for NCP = (1102.77 ; 1341.99)

Minimum Fit Function Value = 4.73

Population Discrepancy Function Value (F0) = 3.49

90 Percent Confidence Interval for F0 = (3.16 ; 3.85)

Root Mean Square Error of Approximation (RMSEA) = 0.18

90 Percent Confidence Interval for RMSEA = (0.18 ; 0.19)

P-Value for Test of Close Fit (RMSEA < 0.05) = 0.00

Expected Cross-Validation Index (ECVI) = 4.28

90 Percent Confidence Interval for ECVI = (3.95 ; 4.63)

ECVI for Saturated Model = 0.69

ECVI for Independence Model = 13.88

Chi-Square for Independence Model with 210 Degrees of Freedom = 4813.28

Independence AIC = 4873.28

Model AIC = 1493.65

Saturated AIC = 480.00

Independence CAIC = 5019.10

Model CAIC = 1663.77

Saturated CAIC = 1646.59

Normed Fit Index (NFI) = 0.66

Non-Normed Fit Index (NNFI) = 0.68

Parsimony Normed Fit Index (PNFI) = 0.64

Comparative Fit Index (CFI) = 0.69

Incremental Fit Index (IFI) = 0.69

Relative Fit Index (RFI) = 0.65

Critical N (CN) = 54.96

Group Goodness of Fit Statistics

Contribution to Chi-Square = 559.78

Percentage Contribution to Chi-Square = 33.94

Root Mean Square Residual (RMR) = 4.12

Standardized RMR = 0.35

Goodness of Fit Index (GFI) = 0.69

Group 2: Testing Moderating Effects (grupo Alto)

Fitted Covariance Matrix

P9_3 P9_5 P9_6 P9_7 P9_13 P9_18

-------- -------- -------- -------- -------- --------

P9_3 19.18

P9_5 4.03 3.32

P9_6 -3.72 -1.27 8.46

P9_7 -4.37 -1.49 5.56 13.40

P9_13 -5.15 -1.75 6.56 7.70 12.84

P9_18 7.20 2.45 -2.26 -2.66 -3.13 5.56

P9_19 -5.63 -1.92 7.17 8.42 9.93 -3.43

P11_3 4.48 1.53 -1.95 -2.29 -2.70 2.73

P11_4 3.25 1.11 -1.41 -1.66 -1.96 1.98

P11_5 2.36 0.80 -1.02 -1.20 -1.42 1.43

P9_12 -2.55 -0.87 3.24 3.81 4.49 -1.55

P9_16 1.22 0.42 -1.56 -1.83 -2.15 0.74

P9_17 1.12 0.38 -1.43 -1.68 -1.98 0.68

P9_20 1.36 0.46 -1.73 -2.03 -2.40 0.83

P9_21 -3.42 -1.16 4.35 5.11 6.02 -2.08

Fitted Covariance Matrix

P9_19 P11_3 P11_4 P11_5 P9_12 P9_16

-------- -------- -------- -------- -------- --------

P9_19 15.72

P11_3 -2.95 23.26

P11_4 -2.14 9.10 7.31

P11_5 -1.55 6.60 4.79 9.71

P9_12 4.91 -1.57 -1.14 -0.82 5.20

P9_16 -2.36 2.58 1.87 1.36 -1.57 9.86

P9_17 -2.16 2.36 1.71 1.24 -1.44 4.34

P9_20 -2.62 2.87 2.08 1.51 -1.75 5.27

P9_21 6.59 -2.10 -1.52 -1.11 4.40 -2.11

Fitted Covariance Matrix

P9_17 P9_20 P9_21

-------- -------- --------

P9_17 6.42

P9_20 4.84 8.66

P9_21 -1.94 -2.35 9.59

Fitted Residuals

P9_3 P9_5 P9_6 P9_7 P9_13 P9_18

-------- -------- -------- -------- -------- --------

P9_3 21.89

P9_5 2.56 -0.06

P9_6 -4.61 -0.29 -0.15

P9_7 -3.62 -0.13 0.38 -1.34

P9_13 -4.23 -1.39 2.09 1.81 7.39

P9_18 7.83 1.74 -1.30 -0.85 -1.58 5.57

P9_19 -4.59 -1.29 3.12 1.87 8.91 -1.44

P11_3 5.80 1.11 2.72 3.12 1.89 4.58

P11_4 5.97 0.90 1.37 -0.25 0.60 3.69

P11_5 4.02 1.32 0.17 -1.91 -1.67 3.49

P9_12 -4.47 -1.05 0.73 1.33 2.47 -2.21

P9_16 3.77 1.09 0.91 -0.79 0.78 4.26

P9_17 2.94 1.02 1.10 1.13 1.12 3.07

P9_20 7.15 1.43 -0.08 0.14 -1.44 4.80

P9_21 -3.19 -1.30 0.02 0.38 1.27 -1.72

Fitted Residuals

P9_19 P11_3 P11_4 P11_5 P9_12 P9_16

-------- -------- -------- -------- -------- --------

P9_19 12.21

P11_3 3.16 14.48

P11_4 1.33 8.52 7.64

P11_5 -2.60 1.70 3.67 5.25

P9_12 3.31 2.25 0.87 -1.70 1.92

P9_16 -2.20 5.08 1.79 1.09 -0.83 8.31

P9_17 -0.89 4.97 2.73 1.58 -0.54 1.97

P9_20 -4.09 4.69 4.16 3.15 -1.87 3.89

P9_21 2.51 2.32 0.52 -0.89 0.69 -0.18

Fitted Residuals

P9_17 P9_20 P9_21

-------- -------- --------

P9_17 0.40

P9_20 1.67 6.29

P9_21 0.46 -0.82 -1.55

LAVAAN results

> HBMStrmodelMG <- '

+ # measurement model

+ AMENPERC =~ P9.19 + P9.6 + P9.7 + P9.13

+ SUSCPERC =~ P9.12 + P9.21

+ BENCONSA =~ P9.16 + P9.17 + P9.20

+ AUTOEFIC =~ P9.18 + P9.3 + P9.5

+ PPFARM =~ P11.3 + P11.4 + P11.5

+ # regressions

+ PPFARM ~ BENCONSA + AUTOEFIC

+ AMENPERC ~ SUSCPERC

+ AUTOEFIC ~ AMENPERC

+ '

> fitSR <- sem(HBMStrmodelMG, data = mydataCFAMarta, estimator = "MLM", group = "V1",

+ group.equal=c("loadings", "regressions", "residuals",

+ "residual.covariances", "lv.variances", "lv.covariances"))

> summary(fitSR, fit.measures = TRUE)

lavaan (0.5-15) converged normally after 50 iterations

Number of observations per group

0 202

1 149

Estimator ML Robust

Minimum Function Test Statistic 330.350 292.720

Degrees of freedom 205 205

P-value (Chi-square) 0.000 0.000

Scaling correction factor 1.129

for the Satorra-Bentler correction

Chi-square for each group:

0 160.314 142.053

1 170.036 150.667

Model test baseline model:

Minimum Function Test Statistic 1978.304 1741.841

Degrees of freedom 210 210

P-value 0.000 0.000

User model versus baseline model:

Comparative Fit Index (CFI) 0.929 0.943

Tucker-Lewis Index (TLI) 0.927 0.941

Loglikelihood and Information Criteria:

Loglikelihood user model (H0) -9188.684 -9188.684

Loglikelihood unrestricted model (H1) -9023.509 -9023.509

Number of free parameters 65 65

Akaike (AIC) 18507.367 18507.367

Bayesian (BIC) 18758.318 18758.318

Sample-size adjusted Bayesian (BIC) 18552.114 18552.114

Root Mean Square Error of Approximation:

RMSEA 0.059 0.049

90 Percent Confidence Interval 0.047 0.071 0.037 0.061

P-value RMSEA <= 0.05 0.104 0.523

Standardized Root Mean Square Residual:

SRMR 0.091 0.091

Parameter estimates:

Information Expected

Standard Errors Robust.sem

Group 1 [0]:

Estimate Std.err Z-value P(>|z|)

Latent variables:

AMENPERC =~

P9.19 1.000

P9.6 0.885 0.061 14.488 0.000

P9.7 0.801 0.059 13.526 0.000

P9.13 0.899 0.054 16.736 0.000

SUSCPERC =~

P9.12 1.000

P9.21 1.164 0.110 10.585 0.000

BENCONSA =~

P9.16 1.000

P9.17 1.015 0.096 10.561 0.000

P9.20 0.831 0.094 8.853 0.000

AUTOEFIC =~

P9.18 1.000

P9.3 0.877 0.075 11.703 0.000

P9.5 0.987 0.098 10.100 0.000

PPFARM =~

P11.3 1.000

P11.4 0.878 0.120 7.299 0.000

P11.5 0.916 0.152 6.028 0.000

Regressions:

PPFARM ~

BENCONSA 0.125 0.058 2.147 0.032

AUTOEFIC 0.073 0.035 2.096 0.036

AMENPERC ~

SUSCPERC 1.113 0.112 9.930 0.000

AUTOEFIC ~

AMENPERC -0.385 0.049 -7.920 0.000

Covariances:

SUSCPERC ~~

BENCONSA -0.453 0.111 -4.067 0.000

Intercepts:

P9.19 3.376 0.158 21.308 0.000

P9.6 4.223 0.151 27.887 0.000

P9.7 3.149 0.157 20.027 0.000

P9.13 3.272 0.146 22.432 0.000

P9.12 3.045 0.138 22.071 0.000

P9.21 3.926 0.133 29.482 0.000

P9.16 4.827 0.125 38.598 0.000

P9.17 5.743 0.111 51.924 0.000

P9.20 5.851 0.104 56.367 0.000

P9.18 5.436 0.116 46.802 0.000

P9.3 5.698 0.111 51.215 0.000

P9.5 4.129 0.136 30.324 0.000

P11.3 6.059 0.094 64.776 0.000

P11.4 6.559 0.058 112.599 0.000

P11.5 6.089 0.090 67.913 0.000

AMENPERC 0.000

SUSCPERC 0.000

BENCONSA 0.000

AUTOEFIC 0.000

PPFARM 0.000

Variances:

P9.19 1.884 0.244

P9.6 2.068 0.264

P9.7 2.654 0.298

P9.13 1.805 0.217

P9.12 1.889 0.223

P9.21 1.319 0.192

P9.16 1.855 0.264

P9.17 0.828 0.138

P9.20 0.827 0.191

P9.18 1.041 0.160

P9.3 1.146 0.148

P9.5 2.116 0.192

P11.3 0.935 0.116

P11.4 0.133 0.050

P11.5 0.865 0.119

AMENPERC 1.140 0.217

SUSCPERC 1.666 0.268

BENCONSA 1.327 0.209

AUTOEFIC 1.242 0.201

PPFARM 0.536 0.124

Group 2 [1]:

Estimate Std.err Z-value P(>|z|)

Latent variables:

AMENPERC =~

P9.19 1.000

P9.6 0.885 0.061 14.488 0.000

P9.7 0.801 0.059 13.526 0.000

P9.13 0.899 0.054 16.736 0.000

SUSCPERC =~

P9.12 1.000

P9.21 1.164 0.110 10.585 0.000

BENCONSA =~

P9.16 1.000

P9.17 1.015 0.096 10.561 0.000

P9.20 0.831 0.094 8.853 0.000

AUTOEFIC =~

P9.18 1.000

P9.3 0.877 0.075 11.703 0.000

P9.5 0.987 0.098 10.100 0.000

PPFARM =~

P11.3 1.000

P11.4 0.878 0.120 7.299 0.000

P11.5 0.916 0.152 6.028 0.000

Regressions:

PPFARM ~

BENCONSA 0.125 0.058 2.147 0.032

AUTOEFIC 0.073 0.035 2.096 0.036

AMENPERC ~

SUSCPERC 1.113 0.112 9.930 0.000

AUTOEFIC ~

AMENPERC -0.385 0.049 -7.920 0.000

Covariances:

SUSCPERC ~~

BENCONSA -0.453 0.111 -4.067 0.000

Intercepts:

P9.19 3.383 0.187 18.135 0.000

P9.6 3.973 0.175 22.694 0.000

P9.7 3.007 0.172 17.527 0.000

P9.13 3.168 0.175 18.055 0.000

P9.12 2.463 0.147 16.782 0.000

P9.21 3.483 0.156 22.359 0.000

P9.16 5.403 0.148 36.550 0.000

P9.17 6.067 0.111 54.496 0.000

P9.20 6.356 0.089 71.520 0.000

P9.18 5.631 0.138 40.768 0.000

P9.3 5.946 0.128 46.310 0.000

P9.5 4.067 0.162 25.136 0.000

P11.3 6.309 0.089 71.063 0.000

P11.4 6.691 0.054 124.402 0.000

P11.5 6.396 0.081 78.703 0.000

AMENPERC 0.000

SUSCPERC 0.000

BENCONSA 0.000

AUTOEFIC 0.000

PPFARM 0.000

Variances:

P9.19 1.884 0.244

P9.6 2.068 0.264

P9.7 2.654 0.298

P9.13 1.805 0.217

P9.12 1.889 0.223

P9.21 1.319 0.192

P9.16 1.855 0.264

P9.17 0.828 0.138

P9.20 0.827 0.191

P9.18 1.041 0.160

P9.3 1.146 0.148

P9.5 2.116 0.192

P11.3 0.935 0.116

P11.4 0.133 0.050

P11.5 0.865 0.119

AMENPERC 1.140 0.217

SUSCPERC 1.666 0.268

BENCONSA 1.327 0.209

AUTOEFIC 1.242 0.201

PPFARM 0.536 0.124

> fitMeasures(fitSR)

fmin chisq df pvalue

0.471 330.350 205.000 0.000

chisq.scaled df.scaled pvalue.scaled chisq.scaling.factor

292.720 205.000 0.000 1.129

baseline.chisq baseline.df baseline.pvalue baseline.chisq.scaled

1978.304 210.000 0.000 1741.841

baseline.df.scaled baseline.pvalue.scaled baseline.chisq.scaling.factor cfi

210.000 0.000 1.136 0.929

tli nnfi rfi nfi

0.927 0.927 0.829 0.833

pnfi ifi rni cfi.scaled

0.813 0.929 0.929 0.943

tli.scaled nnfi.scaled rfi.scaled nfi.scaled

0.941 0.941 0.828 0.832

ifi.scaled rni.scaled logl unrestricted.logl

0.832 0.950 -9188.684 -9023.509

npar aic bic ntotal

65.000 18507.367 18758.318 351.000

bic2 rmsea rmsea.ci.lower rmsea.ci.upper

18552.114 0.059 0.047 0.071

rmsea.pvalue rmsea.scaled rmsea.ci.lower.scaled rmsea.ci.upper.scaled

0.104 0.049 0.037 0.061

rmsea.pvalue.scaled rmr rmr_nomean srmr

0.523 0.221 0.235 0.091

srmr_nomean cn_05 cn_01 gfi

0.097 255.369 271.965 0.993

agfi pgfi mfi

0.991 0.754 0.836

>

I can provide other results if needed.

Thanks a lot for your help

Salvador

yrosseel

unread,

Jan 16, 2014, 3:00:06 AM1/16/14

to lav...@googlegroups.com

On 01/15/2014 03:22 PM, Salvador Ruiz wrote:
> Can anyone tell me if I am use equivalent scripts?

I have the impression that the scripts are equivalent, but the data are
not. Would you be able to provide us with the data, and perhaps the
prelis/lisrel/simplis scripts and lavaan scripts?

Yves.

Message has been deleted

Salvador Ruiz

unread,

Jan 16, 2014, 6:30:44 PM1/16/14

to lav...@googlegroups.com

Hi Yves,

Thanks a lot for your answer. Attached are the scripts for LISREL and lavaan.

For LISREL I used

1. Two databases (one for each group: base1_149.sav and base0_202.sav) to generate the two covariance matrices with scripts base1_149.PR2 and base0_202.PR2

2. A SIMPLIS script for the multi group (NoModera.spl)

For lavaan I used

1. the script for the multi group analysis (AnalysesHBMyPERC_Confirm_Structural.R) and the whole database (base.sav)

Thanks again for your help

Best

Salvador

SEM.zip

yrosseel

unread,

Jan 17, 2014, 2:54:44 AM1/17/14

to lav...@googlegroups.com

On 01/17/2014 12:21 AM, Salvador Ruiz wrote:
> Hi Yves,

> Thanks a lot for your answer. Attached are the scripts for LISREL and
> lavaan.

Thanks. I also need two additional files: base0_202.PSF and
base1_149.PSF in order to replicate your prelis script.

Yves.

Salvador Ruiz

unread,

Jan 17, 2014, 5:12:59 AM1/17/14

to lav...@googlegroups.com

sorry, here they go.

Best

Salvador

base0_202.psf

base1_149.psf

yrosseel

unread,

Jan 18, 2014, 5:19:23 AM1/18/14

to lav...@googlegroups.com

On 01/17/2014 11:12 AM, Salvador Ruiz wrote:
> sorry, here they go.

Thanks. I ran both the prelis and lisrel/simplis scripts (unaltered,
except for the directory names), and I get completely different results
(from lisrel) compared to the output you posted. But, my lisrel results
coincide with the lavaan results. I have no explanation. Which version
of lisrel did you use? I used 8.72. In attach is the full lisrel output.
Here is just a snippet of the global fit indices:

Global Goodness of Fit Statistics

Degrees of Freedom = 205

Minimum Fit Function Chi-Square = 328.03 (P = 0.00)

Note that the chi-square is 328.03, while it was 1649.36 in your output.
Something went wrong, but I am not a lisrel expert. If we use lavaan
with the mimic="EQS" option (often very close to LISREL), we get:

Number of observations per group
0 202
1 149

Estimator ML Robust

Minimum Function Test Statistic 328.036 293.969

Degrees of freedom 205 205
P-value (Chi-square) 0.000 0.000

All estimates are as good as identical between lavaan and lisrel.

Yves.

NoModera.lisrel.txt

Salvador Ruiz

unread,

Jan 18, 2014, 7:27:33 AM1/18/14

to lav...@googlegroups.com

Interesting. Thank you very much. I used lisrel 8.80, but that should not be an issue, I think.

I am checking it again and let you know.

In any case, I am trusting more lavaan now :)

Thanks a lot again.

Salvador

Reply all

Reply to author

Forward