Omega Values are too High for the Bifactor model I created

48 Views Asked by At

I have built a bifactor model in Rstudio by taking items with negative variances off of the items they loaded on until I got a model that converged. However, I need to get the omega statistics for this model and when I use the omegaFromSem function on the fitted model, I get omega values of over 20 when they should be less than 1. Why are my omega values so high?

Below is the code and results to create my model and get the omega:

#Create a model for bifactor analysis
model_bif <- '

#Define the general factor
g =~ DTW_1 + DTW_2 + DTW_3 + DTW_6 + DTW_7 + DTW_8 + DTW_9 + DTW_10 + DTW_11 + DTW_12 + DTW_13 + DTW_14 + DTW_15 + DTW_16 + DTW_17 + DTW_18 + DTW_19 + DTW_20 + DTW_21 + DTW_22

#Define the specific factors
n =~ DTW_1 + DTW_2 + DTW_3 + DTW_4 + DTW_5 + DTW_6
m =~ DTW_7 + DTW_8 + DTW_9 + DTW_10
s=~ DTW_17 + DTW_19 + DTW_21

'

#Fit the model
fit_bif <- cfa(model_bif, data = dataset, orthogonal = TRUE)
summary(fit_bif, fit.measures = TRUE, standardized = TRUE)

lavaan 0.6.15 ended normally after 97 iterations
Estimator                                         ML
Optimization method                           NLMINB
Number of model parameters                        55

                                              Used       Total
Number of observations                           296         302

Model Test User Model:
                                                  
Test statistic                               467.855
Degrees of freedom                               198
P-value (Chi-square)                           0.000

Model Test Baseline Model:

Test statistic                              3266.707
Degrees of freedom                               231
P-value                                        0.000

User Model versus Baseline Model:

Comparative Fit Index (CFI)                    0.911
Tucker-Lewis Index (TLI)                       0.896

Loglikelihood and Information Criteria:

Loglikelihood user model (H0)              -6953.172
Loglikelihood unrestricted model (H1)      -6719.244
                                                  
Akaike (AIC)                               14016.344
Bayesian (BIC)                             14219.313
Sample-size adjusted Bayesian (SABIC)      14044.891

Root Mean Square Error of Approximation:

RMSEA                                          0.068
90 Percent confidence interval - lower         0.060
90 Percent confidence interval - upper         0.076
P-value H_0: RMSEA <= 0.050                    0.000
P-value H_0: RMSEA >= 0.080                    0.006

Standardized Root Mean Square Residual:

SRMR                                           0.056

Parameter Estimates:

Standard errors                             Standard
Information                                 Expected
Information saturated (h1) model          Structured

Latent Variables:
                Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
g =~                                                                  
DTW_1             1.000                               0.155    0.141
DTW_2             2.552    0.993    2.570    0.010    0.397    0.349
DTW_3             2.265    0.918    2.468    0.014    0.352    0.283
DTW_6             2.589    0.988    2.622    0.009    0.402    0.366
DTW_7             2.709    1.085    2.497    0.013    0.421    0.377
DTW_8             3.002    1.202    2.498    0.012    0.466    0.379
DTW_9             2.557    1.032    2.477    0.013    0.397    0.360
DTW_10            2.485    1.032    2.408    0.016    0.386    0.311
DTW_11            3.322    1.246    2.667    0.008    0.516    0.795
DTW_12            3.459    1.303    2.654    0.008    0.537    0.717
DTW_13            3.183    1.208    2.635    0.008    0.495    0.632
DTW_14            3.588    1.349    2.660    0.008    0.557    0.752
DTW_15            3.081    1.170    2.633    0.008    0.479    0.623
DTW_16            2.946    1.103    2.671    0.008    0.458    0.826
DTW_17            3.208    1.204    2.665    0.008    0.498    0.783
DTW_18            3.506    1.330    2.636    0.008    0.545    0.635
DTW_19            3.132    1.182    2.650    0.008    0.487    0.699
DTW_20            3.170    1.203    2.635    0.008    0.493    0.630
DTW_21            3.135    1.178    2.662    0.008    0.487    0.763
DTW_22            3.112    1.172    2.655    0.008    0.484    0.724
n =~                                                                  
DTW_1             1.000                               0.648    0.589
DTW_2             0.617    0.121    5.113    0.000    0.400    0.352
DTW_3             0.643    0.133    4.816    0.000    0.417    0.335
DTW_4             1.038    0.137    7.560    0.000    0.673    0.644
DTW_5             1.063    0.135    7.846    0.000    0.689    0.722
DTW_6             0.748    0.121    6.181    0.000    0.485    0.442
m =~                                                                  
DTW_7             1.000                               0.770    0.691
DTW_8             0.726    0.103    7.030    0.000    0.559    0.454
DTW_9             0.926    0.101    9.136    0.000    0.714    0.646
DTW_10            1.034    0.115    9.020    0.000    0.797    0.642
s =~                                                                  
DTW_17            1.000                               0.224    0.351
DTW_19            1.310    0.228    5.739    0.000    0.293    0.421
DTW_21            1.318    0.240    5.486    0.000    0.295    0.462

Covariances:
                 Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
g ~~                                                                  
 n                 0.000                               0.000    0.000
 m                 0.000                               0.000    0.000
 s                 0.000                               0.000    0.000
n ~~                                                                  
 m                 0.000                               0.000    0.000
 s                 0.000                               0.000    0.000
m ~~                                                                  
 s                 0.000                               0.000    0.000

Variances:
               Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
.DTW_1             0.767    0.078    9.792    0.000    0.767    0.633
.DTW_2             0.977    0.086   11.388    0.000    0.977    0.755
.DTW_3             1.249    0.108   11.534    0.000    1.249    0.808
.DTW_6             0.809    0.075   10.834    0.000    0.809    0.671
.DTW_7             0.473    0.065    7.248    0.000    0.473    0.380
.DTW_8             0.987    0.090   10.958    0.000    0.987    0.650
.DTW_9             0.552    0.065    8.526    0.000    0.552    0.453
.DTW_10            0.757    0.085    8.876    0.000    0.757    0.491
.DTW_11            0.155    0.015   10.528    0.000    0.155    0.368
.DTW_12            0.273    0.024   11.159    0.000    0.273    0.486
.DTW_13            0.367    0.032   11.534    0.000    0.367    0.600
.DTW_14            0.239    0.022   10.930    0.000    0.239    0.435
.DTW_15            0.362    0.031   11.565    0.000    0.362    0.612
.DTW_16            0.098    0.010   10.114    0.000    0.098    0.318
.DTW_17            0.107    0.012    8.823    0.000    0.107    0.264
.DTW_18            0.440    0.038   11.525    0.000    0.440    0.597
.DTW_19            0.161    0.019    8.340    0.000    0.161    0.333
.DTW_20            0.369    0.032   11.541    0.000    0.369    0.603
.DTW_21            0.084    0.016    5.282    0.000    0.084    0.205
.DTW_22            0.213    0.019   11.119    0.000    0.213    0.476
.DTW_4             0.638    0.071    9.011    0.000    0.638    0.585
.DTW_5             0.437    0.059    7.401    0.000    0.437    0.479
g                  0.024    0.018    1.336    0.182    1.000    1.000
n                  0.420    0.088    4.777    0.000    1.000    1.000
m                  0.594    0.096    6.213    0.000    1.000    1.000
s                  0.050    0.014    3.585    0.000    1.000    1.000



#Get omega functions
omegaFromSem(fit_bif, m = NULL, flip = TRUE, plot = TRUE)

The following analyses were done using the  lavaan  package 

Omega Hierarchical from a confirmatory model using sem =  21.24
Omega Total  from a confirmatory model using sem =  2.04 
With loadings of 
          g  F1*  F2*  F3*    h2     u2   p2
DTW_1   1.0 1.00            2.00  -1.00 0.50
DTW_2   2.5 0.62            6.89  -5.89 0.94
DTW_3   2.3 0.64            5.54  -4.54 0.92
DTW_6   2.6 0.75            7.26  -6.26 0.92
DTW_7   2.7      1.00       8.34  -7.34 0.88
DTW_8   3.0      0.73       9.54  -8.54 0.94
DTW_9   2.6      0.93       7.40  -6.40 0.89
DTW_10  2.5      1.03       7.24  -6.24 0.85
DTW_11  3.3                11.04 -10.04 1.00
DTW_12  3.5                11.96 -10.96 1.00
DTW_13  3.2                10.13  -9.13 1.00
DTW_14  3.6                12.87 -11.87 1.00
DTW_15  3.1                 9.50  -8.50 1.00
DTW_16  3.0                 8.68  -7.68 1.00
DTW_17  3.2            1.0 11.29 -10.29 0.91
DTW_18  3.5                12.29 -11.29 1.00
DTW_19  3.1            1.3 11.52 -10.52 0.85
DTW_20  3.2                10.05  -9.05 1.00
DTW_21  3.1            1.3 11.57 -10.57 0.85
DTW_22  3.1                 9.68  -8.68 1.00
DTW_4       1.04            1.08  -0.08 0.00
DTW_5       1.06            1.13  -0.13 0.00

With sum of squared loadings of:
    g   F1*   F2*   F3* 
174.5   4.6   3.5   4.5 

The degrees of freedom of the confirmatory model are  198  and the fit is  467.8555  with p =  0
general/max  38.27   max/min =   1.32
mean percent general =  0.84    with sd =  0.29 and cv of  0.35 
Explained Common Variance of the general factor =  0.93 

Measures of factor score adequacy             
                                                  g  F1*  F2*   F3*
Correlation of scores with factors             5.08 1.50 1.45  2.36
Multiple R square of scores with factors      25.84 2.24 2.11  5.55
Minimum correlation of factor score estimates 50.68 3.48 3.22 10.10

 Total, General and Subset omega for each subset
                                                  g   F1*   F2*   F3*
Omega total for total scores and subscales     2.04 19.18 13.00 13.98
Omega general for total scores and subscales  21.24 18.83 11.63 12.19
Omega group for total scores and subscales     0.33  0.34  1.37  1.79

To get the standard sem fit statistics, ask for summary on the fitted object> 
0

There are 0 best solutions below