Table 2 Fit indices for latent growth mixture models (LGMM) for psychopathological symptoms, COVID-19-related rumination, and stress-related growth.

From: Interrelations of resilience factors and their incremental impact for mental health: insights from network modeling using a prospective study across seven timepoints

Class number

Max LL

AIC

SSBIC

Entropy

BLRT p

% smallest class

Psychopathological symptoms (linear slopes)

1

−26879.34

53782.69

53806.38

–

–

–

2

−25994.30

52024.60

52060.13

0.76

<0.001

33.6%

3

−25776.24

61600.48

51647.86

0.73

<0.001

20.9%

4*

−25706.15

51472.31

51531.53

0.69

<0.001

16.7%

5

−25690.03

51452.06

51523.13

0.73

0.267

2.0%

COVID-19-related rumination (linear slopes)

1

−23634.14

47290.27

47311.99

–

–

–

2

−23276.22

46586.44

46674.00

0.77

<0.001

47.7%

3

−23076.20

46198.40

46243.80

0.73

<0.001

12.2%

4*

−23004.74

46067.48

46124.73

0.72

<0.001

10.4%

5a

−22981.87

46033.75

46102.85

0.64

<0.001

10.3%

6a

−22968.30

46018.60

46099.54

0.61

0.118

10.3%

Stress-related growth (quadratic and linear slopes)

1

−20979.81

41987.63

42015.27

–

–

–

2*

−20807.57

41663.14

41710.52

0.45

<0.001

17.7%

3

−20786.94

41641.87

41709.00

0.47

0.667

8.7%

4

−20807.57

41703.14

41790.00

0.73

1.00

<1%

5a

−20703.38

41514.75

41621.36

0.64

<0.001

2.0%

  1. % smallest class = percentage of respondents in the smallest class.
  2. Max LL maximized log likelihood value, AIC Akaike information criteria, SSBIC sample-size-adjusted Bayesian information criterion, BLRT parametric bootstrapped likelihood ratio test.
  3. *Selected model.
  4. aProblems with model convergence.
Back to article page