Skip to main content

Table 1 Model fit indices for growth mixture models

From: Supervision trajectories of male juvenile offenders: growth mixture modeling on SAVRY risk assessments

Model

BIC

LMR-LRT

Entropy

Smallest class %/(n)

Antisocial behaviour

 1

− 42,552.08

n/a

n/a

n/a

 2

− 43,352.39

p < .001

.91

.039 (202)

 3

− 44,042.78

p < .001

.80

.040 (207)

 4

− 44,209.88

p = .006

.84

.005 (28)

 5

− 44,286.48

p = .33

.83

.006 (29)

Family functioning

 1

− 50,973.55

n/a

n/a

n/a

 2

− 52,575.68

p < .001

.87

.209 (1088)

 3

− 53,646.03

p < .001

.89

.033 (170)

 4

− 54,171.66

p = .16

.87

.027 (140)

 5

− 54,506.86

p = .20

.90

.021 (111)

Personality

 1

− 29,962.82

n/a

n/a

n/a

 2

− 30,254.27

p < .001

.61

.318 (1653)

 3

− 30,519.34

p < .001

.74

.157 (818)

 4

− 30,721.70

p < .001

.69

.043 (222)

 5

− 30,775.80

p = .081

.70

.015 (78)

Social support

 1

− 24,159.37

n/a

n/a

n/a

 2

− 24,714.14

p < .001

.72

.337 (1755)

 3

− 25,136.04

p < .001

.80

.029 (151)

 4

− 25,733.02

p < .001

.82

.053 (278)

 5

− 25,913.54

p = .240

.82

.021 (111)

Treatability

 1

− 18,466.19

n/a

n/a

n/a

 2

− 18,845.24

p < .001

.65

.416 (2165)

 3

− 18,996.01

p < .001

.67

.090 (468)

 4

− 19,470.22

p < .001

.75

.049 (256)

 5

− 19,559.41

p = .240

.76

.022 (114)

  1. Entropy and LMR-LRT are not applicable for the 1-class model. The values in italics indicate the selected model
  2. BIC Bayesian Information Criterion, LMR-LRT Lo-Mendell-Rubin adjusted likelihood ratio test, Entropy average quality of classification