Skip to main content

Table 3 Goodness-of fit statistics for the tested models (n = 772)

From: Friendship networks and psychological well-being from late adolescence to young adulthood: a gender-specific structural equation modeling approach

 

Goodness-of-fit statistics

 

Model 1: Baselinea

Model 2: Social causationb

Model 3: Social selectionc

Model 4: Reciprocald

Males

 RMSEA

.016

.017

.018

.018

 CFI

.995

.994

.994

.993

 TLI

.992

.992

.991

.991

 AIC

12178.444

12179.927

12180.417

12181.908

 BIC

12381.108

12386.565

12387.055

12392.520

 χ2

73.00

72.49

72.99

72.47

  df

66

65

65

64

  p

<.001

<.001

<.001

<.001

 Chi-square difference test

  Comparison with:

-

Model 1

Model 1

Model 1/Model 2/Model 3

  Change in χ2

-

.51

.01

.53/.02/.52

  Change in df

-

1

1

2/1/1

   p

-

.47

.87

.76/.89/.47

Females

 RMSEA

.051

.052

.051

.051

 CFI

.949

.948

.950

.950

 TLI

.929

.928

.931

.929

 AIC

12115.777

12117.570

12114.797

12116.652

 BIC

12316.592

12322.322

12319.549

12325.342

 χ2

131.15

130.95

128.17

128.03

  df

66

65

65

64

  p

<.001

<.001

<.001

<.001

 Chi-square difference test

  Comparison with:

-

Model 1

Model 1

Model 1/Model 2/Model 3

  Change in χ2

-

.20

2.98

3.12/2.92/.14

  Change in df

-

1

1

2/1/1

   p

-

.65

.08

.20/.08/.70

  1. a Only auto-regressive associations and cross-sectional correlations
  2. b Friendship network quality at T1 predicts psychological well-being at T2
  3. c Psychological well-being at T1 predicts friendship network quality at T2
  4. d Friendship network quality and psychological well-being have reciprocal associations