Skip to main content

Table 4 Path analysis results: estimations by generalized least squares (GLS)*

From: Confirmatory factor analytical study of the WHOQOL-Bref: experience with Sudanese general population and psychiatric samples

Structural fit indices

Using 6F & 5F models from 623 general population subjects

Applying WHO 4- & 6-domain models on

 

General population subjects

Psychiatric patients

Family caregivers

Gen population Subjects

Psychiatric pts

Family caregivers

 

6F model

5F model

6F model

5F model

6F model

5F model

4d model

6d model

4d model

6d model

4d model

6d model

No of parameters

16

15

15

16

16

15

12

21

12

21

12

21

Discrepancy (X2)

62.42

9.26

82.69

63.19

63.24

46.08

11.82

121.51

32.12

95.72

36.27

96.28

DF

12

6

12

6

12

6

3

7

3

7

3

7

Discrepancy/DF

5.2

1.54

6.89

8.7

5.27

7.68

3.94

17.36

10.71

13.67

12.09

13.76

GFI

0.97

0.99

0.88

0.94

0.94

0.95

0.99

0.94

0.96

0.91

0.95

0.91

Adjusted GFI

0.93

0.98

0.72

0.79

0.86

0.82

0.96

0.78

0.78

0.63

0.76

0.63

AIC

94.4

39.3

114.7

82.5

95.2

76.1

35.8

163.5

56.1

137.7

60.3

138.3

RMSEA

0.08

0.03

0.14

0.16

0.12

0.15

0.07

0.16

0.18

0.21

0.22

0.21

  1. NOTES: DF = degrees of freedom; 6F mod = 6-factor model; 5F mod = 5-factor model; 4d mod = 4-domain model; 6d mod = 6-domain model; GFI = goodness of fit index; RMSEA = root mean square error; AIC = Akaike index; Gen popn = general population
  2. * Ideal fit indices are: Discrepancy/Df ≤ 5; GFI, Adjusted GFI ≥ 0.9; RMSEA = 0.05 – 0.08; lower AIC