Skip to main content

Table 2 Model indices for confirmatory factor analyses, and measurement invariance across language, ethnicity and chronic conditions

From: Confirmatory factor analysis and measurement invariance of the English, Mandarin, and Malay versions of the SF-12v2 within a representative sample of the multi-ethnic Singapore population

 

WLSMV χ2 (df)

RMSEA

CFI

TLI

SRMR

DIFF TEST χ2 (df), p

ΔRMSEAa

ΔCFIa

ΔTLI

ΔSRMR

Confirmatory factor analyses

          

Model 1

Unidimensional

3994.848 (54)

0.109

0.940

0.926

0.116

–

–

–

–

–

Model 2

Two–factor model

2304.272 (53)

0.083

0.965

0.957

0.082

554.73 (1), p < 0.001

− 0.026

 + 0.025

 + 0.031

− 0.034

Model 3

Two-factor model with correlated residuals

618.043 (48)

0.044

0.991

0.988

0.044

939.031 (5), p < 0.001

− 0.039

0.026

 + 0.031

− 0.038

Confirmatory factor analyses by language subgroupsb

English language only

          

Model 2

Two-factor model

3652.579 (53)

0.118

0.951

0.939

0.078

–

–

–

–

–

Model 3

Two-factor model with correlated residuals

1210.750 (48)

0.070

0.984

0.978

0.046

1422.088 (5), p < 0.001

− 0.048

 + 0.033

+0.039

− 0.032

Mandarin language only

          

Model 2

Two-factor model

543.064 (53)

0.129

0.962

0.953

0.097

–

–

–

–

–

Model 3

Two-factor model with correlated residuals

206.138 (48)

0.077

0.988

0.983

0.057

208.193 (5), p < 0.001

− 0.052

 + 0.026

 + 0.03

− 0.04

Malay language

          

Model 2

Two-factor model

566.189 (53)

0.123

0.987

0.983

0.074

–

–

–

–

–

Model 3

Two-factor model with correlated residuals

310.606 (48)

0.092

0.993

0.991

0.052

169.298 (5), p < 0.001

− 0.025

 + 0.006

 + 0.008

− 0.022

Measurement invariance (MGCFA)

          

Language

          

Configural

895.138 (144)

0.051

0.991

0.988

0.046

–

–

–

  

Metric

879.544 (164)

0.046

0.991

0.99

0.046

54.333 (20), p < 0.001

− 0.005

0

  

Scalar

1178.003 (222)

0.046

0.988

0.99

0.05

378.713 (58), p < 0.001

0

− 0.003

  

Ethnicity

          

Configural

1473.219 (195)

0.065

0.991

0.988

0.048

–

–

–

  

Metric

1354.748 (225)

0.057

0.992

0.991

0.048

33.096 (30), p = 0.32

− 0.008

+0.001

  

Scalar

1398.609 (312) 

0.048

0.992

0.993

0.049

231.257 (87) , p < 0.001

− 0.01

0

  

Chronic physical conditions

          

Configural

680.066 (97)

0.044

0.991

0.987

0.047

–

–

–

  

Metric

654.824 (107)

0.041

0.991

0.989

0.048

39.467 (10), p < 0.001

− 0.003

0

  

Scalar

666.124 (136)

0.036

0.991

0.992

0.053

87.085 (29), p < 0.001

− 0.005

0

  
  1. aOnly change in RMSEA and CFI were evaluated for MGCFA
  2. bThe CFAs within the language subgroups were not weighted using survey weights
Back to article page

Health and Quality of Life Outcomes

ISSN: 1477-7525

Contact us