Skip to main content

Table 3 Testing wording effects for Attitudes Toward Obese Persons (ATOP) and Beliefs About Obese Persons (BAOP) using confirmatory factor analysis (N = 707)

From: Attitudes toward and beliefs about obese persons across Hong Kong and Taiwan: wording effects and measurement invariance

 

ATOP

BAOP

M1

M2

M3

M4

M1

M2

M3

M4

χ2 (df)

1040.72 (170)*

580.25 (163)*

211.38 (149)#

297.82 (150)*

51.97 (20)*

46.42 (18)*

11.32 (11)*

24.33 (12)*

CFI

0.799

0.904

0.986

0.966

0.958

0.963

1.000

0.984

TLI

0.776

0.888

0.982

0.957

0.941

0.942

0.999

0.962

RMSEA (90% CI)

0.086 (0.081, 0.091)

0.060 (0.055, 0.066)

0.024 (0.016, 0.032)

0.038 (0.031, 0.044)

0.048 (0.032, 0.064)

0.047 (0.031, 0.064)

0.006 (0.000, 0.040)

0.038 (0.015, 0.060)

SRMR

0.092

0.069

0.041

0.049

0.050

0.047

0.024

0.035

ECVI

1.601

0.963

0.476

0.597

0.119

0.117

0.087

0.103

  1. df degree of freedom, CFI Comparative fit index, TLI Tucker-Lewis index, RMSEA Root mean square error of approximation, SRMR Standardized root mean square residual, ECVI Expected cross-validation index
  2. *p < .001; #p = .001
  3. M1: One trait factor (ATOP or BAOP) model
  4. M2: One trait (ATOP or BAOP) and one method (negative wording) factors model
  5. M3: One trait (ATOP or BAOP) and two correlated-method (positive and negative wordings) factors model
  6. M4: One trait (ATOP or BAOP) and tow uncorrelated-method (positive and negative wordings) factors model