Table 1
Fit indices for multi-group measurement invariance models across studies.
| TESTED MODELS | CHARACTERISTICS | χ2 | DF | CFI | RMSEA |
|---|---|---|---|---|---|
| Study 1 | |||||
| Configural | Unconstrained model | 544.287 | 285 | .937 | .039 |
| Metric | Equal factor loadings | 641.581 | 313 | .918 | .042 |
| Scalar | Equal intercepts | 944.296 | 345 | .854 | .053 |
| Partial scalar | Partial equal intercepts | 720.014 | 321 | .903 | .045 |
| Study 2 | |||||
| Configural | Unconstrained model | 268.609 | 156 | .959 | .031 |
| Metric | Equal factor loadings | 336.629 | 178 | .942 | .035 |
| Scalar | Equal intercepts | 868.159 | 204 | .758 | .067 |
| Partial scalar | Partial equal intercepts | 403.856 | 186 | .920 | .040 |
Table 2
Descriptive statistics and zero-order correlations for self-objectification, body surveillance, and body shame across studies and countries.
| BODY SURVEILLANCE | BODY SHAME | MEAN (SD) | ||
|---|---|---|---|---|
| Belgium(Study 1) | LSOQ | .39** | .21** | –0.48 (1.72) |
| Body Surveillance | – | .39** | 4.77 (1.05) | |
| Body Shame | – | – | 3.00 (1.42) | |
| UK(Study 1) | LSOQ | .52** | .41** | .50 (1.94) |
| Body Surveillance | – | .59** | 5.14 (1.11) | |
| Body Shame | – | – | 4.32 (1.29) | |
| US(Study 1) | LSOQ | .63** | .44** | 0.17 (2.21) |
| Body Surveillance | – | .46** | 4.96 (1.17) | |
| Body Shame | – | – | 3.61 (1.17) | |
| Belgium(Study 2) | SOQ | .50** | .30** | 1.38 (13.71) |
| Body Surveillance | – | .53** | 4.77 (1.02) | |
| Body Shame | – | – | 3.09 (1.15) | |
| Israel(Study 2) | SOQ | .15** | .08 | –1.89 (8.82) |
| Body Surveillance | – | .46** | 4.83 (1.07) | |
| Body Shame | – | – | 3.22 (1.21) | |
| Thailand(Study 2) | SOQ | .31** | .19** | –5.41 (10.86) |
| Body Surveillance | – | .24** | 4.00 (0.98) | |
| Body Shame | – | – | 3.44 (0.86) |
[i] Note: **p < .01. LSOQ = Likert Version of the Self-Objectification Questionnaire. SOQ = the Self-Objectification Questionnaire. For descriptive purposes, means and correlations are reported using manifest scores computed from the full original item sets of each scale. Inferential analyses were conducted using latent variables as specified in the SEM models.
Figure 1
Standardized regression coefficients for the relationship between self-objectification and body shame as mediated by body surveillance among Belgian (a), US (b), and UK (c) women (Study 1).
Figure 2
Standardized regression coefficients for the relationship between self-objectification and body shame as mediated by body surveillance among Belgian (a), Israeli (b), and Thai (c) women (Study 2).