Table 3.
Effects of country on first-order norms latent variables, second-order norms latent variables and observed indicators, standardized regression coefficients.
| Latent variable/Observed indicator | Baseline MIMIC model | DIF corrected model | ||
|---|---|---|---|---|
| Β (SE) | p value | Β (SE) | p value | |
| Second-order latent variables | ||||
| Anti-Smoking/Vaping Norms | 0.30 (0.09) | < 0.001 | – | – |
| First-order latent variables | ||||
| Expt. Inj. Norms | − 0.21 (0.08) | 0.007 | − 0.11 (0.07) | 0.15 |
| Sur. Inj. Norms | − 0.21 (0.07) | 0.004 | − 0.35 (0.08) | < 0.001 |
| Expt. Desc. Norms | − 0.31 (0.07) | < 0.001 | − 0.31 (0.07) | < 0.001 |
| Sur. Desc. Norms | 0.30 (0.12) | 0.008 | 0.03 (0.09) | 0.75 |
| Indicatorsa | ||||
| P2S2 | – | – | − 0.10 (0.01) | < 0.001 |
| P2S5 | – | – | − 0.14 (0.02) | < 0.001 |
| IN1 | – | – | 0.15 (0.03) | < 0.001 |
| IN4 | – | – | 0.15 (0.04) | 0.001 |
| DN2 | – | – | 0.38 (0.06) | < 0.001 |
| DN3 | – | – | 0.22 (0.07) | 0.001 |
MIMIC multiple indicators multiple causes, DIF differential item functioning.
aControlling for country differences on the underlying first-order latent variable (0 = Northern Ireland, 1 = Colombia).