Table 4.
Multilevel binomial logistic regression (with logit link) results predicting likelihood to commit positive ERBs in Experiment 2.
| Predictors | Odds ratios | CI | p |
|---|---|---|---|
| (Intercept) | 0.72 | 0.60–0.86 | < 0.001 |
| Non-environmental news group | 0.90 | 0.71–1.16 | 0.429 |
| Negative environmental news group | 1.02 | 0.80–1.31 | 0.855 |
| Positive trait affect | 1.45 | 1.24–1.69 | < 0.001 |
| Non-environmental news group × positive trait affect | 0.75 | 0.60–0.95 | 0.017 |
| Negative environmental news group × positive trait affect | 0.67 | 0.54–0.85 | 0.001 |
| Age | 1.01 | 1.00–1.02 | 0.066 |
| Social desirability scale | 1.03 | 1.00–1.06 | 0.022 |
| Gender | 0.88 | 0.79–0.97 | 0.013 |
| Biospheric values | 1.12 | 0.99–1.26 | 0.070 |
| Egoistic values | 1.02 | 0.91–1.15 | 0.739 |
| Random effects | |||
| σ2 | 3.29 | ||
| τ00 pcpID | 0.17 | ||
| ICC | 0.05 | ||
| NpcpID | 328 | ||
| Observations | 2190 | ||
| Marginal R2 | 0.040 | ||
| Conditional R2 | 0.088 | ||
The grouping variables are effects coded such that the positive environmental news group represents the “baseline”. For comparative purposes, a similar model without the covariates, and without the interactions, is shown in Supplementary Table S4.
ERBS Environmentally relevant behaviors.