Table 1. Path coefficients between experimental variables and the model mediator.
| Path Estimates | Processing Fluency B (SE) | 95% Confidence Interval (Lower Limit, Upper Limit) |
|---|---|---|
| Constant (COVID-19 referent) | 5.62 (0.17) | 5.28, 5.97 |
| Jargon (a1) | 0.11 (0.22) | -0.31, 0.54 |
| Flood Topic (a21) | -0.29 (0.22) | -0.71, 0.14 |
| Policy Topic (a22) | 0.04 (0.22) | -0.39, 0.47 |
| Jargon x Flood (a31) | -0.59 (0.31) | -1.19, 0.02 |
| Jargon x Policy (a32) | -0.85 (0.31)** | -1.46, -0.24 |
| Sample | -0.67 (0.13)*** | -0.92, -0.41 |
| F | 9.50*** | |
| R2 | .13 | |
| Conditional Effects | ||
| Covid-19 | 0.11 (0.22) | -0.31, 0.54 |
| Flood Risk | -0.47 (0.22)* | -0.90, -0.04 |
| Emergency Policy | -0.74 (0.22)*** | -1.17, -0.30 |
All (a) paths estimated with 10,000 Bootstrapped resamples from Hayes’ (2013) PROCESS Model 7. These estimates come from the model predicting motivated resistance to persuasion. All models yield slightly different estimates due to Bootstrapping. The sample covariate represents whether participants were compensated $0.80 (0) or $2.00 (1).
*p < .05
**p < .01
***p < .001.