Table 2.
Negative Binomial Regression Results Predicting Government and EM Agencies’ Facebook Posts Likes, Comments, and Shares during Hurricane Harvey (N = 1,360).
| Message Structure-Level Features | # of Likes | # of Comments | # of Shares |
|---|---|---|---|
| (Intercept) | 4.05 (0.23) *** | 2.13 (0.30) *** | 5.85 (0.29) *** |
| Information specificity | |||
| Word Count | 0.00 (0.00) | 0.00 (0.00) *** | 0.00 (0.00) *** |
| Time markers | −0.31(0.08) *** | −0.13 (0.10) | 0.06 (0.10) |
| Location markers | 0.21 (0.08) ** | 0.00 (0.10) | −0.06 (0.10) |
| Media Richness | |||
| Text-only | −0.59 (0.15) *** | −0.84 (0.19) *** | −1.31 (0.18) *** |
| URLs | −1.28 (.015) *** | −1.89 (0.19) *** | −1.87 (0.19) *** |
| Photos and other visual | −0.48 (0.13) *** | −1.10 (0.17) *** | −1.00 (0.17) *** |
| Videos (reference group) | |||
| Context-Specific Topical Features | |||
| Risk forecasting and updates | −0.46 (0.10) *** | −0.40 (0.13) ** | −0.14 (0.13) |
| Correcting | 0.74 (0.17) *** | 0.72 (0.22) *** | 0.91 (0.21) *** |
| Confirming | 0.75 (0.09) *** | 0.94 (0.11) *** | 0.62 (0.11) *** |
| Connecting | 0.05 (0.09) | 0.38 (0.12) *** | 0.03 (0.11) |
| Assisting Relief | −0.22 (.12) | 0.15 (0.15) | 0.41 (0.14) ** |
| Growing | 0.59 (0.10) *** | 0.15 (0.13) | −0.42 (0.13) ** |
| Storytelling | 0.16 (0.28) | 0.90 (0.36) * | 0.56 (0.35) |
| Linguistic Features | |||
| Dialogic loop | 0.09 (0.10) | −0.10 (0.12) | 0.56 (0.12) *** |
| Empathic tone-positive | 0.04 (0.01) *** | 0.02 (0.01) | 0.04 (0.01) ** |
| Empathic tone-negative | 0.01 (0.01) | 0.01 (0.02) | - 0.02 (0.02) |
| Analytic tone | −0.00 (0.00) *** | 0.00 (0.00) | −0.01 (0.00) ** |
| Informal tone | 0.05 (0.02) * | 0.05 (0.02) * | 0.01 (0.02) |
| Control Variables | |||
| Pre-disaster | 0.53 (0.15) *** | 1.26 (0.18) *** | 0.48 (0.18) ** |
| During disaster | 0.64 (0.09) *** | 0.93 (0.11) *** | 0.56 (0.11) *** |
| Post disaster (reference group) | |||
Note: *** p < .001, ** p < .01, *p < .05.