Skip to main content
Dryad

Data from the article: Narrative Risk Communication as a Lingua Franca for Environmental Hazard Preparation

Cite this dataset

Raile, Eric D. et al. (2021). Data from the article: Narrative Risk Communication as a Lingua Franca for Environmental Hazard Preparation [Dataset]. Dryad. https://doi.org/10.5061/dryad.47d7wm3dw

Abstract

Incorporating narrative elements into risk communication may encourage preparation for environmental hazards in ways that scientific language alone does not. We integrate narrative theory, narrative persuasion, and risk theories into a Narrative Risk Communication Framework and then assess the effectiveness of character selection as a narrative mechanism in scientific risk communication as compared to conventional science messaging alone. We utilize a survey experiment with residents along the flood-prone Yellowstone River in Montana and analyze the resulting data with a parallel and serial mediation statistical model. We find that positive affective response mediates the influence of narratives featuring hero character language. Positive affective response appears to overcome the risk perception paradox both by circumventing rational analysis of risk and by shaping risk perception. Overall, the results suggest that inspirational hero language is superior to language of fear or victimization in encouraging preparation – an important lesson for practitioners working to help citizens prepare for environmental disasters.

Methods

The dataset resulted from a survey experiment conducted with residents who lived within 800 meters of the Yellowstone River in Montana. The survey yielded 3,320 responses out of 8,721 mailings, for a response rate of 38.1%. After cleaning to remove addresses that did not contain a residence or for which the respondent was not an owner, trustee, or administrator of the property, 2,901 responses remained as valid. The primary data analysis involved a parallel and serial mediation model. Data and instructions are supplied to replicate that model. The number of observations used in the modeling, after eliminating missing data, is 1,938.

Usage notes

Observations with missing data have been removed from the comma-separated data file (Lingua_Franca_Dataset_REPLICATION.csv) for replication purposes. A codebook (Lingua_Franca_Codebook_REPLICATION.docx) and instructions for running the parallel and serial mediation model (Lingua_Franca_Instructions_REPLICATION.docx) are supplied along with the materials. The materials also include a more comprehensive ReadMe file (LINGUA_FRANCA_Readme.txt).

Funding

National Science Foundation, Award: 1635885

National Institute of Food and Agriculture, Award: Hatch Project Number 1015745

National Institute of General Medical Sciences

National Science Foundation, Award: EPS-1101342; OIA-1443108; OIA-1757351