Research on Causal Attribution of Wildfire: An Exploratory Multiple-Methods Approach

Abstract

Although studies show that actions by property owners, such as maintaining a defensible space, are generally the best means of protecting property from wildfire, victims often blame government agencies and others for property damage, injury, and death. This article describes a multiple-methods approach for investigating factors that influence how people who experience wildfire perceive the cause of wildfire damage. Phase I and II mail surveys and real-time field interviews were conducted in communities on the western slope of the Sierra Nevada. Generally speaking, people who had experienced wildfire attributed damage to other people's actions more than people who had not. Whether residents incurred damage or not, having maintained a sense of control or interacting with firefighters also appears to have influenced attributions. We argue that multiple-methods approaches to such questions have the potential to reveal more about such phenomena than approaches based on any single method.

Keywords:

• causal attribution
• multiple methods
• perceptional difference
• wildfire

Notes

1 It is important to note that in theoretical sampling, one attempts to include representatives from all the substantively important categories of, in this case, affected community residents. It is also important to note that categories themselves are not necessarily clear at the beginning. In this case, the field-worker began by sampling all three households that were actually burned out (clearly a relevant category), a number of households that were relatively near the fire damage, and a number that were further away from actual damage. The field-worker also paid attention to gender, age, ethnicity, and length of residence in the community, all of which potentially could have produced different responses. He discontinued his fieldwork only when novel responses to key questions ceased to be forthcoming and, after consulting with key informants in the community concerning the diversity of his sample, was satisfied the relevant categories of affected people in the two communities had been covered.