**4. Methods**

Key informant interviews were conducted in each of these four states during 2012 (**Table 1**). Interviews were administered to individuals knowledgeable about WUI issues and local affairs; moreover, these people were broadly representative of private forest landowners (PFLs), forest industry, government, local political and social factions, and social status [51]. We conducted 33 interviews in Georgia, 35 in New Mexico, 35 in Minnesota, and 45 in Oregon for a total of 148. Initial key informants were identified using local directories and Internet sources. To ensure


**Table 1.** *Key informant types (n = 148).* comparability across urban and rural counties of each state, informants representing each of the following perspectives were interviewed in each community: (1) federal and state land manager; (2) extension agent; (3) local planner and/or natural resource manager; (4) emergency services professional; (5) elected official; (6) business leader; (7) landowner; (8) religious leader; (9) journalist; (10) consultant or industrial forester; (11) environmental activist; and (12) citizen activist. Additional informants were identified using snowball sampling with purposive selection to encourage diverse perspectives, including an underrepresented or marginalized segment of local society [52].

Each interview covered (1) awareness of past and proposed fuel treatments in the area; (2) range of values associated with the WUI; (3) perceptions of wildfire risk; (4) public response to wildfire risk and occurrence; and (5) constraints on implementing wildfire reduction treatments. Open-ended questions encouraged informants to volunteer information, rather than simply responded to queries. Their rich and spontaneous replies provided a reality view of a place, including broad relationship patterns among actions and actors with the local environment [53].

Interviewers' notes were analyzed for emergent themes using a two-step coding process involving reading the notes and then coding into thematic categories [54]. Themes were compared within and across cases in each state and then over the four states. Each author reviewed the data and added additional interpretation to improve reliability.
