The questionnaire was composed of mostly closed-ended questions. Demographic questions pertained to age, gender, and marital status at two points in time; racial background; total household income for 2011; home or rental insurance; and property damage information and evacuation experience during the wildfire.

Received social support was measured by the Inventory of Postdisaster Social Support scale (Kaniasty & Norris, 2000; Norris, Murphy, Kaniasty, Perilla, & Ortis, 2001) . Participants were asked to estimate how often they received different types of social support within a clearly established time frame anchored between the October wildfires (Oct 21, 2007) and around New Year’s Day 2008. This support was related to the wildfires . It was modified to include 30 items to measure three types of support: emotional (three questions), informational (three questions), and tangible (four questions). Emotional social support assessed actions such as being comforted or shown signs of affection, someone expressing concern about your well-being , and knowing that others would be around if they were needed. Informational social support measured someone suggesting some action to take, understanding a situation you were in, and being given information on how to do something. Tangible social support referred to being given/loaned/offered money, furniture tools, food, and a place to stay. Three distinct sources of support were measured: family (ten items), friends, including neighbors and coworkers (ten items), outsiders, including people outside respondent’s immediate support circle such as community leaders , voluntary organizations, and professional service providers (ten items). These questions were scored on a four-option response set ((0) = never to (3) = many times). In the current study, there was very good total internal consistency with a Cronbach alpha coefficient of 0.95. Others have found internal consistency coefficient of 0.93 with 36 total items (Kaniasty & Norris, 2000) .

The Center for Epidemiologic Studies Depression Scale (CES-D; Radloff, 1977) was used to assess nonclinical depressive mood and symptoms and is intended for the general population. A 20-item scale was used and scored on a four-option response set ((0) = rarely or none of the time to (3) = most or all of the time). In the general population, a score of 16 or greater suggests a high level of depressive symptoms, and a score lower than 16 reflects lower levels of symptoms (Myers & Weissman, 1980) . In the current study, the Cronbach’s alpha coefficient was 0.85.

Researchers included a prompt at the end of the questionnaire asking for additional comments related to participants’ experiences after the wildfires . Narrative data were collected from 46 participants.

The snowball sampling technique was used to disperse the seven-page questionnaire during the months of May through July 2012. Undergraduate research assistants and I distributed the questionnaire to adult community residents who then distributed it to other community-residing adults and so forth. An information sheet was attached to each questionnaire. This information sheet invited community-residing adults to participate in the study if they lived in the area that was involved in the 2007 northern San Diego County wildfires and if they were at least 25 years old in October 2007. These adults were told that every resident’s experiences were wanted whether they experienced damage to their homes or apartments or had no damage and whether they had to evacuate or did not have to evacuate. They were told that the survey would take about 30 min to fill out and asked questions about social support encountered during and after the wildfires and about their experiences during the wildfires and in 2012.

Participants were not compensated for their time. They completed the questionnaire in their place of their choice and mailed it back to the main researcher in an attached addressed and stamped envelope. There was a response rate of 45.7 %.

When victim status groups and participants’ gender were compared with CES-D scores, no main effects were found, but there was an interaction, F(2, 176) = 3.93, p = 0.02. Specifically, women respondents’ CES-D scores declined as their exposure to the wildfire declined; men’s CES-D scores increased as their exposure to the wildfires declined. Table 2.3 presents CES-D mean and SD, gender, and victim status data.

Women (M = 9.58, SD = 8.83; range = 0–46) of all ages showed more depressive symptomology than the men (M = 7.92, SD = 6.21; range = 0–37); however, there was no significant effect between females and male participants, t(181) = - 1.49, p = 0.14 (two-tailed).

Although primary victims (M = 10.11, SD = 10.42, range = 0–46) reported more depressive symptomology during a week in 2012 than secondary victims (M = 8.01, SD = 6.79, range = 0–30) and non-victims (M = 9.65, SD = 7.66, range = 0–37), analysis of the three victim status groups and their mental health at 4.5 years after the wildfire yielded no significant effect, F(2, 163) = 1.51, p  = 0.22. After looking at these data, the large SD for the primary victim status group showed scores ranging from 0 to 46 with six participants’ scores being 16 and above and eight participants’ scores being 3 and below; therefore, it confirmed a wide range of values indicative of a large SD.