Household preparedness motivation in lahar hazard zones: assessing the adoption of preparedness behaviors among laypeople and response professionals in communities downstream from Mount Baker and Glacier Peak (USA) volcanoes

TL;DR

This survey of Skagit Valley residents and response professionals finds that perceived self-efficacy and ascription of responsibility, rather than perceived response-efficacy or protective response costs, correlate with household lahar preparedness. Professional hazard-management participation improves self-efficacy, confidence in officials, and information seeking but minimally improves household preparedness or hazard-map knowledge.

Key findings

• Few respondents endorsed proposed barriers: lack of hazard knowledge 35%, lack of preparedness knowledge 28%, cost 26%, time 23%; low response-efficacy (“items will not help me protect myself”) 7%, reliance on neighbors/community members 6%, reliance on emergency services 3%. Self-efficacy and responsibility correlated with preparedness: self responsibility ρ = 0.19*, local emergency services ρ = −0.14, FEMA ρ = −0.14, and “knowledge and skills to ensure that I am prepared for a natural hazard” ρ = 0.43*.
• Response professionals (n = 73) had higher average composite preparedness (CP) scores than laypeople (n = 383): 0.17 vs −0.03 (p = 0.036), driven by the action indicator (0.21 vs −0.03, p = 0.042); planning and supplies indicator differences were not statistically significant. More professionals had a family member who learned first aid (+26%), knew vulnerable community members (+26%), sought local volcanic hazard information (+21%), identified a non-local emergency contact (+13%), and installed a smoke detector (+12%).
• Only 37% of response professionals and 45% of laypeople correctly identified only lahars as impacting the Skagit Valley (not statistically significant); about 60% of both groups correctly said the lahar hazard decreases downstream. Response professionals expressed greater confidence in officials and higher self-efficacy, with agreement on self-efficacy statements increasing by an average of 21% (range 15–24%) among professionals.
• Overall, respondents had prepared eight to nine recommended items on average; the most commonly prepared items were first aid kit (81%), non-electric can opener (78%), blanket or sleeping bag (75%), smoke detector (74%), and flashlight with extra batteries (71%). For planning and information seeking, 40% had a family contact plan, 46% had an out-of-area emergency contact, 63% had someone in the family who knows first aid, and only 25% had sought local volcanic hazard information, while over three-quarters were interested in learning more.

Methods

• An anonymous, voluntary online questionnaire was administered between August and December 2014 using non-random convenience sampling; 507 individuals living or working in the Skagit Valley participated. Recruitment included postcards, phone calls and emails, a newspaper article, and social media posts.
• The questionnaire assessed hazard knowledge, risk perception, preparedness behaviors, barriers, ascription of responsibility, and self-efficacy. Preparedness was measured from six activities and fourteen items grouped into supplies, planning, and action indicators; raw scores were standardized to z-scores and averaged into an equally weighted composite preparedness (CP) score.
• Data were analyzed in R version 3.3.2 using chi-square tests, t-tests, and Spearman’s rho correlations. Likert-type responses used a 5-point scale, with 1–2 and 4–5 rankings combined for chi-square tests.

Limitations

• Stated in the paper: the convenience sample lacks randomness, may suffer selection bias, and cannot yield a response rate; the questionnaire was English-only, likely underrepresenting Spanish-speaking residents (5.1% of adults in Skagit County speak Spanish or Spanish Creole at home); the online platform limited participation by those without computer access or literacy, including elderly and lower-socioeconomic-status residents; respondents with household incomes below $50,000 were underrepresented by 7%.
• Stated in the paper: the “response professionals” category combines varied occupations, limiting subgroup conclusions; demographic data on response professionals in the Skagit Valley were lacking; convenience sampling may overestimate participation effects because more engaged individuals may respond. The study also did not distinguish perceived from actual barriers and used a coarse grouping of preparedness measures.
• Evident in the text: the design is cross-sectional and self-report, with no independent verification of preparedness behaviors; findings come from one valley and two volcanoes, and the professional subgroup is small (n = 73), limiting detailed occupational comparisons.