Recruitment Procedure
We adopt the definition of local “first responder” from Section 2 of the Homeland Security Act of 2002 (6 U.S.C. § 101) for the recruitment criteria. We will target a pool of approximately 10,000 first responders nationwide in this study. Assuming a 20-30% response rate typical of management research, which should provide approximately 2,000 - 3,000 matched responses.
Because a majority of fire stations are comprised of either 4- or 8-person shifts depending on the number and type of apparatuses at the station, we will need a minimum of 3 respondents from each station and shift (i.e., one crew) to perform the analysis. It is our hope that 3,000 matched responses will result in approximately 200 unique crews.
With appropriate funding, our research team hopes to engage in future stratified sampling based on the organizational types (public safety, fire, emergency response, and emergency medical) and a mix of career, volunteer, and combination departments serving rural and urban areas. Our aim would be to balance data collection across geographic-economic regions, between rural and urban areas, and between career and volunteer organizations.
Women and minorities are often underrepresented in fire and other first-response services. To properly address this issue, we plan to oversample both groups in these steps:
1. Identify the participating organizations with higher portions of both groups than average
2. Open the survey only to both groups until the number of respondents reach a preset level
3. Open the survey to the entire organization (Vaughan, 2017)
A random oversampling technique may be adopted during the data analysis if deemed necessary to increase the weights of each group.