By Matt Marietta
Fire departments have many expectations they are trying to meet, including job performance requirements for every role, and turnout and response times benchmarked through national standards, such as NFPA 1710. These standards are backed up by numerous studies into fire behavior in modern structures and analyses of how many firefighters it takes to conduct various fireground operations (not to mention years of practical experience of fire departments across the nation).
With fierce budgetary competition from other services (e.g., law enforcement, roads, parks), we need to justify every new station request and every apparatus, to say nothing of full-time employees. On top of this, simply providing run numbers and average response time is not enough in many jurisdictions.
Government budgets have multiple competing priorities. While public safety is the core function of any government, there are a lot of other areas that impact a community’s long-term health that are also screaming for attention. Whether we believe it or not, functions such as community development and even parks have an impact on a community’s long-term viability and, therefore, have a seat at the financial table with us. We need to be prepared to justify our needs with something more than a good community reputation (although that definitely helps).
Statistical support
So how do you statistically support proposed outcomes for programs and equipment? It is clearly no longer enough to rely on the fact that everyone loves firefighters or to issue dire warnings that people will die if the new $1 million ladder truck isn’t funded. This is the age-old problem that departments face when trying to quantify how many houses didn’t burn down. Harder still is to quantify how many fires won’t occur in the future because of a mitigation effort, but this is the type of justification driving many budgets now. When city managers or mayors take a strong scientific management approach, we need to be prepared to put together a good, quasi-predictive argument for outcomes based on budget expenditures.
In addition to always ensuring you are getting complete and accurate information from the firefighters in the field, fire service managers must remember that the larger the data set (e.g. the more runs or inspections you have in your analysis), the more likely you are going to get generalizable results.
For example, if you only run five full alarms in one year, then one extra-long or extra-short response time could completely throw off the resulting average or benchmark. If you run 500 full alarms, then outliers will not have a significant impact on the response time analysis, or on the ability of the fire chief to say something more definitive about the department and their response profile both now and in the future.
Use anecdotal evidence to support your figures
Sometimes, we have no choice but to run with small data sets. When you are limited by your available data, make sure to let your audience know. I usually include this as a note on anything released for public consumption or as a justification for a new initiative or capital budget request.
To turn smaller numbers into a measurable outcome for the community in a smaller department, it may be better to rely on a combination of traditional response data and well-reasoned connections to scientifically-based consensus standards (like NFPA 1710), as well as the anecdotal evidence. In statistical terms, the limitation of the quantitative data needs to be supplemented by qualitative analysis.
Lack of available data and shrinking budgets can impose limitations on departments. Marry traditional stories of outcomes – especially where there are several repeated similar outcomes – to changes in deployment or policy, or to specific output data. This will make it easier to identify small trends and potentially extrapolate to the broader fire safety context in the community served.
Put simply, smaller departments can overcome limited data by combining flashy life-saving stories with a thorough explanation of the output data and the demands of the community. Showing a deep understanding of the data will support an anecdotal story application being used to address a specific safety problem or budget expenditure.