Supporters of Measure B Rely on Faulty Data Analysis

After viewing the Fire Chief’s Pro Measure B slide presentation at the Grass Valley City Council meeting, I brought to his attention that his yearly graphs illustrating total incident calls (Fire + Emergency/EMS) were being misinterpreted and are misleading.

The City’s analysis does not account for a downward trend pre-Covid, contracted incident calls from Nevada City, the error rate between the old and new system of data tracking, and the anomalous data during Covid years.

The City is using faulty data analysis to back up support of Measure B (proposed sales tax increase for Grass Valley, which appears on the 2024 Primary ballot), and this faulty analysis is misleading the public into believing that this sales tax measure is needed when it is not.

Using one-time spikes in trend analysis is misleading

First, the use of yearly data points (2012-2022), instead of standard six-year moving averages to identify trends, can over exaggerate the optics of the graph as we find during the Covid years (191% greater difference, see Graph 1)

When looking at trends, moving average calculations are used to reduce the exaggerated effects that individual data points on graphs can display which is critical when making long term costly decisions. The reason to use a moving average with time series data is that it shows longer-term trends or cycles without being misled by short-term, atypical spikes. The presentation graphs show very high and unusually steep upward trends (64%) of incident calls during the Covid years. My reanalysis using moving averages showed significantly fewer upward trends (22%) during Covid, indicating this specific time period may be an outlier and additional years of follow up and data collection post Covid is warranted to confirm the direction of the trend.

Second, in years 2020 to 2022 (Covid years), the increase in incidents (64% increase, see Graph 1) were attributed to the following - without any financial details or data to justify placing Measure B on the ballot:

New Grass Valley/Nevada City contract in December 2020
New and more efficient data collection software in 2020 (Image Trend)
Increase in yearly service demand

Note: The increase in EMS calls due to Covid was not considered since this information is not available to the department.

For four ‘normal’ years, the number of calls actually decreased

Third, years 2016-2020 (pre-Covid), which show a declining trend in calls, were left out of the presentation. These years reflect an 11% drop in calls; see Graph 2.

Also, in the pre-Covid years of 2018-2020, the trend line was showing a downward trend in calls. It would lead one to predict that this trend could either continue down slightly or show a slight upward trend after the pandemic ends. More time is needed to evaluate this trend.

The increase is partly due to a new contract with Nevada City

The City’s graph also deceptively includes incidents from a new contract for services between Grass Valley and Nevada City. This contract does result in an increase in the number of calls but what you may not know is the contract specifies Grass Valley receives nearly $900,000 per year with an additional max 3% increase in funding yearly from Nevada City. Since the city is receiving money for services, these incident calls should not be included in the data. The money obtained from Nevada City should include enough money to cover the extra calls and manpower.

Data between years aren’t truly comparable due to change in tracking software

The other reason stated for the high spike in calls after 2020 is the change in tracking software. The new system is more efficient because it directly enters data into the CAD system. This is in contrast to having dispatch entering by hand. Although no one disputes that hand entry is not as efficient as auto entry, no information was given as to the error rate of entry. Due to the difference between data entry methods, this is a classic case of comparing apples to oranges.

Lastly, the City claims that service demands increase yearly placing increased workload on the agency, its personnel and service delivery. Again, these numbers were never made public. We need more transparency before we vote for an increase in sales tax that will cost us (or drain money from our wallets) for years to come.

We won’t be misled by faulty data analysis

In summary, to ignore the downward trend pre-Covid, to include the contracted incident calls from Nevada City, not to disclose the error rate between the old and new system of data tracking, and not to consider a short-term anomalous event (Covid) starting in 2020 indicates to me that having the fire department analyze, interpret, and present important data to recommend higher taxes without the guidance of data experts will lead to false conclusions and the wrong decisions.

Businesses and residents have already suffered financially through the Covid lockdowns, so it is critical that business leaders and voters get this right. I believe the City needs to present all the data to the voting public and follow the trend for several more years before proceeding forward.