Friday, May 3, 2013

Consider the denominator

This is a follow up to my post Tuesday about the declining number of false alarms in Lincoln. A reader, Michelle, who is an alarm administrator for a Texas law enforcement agency, left a comment telling me that the industry standard for assessing false alarms in the ratio of false alarm dispatches to alarm systems. This makes sense. The denominator provides the context for comparison., just like population is used as the denominator to calculate crime rates.

There are 5,554 registered alarms in Lincoln, so last year's 2,383 false alarm calls would represent a ratio of .43. We do not know how many alarms were in operation in 2002, the year we reached our high of 4,848 false alarms. I think we can reasonably assume that there are significantly more today, because the population of Lincoln is more than 30,000 greater today, and it is also my impression that the number of alarmed premises has soared in recent years.

Lacking specific data on the number of registered alarm systems in 2002, if we just assume that the number of alarm systems was the same as in 2012 for the sake of this exercise, the ratio of false alarms to alarm systems would have been 4,848 to 5,554, or.87. If I am right about the likely increase in the number of alarmed locations in the past dozen years, that ratio would actually have been considerably higher than .87, thus the decrease in the ratio to .43 in 2012 would be even more dramatic.

From past comparisons, I already knew that false alarms in Lincoln are on the low side when population is used as the denominator.  I was curious, though, if the same would hold true using Michelle's ratio of alarms to alarm systems.  In her comment, she had mentioned the website of the False Alarm Reduction Association, faraonline.org, so I went there in search of any data of this type from other jurisdictions. It looks like the site has lots of resources, but most are for members only, and I wasn't quite ready to fork over $150 to get a peak right at the moment.

I did, however, find a document on FARA's site that sheds some light on how Lincoln's ratio of false alarm calls to alarms system stacks up.  It was this description of the winners of the 2011 FARA Achievement Awards.  The top award went the the Charlotte-Mecklenburg Police, which reduced the false alarm ratio from .42 to .35 over a three year period. One of the runner ups, the Riverside, CA Police, was recognized for a four year reduction from .69 to .41.  With Lincoln's ratio standing at .43, it appears we compare quite favorably with the award winners. 

7 comments:

Steve said...

Maybe I'm missing something, but you say the standard is the ratio of false alarms to the number of alarm systems. That would mean that for Lincoln, using your figures, the ratio was 2,383/5,554. On my calculator that comes to approximately .42, or nearly double what you are claiming. Even so, it is a significant improvement over the .87 or higher from years past.

Tom Casady said...

7:15,

Thanks. I was lost in my own notes. More coffee!

Steve said...

As a math mentor for LPS, it's my job to catch these kinds of mistakes. :)

Michelle said...

Texas? I can do a drawl if I need to. Although that might get ATB/PMAM a bit excited since I would be in their backyard and I write about their competitor in the alarm billing/outsourcing market.
It might be interesting to know if more alarm systems are registered as a result of having an alarm or because your citizens are compliant. If it is the former, your dispatch rate may still be artificially high as a result of unregistered alarm systems. Our dispatch rate was .32 last year and I register several alarm systems every day because they had a dispatch request. Because of this, I am sure there are hundreds of unregistered systems out there. By working with alarm companies to get accurate customer lists and educating the public about the need to register, I hope to gain a better picture with my numbers for 2013.

Anonymous said...

When I lived in Abel Hall at UNL in 1974 (the first year that women could live in the hall as previously it was an all-male residence hall), it seemed like we had a fire alarm almost every night. It got to the point were no one evacuated because it was just a waste of time. I was on the third floor and figured if it were a real fire, I could just jump, and if I died, no loss to humanity anyway. If you pulled out the fake alarms from UNL, I bet the number of false alarms would be even fewer.

Tom Casady said...

9:19,

This was a terrible problem during the 1970s and 80s at UNL, but no longer. Sometime back in the mid to late 80s, UNL got very, very tough on those caught pulling false alarms, and the problem was virtually eliminated.

Tom Casady said...

Michelle,

Your comments have been languishing in the city's spam filter, for some reason, but were liberated this morning! Regrets.