Mayor Godfrey has made his effectiveness as a crimefighter a centerpiece of his mayoral re-election campaign. Yesterday, our household received a glossy mailing that touted his crimefighting skills, his fecundity and his ability to read bedtime stories to his daughter. All good qualities for a mayor, I suppose, but personally I find the only relevant or useful part of the mailing to be his claims regarding his crimefighting skills.
An earlier WCF article examined the problems with this claim, including his unprovoked mugging of the rules of logic and argument (the post hoc ergo propter hoc fallacy) and some chicanery with numbers.
In Mayor Godfrey’s latest fusillade against logic and reason, he finally reveals the “raw” data underlying his startling campaign claims: a 23 percent reduction in overall crimes from 1999 to 2006, and a 43 percent reduction in violent crimes over the same period.
Let’s set aside for a moment the questionable ethics of using taxpayer money to buy a full-page ad in the Ogden Standard-Examiner for what is essentially a campaign document. We’ll also not beat a dead horse on the post hoc logical fallacy, even though it still casts doubt on his claims of crimefighting prowess.
What I want to talk about today is the fact that he is guilty of depraved indifference to the truth. Either he has negligently failed to check the data underlying his claim, or he is a liar who fabricated numbers to make himself look good.
There are three sources of data which I’ve used for this analysis.
1) An undated report signed by Chief Greiner, apparently from 2004, available as an Acrobat version of a Powerpoint presentation on the Ogden City website. Click on the “12-Year Crime Report” link. (Greiner 2004).
2) The Federal Bureau of Investigation’s Uniform Crime Report, published annually with data breakdowns for all cities, states and regions (FBI UCR).
3) Mayor Godfrey’s personal advertising space, under the guise of an Ogden City communication with its citizens, published in the Thursday, August 30, 2007 Standard-Examiner (Godfrey 2007).
The Ogden City advertisement repeats exactly, almost verbatim, the claims contained in Mayor Godfrey’s campaign material:
“Over the last eight years, since Matthew Godfrey became mayor, Ogden has hired 18 new police officers and crime has dropped more than 23% — including a drop in violent crimes of 43%. That’s no coincidence.” [Campaign mailing, paid for by the People to Re-elect Matthew Godfrey]I’ve taken the FBI’s Uniform Crime Report statistics and converted them to the same basis used in the Greiner and Godfrey reports (crimes per 1,000 population): I divided by 100 the FBI metric (number of crimes per 100,000). I dropped the 2003 data from the Greiner report because it was marked “projected” and therefore I did not think that Chief Greiner should be held accountable for its veracity.
“According to statistics from the Ogden Police Department, the overall crime rate in Ogden has dropped by 23 percent in the last eight years. Non-violent crimes are down by 21 percent, and data on violent crimes proves even more significant.
“In a report calculating incidents based upon crimes per thousand population, violent crimes recorded in 1999 were 6.7 compared to 3.8 reported last year — in other words, violent crime in Ogden is down by 43 percent.
…“[There is a] remarkable drop in crime that is already evident in Ogden” [Ogden City advertisement, paid for by the People who May or May Not Want to Re-elect Matthew Godfrey But Who Must, By Law, Pay Taxes to Publish This Unadulterated Nonsense]
The Godfrey and Greiner numbers (red and blue) are in general agreement, except for a troubling increase in 1999 crime in Godfrey’s data which is not shown in Greiner’s data. The Godfrey and FBI UCR data are also in good general agreement, except for that same spike in 1999 which Godfrey claims (and which he uses as a baseline for his “23% drop” claim) but which is not reflected in the federal reports.
I’ve already discussed in my previous article why the apparent drop in total crime is not so startling, since total crimes in Utah dropped at an greater rate over the same period, probably due to demographic changes.
Here’s a comparison for the “remarkable” drop in violent crime during Mayor Godfrey’s administration:
Evidence of data manipulation is manifest here. The 1999 number is far higher than we’d expect from either the FBI Uniform Crime Report or from the Ogden City Police Department’s own published report, and the 2005 number is far lower. The 2002 spike in the FBI data for violent crimes is inexplicably absent in both Greiner’s and Godfrey’s data. The 2000, 2003 and 2004 data are in good general agreement, and the 2001 number is identical in all three reports. The 2002 spike is probably a statistical aberration, as is the 2001 dip, since they both lie away from the trend line. Still, Godfrey’s and Greiner’s data preserve the artificially low 2001 figure but “shave off” the artificially high 2002 figure.
As far as I’m aware, the Ogden City Police Department is legally obligated to provide accurate data to the FBI for their Uniform Crime Report. They seem to have done a good job doing so; there are minor discrepancies in the data which may be explained by methodological differences (such as different population estimates, different types of crimes included, and the like).
But how in the world can we possibly explain an artificially inflated total crime rate in 1999, before Mayor Godfrey was elected?
How can we possibly explain an artificially inflated violent crime rate in 1999, and an artificially low violent crime rate in 2002 and in 2005, the last year for which we can compare publicly available data? Isn’t it odd that the “43% drop in violent crime” claim simply reflects a comparison of two faulty datapoints (1999 and 2005)? How odd that the faulty data is in exactly the direction that makes Mayor Godfrey look more effective at reducing crime than he actually is.
(A full version, including city-by-city breakdown, of the 2006 FBI UCR data has not been released yet, so I did not include the Mayor’s claim for 2006 data of a rate of 3.8 violent crimes per 1000 population. His “43% drop” claim is apparently based on comparison of the bogus 1999 datapoint, 6.7 per 1000, versus the unverifiable 2006 datapoint, 3.8 per 1000. 3.8/6.7 is 57%, and 100% minus 57% is a drop of 43%.)
The real drop, using real, verifiable statistics, is 13%, as I’ve argued earlier. There is a lot of year-to-year fluctuation, so I took the average of the pre-Godfrey years for which FBI UCR data were available (1995-1999) and compared them to the average of the Godfrey years (2000-2005). I think it’s disingenuous to use only two favorable datapoints in making comparisons when you have yearly data for an eight-year mayoral administration.
To get a 43% drop in violent crime, the author of these statistics had to artificially inflate the 1999 number and artificially deflate the 2005 number.
Last year Mayor Godfrey explained to me that we were in a Republic, not a Democracy. This year, based on the letters generated by his toadies, it appears we are employees (or maybe stockholders, I’m not sure) of the Ogden City Corporation, which like all corporations is looking out for the commonweal, like water and sewer and parkland and police services.
It seems to me that the CEO of a $120 million corporation, a Captain of Industry, what Tom Wolfe would have called a Master of the Universe, would want to have accurate data in order to make his Important Corporate Decisions that we mere mortals could not possibly understand.
It seems to me that the CEO of a $120 million corporation would be tarred and feathered by either his board or the stockholders if he were found to have fabricated accounting data in order to make it appear the corporation was more successful than it actually was.
Was this done by Mayor Godfrey, or was it done by one of his employees?
Will the theft of lawn signs be included in next year’s statistics?
Update 9/1/07 11:09 a.m. MT: At the request of one of our readers, we have created a printer-friendly page for readers who would like a hard-copy printout of this article. The page can be found here. Simply navigate to the page, and send the document to your printer.