Kozak Says Collins’ Firing Decision Was Based on Inaccurate Data
Cheyenne Police Chief Brian Kozak says Mayor-elect Patrick Collins' decision not to retain him due to crime numbers was based on inaccurate data caused by a computer software failure.
The police department issued the following news release Friday afternoon:
Through media reports, the Cheyenne Police Department (CPD) learned Mayor-Elect Collins would not reappoint Chief Kozak due to low index crime clearance rates from 2015-2020. Since crime is a sociological phenomenon influenced by a variety of factors, the FBI strongly discourages using their data to rank locations or make comparisons as a way of measuring law enforcement effectiveness.
In theory, the clearance rate is obtained by calculating the percentage of arrests for the total number of index crimes. Index crimes include murder, rape, aggravated assault, burglary, auto theft and shoplifting.
“Mr. Collins obtained the information from the Wyoming Division of Criminal Investigations (DCI) website for the FBI Uniform Crime Reports (UCR) and did not confirm the data with CPD,” said Chief Kozak. “In 2016, CPD realized the UCR data for crime clearances was inaccurate. We determined a records management software update on April 20, 2015 caused an error, preventing proper data transfer.”
Kozak shared that the computer vendor was unwilling to repair the error because the FBI was phasing out UCR, due to its unreliability. The FBI and CPD will be implementing a more reliable National Incident-Based Reporting System (NIBRS) system in January 2021, which will eliminate future computer errors.
In response to the computer failure, CPD has maintained a manual accounting of accurate clearance rates from 2015-2020:
The updated department statistics reveal that Cheyenne police officers, in fact, have an index crime clearance rate higher than state average.
Chief Kozak said, “Once again, the data reveals that our officers are among the most efficient crime fighters in the United States. We should never rely strictly on data, but should evaluate three areas when deciding if a police force is performing well: Does the community feel safe? Does the community support the police department? Do the officers have good morale?”
Since the results of UCR data can be deceiving and inaccurate, the FBI strongly warns against using these statistics to measure law enforcement effectiveness. For more information about the pitfalls of ranking with UCR data, see the FBI warning brochure, ucr-statistics-their-proper-use (fbi.gov).
The FBI warns that UCR data may not be comparable to previous years because of differing levels of participation over time. For example, in 2015 CPD experienced a sharp increase in shoplifting when three box stores opened. Over the last several years, the department has worked hard to decrease that rate, to one that is now the lowest in decades. In another example, the community saw a spike in aggravated assaults in 2017 after the Department trained its officers to investigate strangulation domestic violence crimes. Additionally, in 2016, the agency reported 19 robberies to DCI, who recorded by typo, 115 robberies for the year. DCI never corrected the error.
Another major flaw with UCR data is that it only records a snapshot of a particular moment. Investigations change the status of a crime, and the crime may be cleared months later, which is never reflected with UCR. As an example, Mr. Collins said the most angering and shocking statistic was the city’s rape statistic, citing 2019 UCR statistics which list 43 rapes with only 2 clearances. The actual status of rape cases in 2019 are listed below:
Actual 2019 UCR forcible rapes = 38
Arrest 2
Sent to prosecutor for prosecution 3
Transferred to an outside agency 4
Unfounded (false report) 12
Cleared by exception 1
Inactive (no evidence to proceed) 12
Open and still under investigation 4
In an interview, Collins stated, “We were really doing a good job of clearing crimes for the first few years the chief was here. But since then, we’ve had a precipitous drop in our crime statistics, and I just thought we should have a higher level of performance from our police department, and I think the police department follows the lead of their chief, and so I thought it was time to replace that person.”
In response, Chief Kozak said, “A good leader should always confirm data before making important decisions. Mr. Collins just insulted every employee in this agency who have dedicated their lives to help people. We must always remember that there are people behind statistics.”
This topic, again, highlights the need to honor the police department’s request for a crime analyst. Chief Kozak has requested the City Council to fund this position for several years.