Big Government Provides Some Data

The FBI usually released crime data in September. Researchers and policy wonks would pour over the data, looking for any hint of a chance for a policy intervention or investigating long-term trends. This year will be different. The FBI announced in 2015 that the underlying reporting system for crime data would change by 2021 but not all agencies nationwide are quite ready for the new system. Implementation remains a work in progress. 

The Uniform Crime Report was conceived in the 1920s and formally launched in 1927 under the Committee on Uniform Crime Reporting. Responsibility for the collection and reporting of data was delegated to the FBI three years later, in 1930. For 90 years, not much changed. The new system, incident based reporting, promises to be a better and more accurate system than what sufficed for ninety years. The National Incident Based Reporting System (NIBRS) reports up to ten offenses within each incident, while the current system only reports the most egregious crime.

For example, consider a robbery that ends in murder. Under the old system, only murder would be reported in the annual crime tally. The new system promises the capability to report both the robbery and the murder.

The current hang-up is that the old system really isn’t the “old” system. Only 57% of law enforcement agencies in the country are participating in NIBRS as of September 2, 2021, though the deadline for switching to NIBRS was January 1, 2021. The change was announced in 2015.

Five states are not yet certified in NIBRS, meaning they do not (yet?) have the technical infrastructure in place to support the new reporting system. Five out of 50 isn’t bad, right? These states are Alaska, California, Florida, New Jersey, and New York – representing a sizeable chunk of the total population. That does not include Alabama, which has no state program, or the eighteen states with less than a majority of their law enforcement agencies participating.

This is data that law enforcement agencies should have on hand. It’s crime data. States have had at least six years to develop the new capabilities, and law enforcement agencies capture the data by the very nature of their work. It is their work.

The difficulties with NIBRS don’t end with low participation. Counting multiple offenses within a single incident will naturally lead to higher crime totals. Any comparison of data aggregated in this manner compared to the single-offense tally system will look like a spike in crime. The spike will, of course, simply be the new parameters in effect.

We saw this effect when the FBI changed the definition of rape years ago. The change limited the ability to trend that crime, particularly at the state level.

We may see similar difficulties trending NIBRS data, though the FBI has said it will continue to convert NIBRS data to the old format for this purpose through the transition period. We do not yet know how long the transition period will last. Once the definition or parameters change, trending becomes much more difficult. We’ll either default to the old system through conversions, or force two incompatible data collection methods together.

It’s too early to say that the FBI has botched the new system. It may very well be useful; it certainly has promise – provided participation increases. The FBI has rolled out a very shiny data exploration tool, though it does not lend itself easily to state-to-state comparisons. Our own NRA-ILA crime data tool goes back to 1960 and specifically allows state-to-state comparisons. It’s at the bottom of this fact sheet about crime. All of the data is from the FBI, but it’s simpler than the FBI’s tool.

Isn’t that what we all want – simple? We expect the FBI, state, and local law enforcement agencies to know how much crime happened in their jurisdiction – or, at least how much crime was reported to law enforcement in their jurisdictions. The FBI is in the 91^st year of managing national crime data, and five years has not been enough time to get the well-publicized new system fully operational.

This is a recurring problem with the federal government. Remember, the CDC annually published absurd estimates on non-fatal gunshot injuries up until they were called out publicly by The Trace. The CDC has since ceased providing that estimate entirely, though anti-gun researchers continued to cite the ridiculous number long after the hopelessly flawed methodology was exposed.

Hospitals know the reasons for their patients’ visits, and law enforcement agencies are aware of the crime reported to them. Why does the federal government have such difficulty aggregating and reporting this data? Researchers and politicians rely on this data to understand the world and develop policy. Anti-gun activists – including those masquerading as earnest researchers – have shown themselves willing to misconstrue data to advance their cause, and this new data system may offer them just such an opportunity. It’s all too easy to see activists comparing incompatible data points without offering a disclaimer.

Before essentially getting its own house in order with basic data collection, anti-gun politicians in Congress and unelected bureaucrats want the federal government to fund research projects. That funding has gone to well-known and well-funded anti-gun research programs and professors at universities with multi-billion-dollar endowments.

Maybe – just maybe – the federal government should focus on getting the data right. Publishing research is far more advantageous to the anti-gun agenda than simply collecting the right data, but we’d all like to live in a country in which government bureaucrats are not actively seeking to infringe on Constitutional rights.