A region's racial composition is THE most reliable predictor of its rate of violent crimes. And yet, apparently, the FBI does not adjust for racial factors when it seeks to impute its guesswork about crime rates, in an attempt to compensate for partial or inefficient crime reporting by local jurisdictions. The FBI's "imputation algorithm" makes allowances for comparatively trifling factors, but (from what I've been able to gather from the web) not for race.

In 1994, the FBI changed the way it copes with the fact that some police jurisdictions are more efficient than others at reporting arrests for violent crime activity to the federal agency. Earlier, the FBI merely reported known arrests and labeled the gaps as such. But since 1994, they have begun "guessing" about the data that "should" have been reported, if the less-efficient jurisdictions had performed more efficiently. You readers who have had a science course or two know the invalidity of substituting guesswork for experimental data, and those of you who understand basic numerical analysis understand the risks of error in extrapolation.

Nonetheless,

What would you care to bet that the ADL wasn't foremost in those "discussions by UCR file users", or that the ADL isn't influential with the "Inter-university Consortium for Political and Social Research," or that the "Imputation Algorithm" and the "Coverage Indicator" are really devices to ensure greater accuracy in the crime-related information presented by the Uniform Crime Reports? I wouldn't wager a nickle on any of those things.

Notice that the racial composition of a jurisdiction isn't among the factors that the FBI will consider when making its "guestimates" about how to fill in the gaps in the data reported by local police and county sheriff's departments. Let's consider two counties in about the same area: Fulton County (pop. 650,000) and Gwinnett County (pop. 353,000), both near Atlanta, Georgia. The two populations have large differences in racial composition, with Fulton County being mostly Black and Gwinnett County being mostly White. Predictably, they also have large differences in violent crime.

The reader is invited to compare the italicized numbers in the last row with: 7.9, 4.4, 9.6, 3.9 (respectively); those are the ratios of Black-to-White per capita rates in those same crime categories for the entire U.S., averaging data from 1993 and 1995. The urbanization of both counties may be affecting the ratios slightly, especially for assault, but otherwise the match is a good one and demonstrates the fact that many behavioral tendencies are racial and hereditary.

In 1990, Gwinnett County had a total population of 352,910 — of which 89.4% were Whites (real ones) and 5.2% were Blacks. Gwinnett County had 13 murders, 94 rapes, 372 robberies, and 637 aggravated assaults reported to the police during 1990.

In 1990, Fulton County had a total population of 648,779 — of which 46.8% were Whites (real ones) and 50.1% were Blacks. Fulton County had 227 murders, 794 rapes, 6542 robberies, and 9178 aggravated assaults reported to the police during 1990.

There you have it: two counties in the same state, side-by-side, sharing portions of the same major American city, having much history and commerce in common. The biggest difference between them is racial demographics. The mostly Black county has much more crime than the mostly White county does.

The numbers suggest, furthermore, that many of the violent crimes in mostly White Gwinnett County are being perpetrated by the Blacks who do live there. Consider: Fulton County had 17.7 times as many Blacks as Gwinnett County did. Multiplying the crime totals for Gwinnett County by 17.7, we get these predictions for Fulton County's crime totals: 230 murders (the actual number was 227), 1664 rapes (the actual number was 794), 6584 robberies (the actual number was 6542), and 11275 aggravated assaults (the actual number was 9178). The relative errors for these predictions are: 1.3 percent for murder, 109.6 percent for rape, 0.6 percent for robbery, and 22.8 percent for aggravated assault. So the guess that Blacks are responsible for most of the crimes that do occur in Gwinnett County is a lousy guess for rape, but it is a pretty good guess for aggravated assault, and it is an excellent guess for murder and robbery. (It should be remembered, however, that rape is an underreported crime and possibly a fair number of Black female rape victims are keeping quiet about their abuse.)

But suppose that the FBI, prodded perhaps by the ADL, chose to believe that the wide variation in the per capita rates of violence in Fulton and Gwinnett Counties were due to an incomplete reporting of violent crime in the mostly White Gwinnett County because of a lack of efficiency of the police there, instead of putting the blame on race, where it really belongs. To reduce the "imbalance" in the crime rates, the FBI might "adjust" Gwinnett County's total number of violent crimes upwards and present their additions to us as factual data. We would have no way of knowing the truth without actually going to Gwinnett County and doing a lot of on-the-spot investigation ourselves! And which of us has enough money to keep doing that all around the country?