One way to get your message into the news is to come up with a number so amazing it becomes a staple of kitchen and office discussions.
Unfortunately, not all such numbers are correct. Take, for example, two recent factoids put out by the National Center on Institutions and Alternatives.
To dramatize that ''the war on drugs is racially biased on all fronts,'' the NCIA produced reports showing that 43 percent of all young black men in Washington, D.C. -- and 56 percent in Baltimore -- are either in jail, on probation or parole, awaiting trial or being sought by the police.
These dramatic numbers, which received widespread coverage, are feeding the worst racial stereotypes. As syndicated columnist William Raspberry wrote, if ''well over half the young black men in your city are certified criminals, the safest assumption may be that all young black men are criminals . . . do you want these people working around you, living in your neighborhood, or browsing in your store?''
Among blacks, the numbers play to another kind of prejudice: that anyone who favors more aggressive crime control is either a traitor or a racist.
A good rule of thumb is that if a number sounds unbelievable, read the fine print. In the case of the NCIA's study, the errors and bias compound each other.
* The Inflated Numerator. The NCIA's percentage comes from a formula that assumes everyone arrested in D.C. or Baltimore is a city resident. With the population of these cities swelling both day and night by several hundred thousand people from Maryland and Virginia -- and with the vast majority of street drug markets located within the cities -- this is an obvious error.
To include non-city criminals in the numerator and exclude suburban residents from the denominator magnifies the bias. Ask any D.C. cop: half the tags in our most vibrant drug markets are from Maryland and Virginia.
* The Deflated Denominator. In addition, the NCIA did not adjust its population base data to correct for the urban undercount, which was 8.5 percent among all black men in the 1990 Census -- and undoubtedly higher for young black men in central cities.
* Lumping murderers with pot smokers. Even accepting the NCIA's assumptions, only one-third of the young people in their population of offenders are in jail for serious offenses; the rest are on probation or parole. To lump rebellious youths who get a lecture and probation for pot-smoking with murderers and rapists -- as the NCIA's figure does -- is terribly misleading.
* Ignoring other causes. Having discovered that arrest and incarceration rates of whites and blacks diverge, NCIA leaps to the conclusion that ''systematic racism'' in the criminal justice 00 system is the sole explanation.
Yet young black men in Washington also are three times more likely to earn less than $15,000 annually, and five times more likely to have dropped out of high school than their white counterparts. Social scientists have demonstrated beyond any doubt that people who are poor, regardless of race, tend to be arrested and convicted more often than the middle-class or rich.
Such selective analysis of causes can only hamstring the search for solutions. It may be more satisfying to cry racism than to talk about how to get black men into the economic mainstream, but '' the latter may be more productive.
* Comparing Apples and Oranges. The NCIA's strongest evidence of bias is that, while blacks and whites use drugs at the same rate, blacks are much more likely to be incarcerated for drug-related crimes. But use is not the same as sales. Ninety-six percent of all D.C. inmates incarcerated for drugs are locked up for dealing, not use or possession -- an allocation NCIA surely supports. But street drug selling is not an integrated profession.
According to recent reports from the Rand Corporation, almost all street-level drug dealing in Washington is conducted by blacks. White do sell drugs, but they generally do so in the private ''referral marketplace'' of schools, construction sites, bars, and living rooms. Since making an arrest in referral markets is notoriously difficult (remember Marion Barry?), and flagrant drug markets generate more neighborhood complaints than surreptitious sales, police focus on street drug sales. This is pragmatism, not racism.
The NCIA is an organization whose heart is in the right place; they are shocked and saddened to see so much human waste. As reported in "The Winnable War, a Community Guide to Eradicating Street Drug Markets," law enforcement needs to focus more on the conditions which make the drug markets possible rather than relying so heavily on ''buy-and-bust'' operations. We also need drug treatment in prisons, drug testing in place of incarceration for first offenders, and stiffer penalties for buyers from street dealers.
Where the NCIA blunders is by promoting flawed numbers which give comfort to both racists and drug-legalization apologists. Intending to shoot down ''lock-em up'' reactionaries, they have succeeded instead in assassinating the character of all black men.
Roger Conner is executive director of the American Alliance for Rights and Responsibilities and co-author, with Patrick Burns, of "The Winnable War: A Community Guide to Eradicating Street Drug Markets."