Terrence Barlow, an African American man, was convicted of possession with intent to distribute cocaine base in violation of 21 U.S.C. § 841(a)(1) and carrying a firearm during a drug trafficking crime in violation of 18 U.S.C. § 924(c). Barlow sought to bar his prosecution on grounds of selective enforcement and filed a motion for discovery on this issue under United States v. Armstrong, 517 U.S. 456, 116 S. Ct. 1480, 134 L. Ed. 2d 687 (1996), which the district court denied. On appeal, Barlow challenges the denial of his Armstrong motion and argues that the jury instructions given at his trial were faulty. We affirm.

On October 29, 1999, Barlow approached the ticket counter at Chicago's Union Station and purchased two one-way tickets to Topeka, Kansas on Amtrak's Southwest Chief, one for himself and one for his friend, William Guidry. Drug Enforcement Administration ("DEA") Transportation Task Force Agents Eric Romano and Patrick Murphy, both working undercover in plain clothes, observed Barlow and Guidry in the waiting area of Union Station. Barlow and Guidry, each carrying a garment bag, kept glancing over their shoulders at the agents and whispering to one another. Their suspicions raised, the two agents followed Barlow and Guidry to the boarding area for the Southwest Chief and asked to speak with them. Romano and Murphy identified themselves as law enforcement officers and briefly interviewed Barlow and Guidry. The agents then asked for and received consent to search Barlow's and Guidry's bags. The agents found in Barlow's garment bag a package containing 485 grams of cocaine base and recovered loaded handguns from both men's luggage; they immediately placed Barlow and Guidry under arrest.

Barlow was indicted on one count of possession with intent to distribute cocaine base and one count of carrying a firearm in relation to a drug trafficking crime. He pleaded not guilty. Guidry did not face federal charges.

In May 2000 Barlow filed a motion for discovery and a hearing under United States v. Armstrong, 517 U.S. 456, 116 S. Ct. 1480, 134 L. Ed. 2d 687 (1996), which articulates the standard a defendant must meet to obtain discovery on a claim that he was singled out for prosecution based on his race. In order to state a constitutional violation, a selective prosecution claim must meet the "ordinary equal protection standards" established by the Supreme Court's jurisprudence on racial discrimination. Id. at 465, 116 S. Ct. 1480. That is, the defendant must demonstrate that the prosecutorial policy in question had both a discriminatory effect and a discriminatory purpose. Id.

Barlow's motion contended that he had been "pursued, stopped, interviewed, and investigated by Drug Enforcement Administration agents based on his race." In his accompanying discovery request, Barlow requested "the names and races of all individuals stopped by all agents and officers detailed to the DEA Transportation Task Force during the years 1995-2000, including, but not limited to, date and time of stop; length of stop; reason for the stop; location of the stop; and outcome of the stop and name [sic] of all agents or officers involved in the stop; or records from which this data can be obtained." Essentially, Barlow contended that, in singling him out for an interview and search, the two agents had engaged in unconstitutional "racial profiling," a form of selective enforcement.

In support of his Armstrong motion, Barlow submitted the affidavit of Dr. John Lamberth, a psychologist and statistician, who has served as an expert witness on several racial profiling cases. With the goal of substantiating Barlow's claim, Dr. Lamberth supervised a field study of law enforcement activity in Union Station.

From February 28 through March 10, 2000, investigators working for Dr. Lamberth conducted surveillance in Union Station to determine whether race played a role in law enforcement decisions to approach or stop travelers. These investigators counted the total number of passengers who entered the departure gate for the Southwest Chief and the subset of African Americans in that group. They also recorded the race of those individuals from the total number of travelers who were approached by law enforcement agents.

Dr. Lamberth's investigators reported only one incident involving a law enforcement stop or interview. On February 29, 2000, the investigators observed an Amtrak porter point out an African American couple to an Amtrak police officer. The Amtrak police officer spoke to the couple, who were subsequently escorted from the waiting area by two uniformed officers and two plain-clothes officers. The investigators did not see law enforcement officials approach or interview any other passengers at any other time during their surveillance.

The only individuals known to have been approached by law enforcement officials in Union Station — Barlow and Guidry, and the couple — were African American. Dr. Lamberth opined that this pattern of law enforcement stops of individuals boarding the Southwest Chief at Union Station suggested that law enforcement agents could be engaging in racial profiling when approaching and stopping travelers.

The district court denied Barlow's motion without a hearing, finding "statistically indefensible" Dr. Lamberth's inclusion of Barlow and Guidry in the data pool for his ten-day study of law enforcement behavior in Union Station. Barlow's case proceeded to trial, and the jury returned a verdict of guilty on both counts. Barlow was sentenced to 151 months' imprisonment for possession with intent to distribute and 60 months' imprisonment on the firearm charge, to be served consecutively.

Barlow first argues that the district court erred in denying his motion for discovery because he produced sufficient evidence to warrant further investigation of his claim that the DEA agents had engaged in racial profiling. We review the denial of a motion for discovery in a criminal case for abuse of discretion. United States v. Bastanipour, 41 F.3d 1178, 1181 (7th Cir.1994).

Barlow's motion for discovery invoked Armstrong, in which the Supreme Court defined the showing necessary for a defendant to obtain discovery on a selective prosecution claim. 517 U.S. at 465, 116 S. Ct. 1480. Barlow complains not of selective prosecution, but of racial profiling, a selective law enforcement tactic. But the same analysis governs both types of claims: a defendant seeking discovery on a selective enforcement claim must meet the same "ordinary equal protection standards" that Armstrong outlines for selective prosecution claims. See Armstrong, 517 U.S. at 465, 116 S. Ct. 1480; Chavez v. Ill. State Police, 251 F.3d 612, 635-36 (7th Cir.2001); United States v. Hayes, 236 F.3d 891, 895 (7th Cir.2001). To prevail on his motion, therefore, Barlow needed to demonstrate that the agents' actions had a discriminatory effect and that the agents had a discriminatory purpose when they approached him in Union Station. Armstrong, 517 U.S. at 465, 116 S. Ct. 1480; Chavez, 251 F.3d at 635-36; Hayes, 236 F.3d at 895.

Law enforcement has a racially discriminatory effect when members of a protected racial group — in this case African Americans — receive less favorable treatment than nonmembers. See Armstrong, 517 U.S. at 465, 116 S. Ct. 1480; Chavez, 251 F.3d at 636; Hayes, 236 F.3d at 895. In other words, to establish discriminatory effect, an African American claimant must demonstrate that a law or regulation was enforced against him, but not against similarly situated individuals of other races. Armstrong, 517 U.S. at 465, 116 S. Ct. 1480; Chavez, 251 F.3d at 636; Hayes, 236 F.3d at 895. Barlow contended that the DEA agents had enforced the law selectively by choosing to approach, interview, and search African Americans but not Caucasians, i.e., by engaging in racial profiling. To obtain discovery on this claim, Barlow was required to present evidence that DEA agents chose not to approach whites to whom he was similarly situated. Armstrong, 517 U.S. at 468-69, 116 S. Ct. 1480; Chavez, 251 F.3d at 638; Hayes, 236 F.3d at 895. A finding that DEA agents did not approach whites who rode the Southwest Chief as frequently as African American travelers would not automatically establish that the agents' investigatory tactics were discriminatory; Barlow needed to show also that at least some of these whites not approached were similarly situated to him. Armstrong, 517 U.S. at 468-69, 116 S. Ct. 1480; Chavez, 251 F.3d at 638; Hayes, 236 F.3d at 895.

Barlow introduced Dr. Lamberth's statistical analysis in an attempt to demonstrate that the DEA agents had a practice of approaching African American travelers but not similarly situated white travelers. Statistical data has proven a useful tool in some high-profile state racial profiling cases. See, e.g., State v. Soto, 324 N.J.Super. 66, 734 A.2d 350 (1996) (statistical evidence that blacks were 4.85 times more likely than whites to be stopped for traffic violations established prima facie case of discriminatory effect). And although statistics alone rarely establish an equal protection violation, they may be sufficient to establish the discriminatory effect prong of the Armstrong test. Chavez, 251 F.3d at 640. But such statistics must be relevant and reliable, id., and the ones Barlow provided were neither.

Dr. Lamberth's statistical conclusions rely heavily upon the fact that, in the ten days his investigators observed the Southwest Chief waiting area, law enforcement officials intercepted only two individuals — an African American couple. But the African American couple was approached by a uniformed Amtrak police officer whereas Barlow was approached by two plain-clothes DEA Transportation Task Force agents. Barlow leveled his allegations against the DEA, not Amtrak; observations of Amtrak's law enforcement activities are irrelevant to a claim that the DEA engaged in racial profiling. Moreover, Dr. Lamberth's investigators could provide no information as to why the Amtrak officer approached the African American couple; Barlow has not even established that the officer did so for a law enforcement purpose.

Even more problematic is Dr. Lamberth's flawed statistical methodology: Dr. Lamberth's investigators counted 726 travelers between February 28 and March 10, 2000, of whom 119, or 16.4%, were African American. Only two of the 726 travelers were approached by law enforcement agents, and both were African American. But Dr. Lamberth did not simply calculate the statistical significance of the fact that two out of two travelers approached were African American. He added Barlow and Guidry to the subset of travelers who had been approached by law enforcement and calculated that the probability that all four individuals approached would be African American to be less than 8 times in 10,000, which he considered a "highly statistically significant" result. But the incident involving Barlow and Guidry occurred on October 29, 1999, months before Dr. Lamberth's investigators began their surveillance in Union Station. Presumably, travelers other than Barlow and Guidry entered the Southwest Chief departure gate on October 29th. We do not know, however, whether law enforcement officials approached additional travelers that day, or, if they did, the race of these travelers. For all we or Dr. Lamberth know, several white individuals could have been approached that day in addition to Barlow and Guidry; if other individuals of any race were approached that day, their addition to the data pool could drastically alter Dr. Lamberth's statistical results.

Barlow argues that Dr. Lamberth's conclusions were sound, pointing out that Dr. Lamberth was an expert witness in Soto, and that the New Jersey court in Soto accepted his noncontemporaneous statistical analysis as evidence of racial profiling. But Dr. Lamberth did not use the same methodology here as he did in Soto. 734 A.2d at 352-53. Soto involved claims that the New Jersey State Police engaged in selective enforcement in traffic stops between April 1988 and May 1991. Id. at 352. Dr. Lamberth conducted his research two years after the time period relevant to the selective enforcement claims had ended: In June 1993 he recorded the number of drivers stopped by police on the New Jersey Turnpike and the subset of those drivers who were African American. His statistical analysis of these numbers suggested that the state police used racial profiling when enforcing traffic laws. Id. at 352-53. Given that there was no evidence that traffic patterns had changed between 1991 and 1993, the court accepted Dr. Lamberth's 1993 statistics as evidence that the stops between 1988 and 1991 resulted from selective enforcement. Id. at 352, 360-61. Significantly, Dr. Lamberth did not add the allegedly discriminatory stops from 1988 to 1991 to the pool for his June 1993 traffic survey, id. at 352; instead he relied on the 1993 data only as a baseline by which to evaluate the period relevant to the defendants' claims. Id. at 352-53. In Barlow's case, on the other hand, Dr. Lamberth did not merely use the ten-day observation period to evaluate the possibility that the DEA agents engaged in racial profiling when they approached Barlow and Guidry; instead he added Barlow and Guidry to the pool of travelers observed during the ten-day period, thereby altering the significance of his data.

Even if we accept Dr. Lamberth's conclusions as statistically valid, however, Barlow has still presented no evidence that he received less favorable treatment than similarly situated white travelers. To meet his burden under Armstrong, Barlow needed to present evidence that the DEA agents observed whites engaging in the same behavior as Barlow — i.e., looking nervously over their shoulders — but chose not to approach them. 517 U.S. at 465, 116 S. Ct. 1480. But Dr. Lamberth's data tells us nothing about the behavior of the white travelers in Union Station; we therefore have no basis for concluding that any of these white travelers were similarly situated to Barlow.

Moreover, Barlow has not demonstrated that the DEA agents acted with discriminatory purpose when they approached him. Chavez, 251 F.3d at 645. The agents made no racial comments during their encounter with Barlow, and there is no evidence of a DEA Transportation Task Force policy, either actual or de facto, encouraging racial profiling. Without evidence of both discriminatory effect and discriminatory intent on the agents' part, Barlow could not make the threshold showing required in Armstrong; the district court therefore did not abuse its discretion in denying Barlow's motion for discovery.

Barlow also argues that his conviction must be reversed because the district court instructed the jury that the government had to prove only that Barlow knew that he was carrying some kind of prohibited drug, not that he knew specifically that he was carrying cocaine base. Barlow contends that the Supreme Court's decision in Apprendi v. New Jersey, 530 U.S. 466, 120 S. Ct. 2348, 147 L. Ed. 2d 435 (2000), requires that a jury find beyond a reasonable doubt that a defendant knew the exact substance involved in a drug offense. But Apprendi does not alter the fact that actual knowledge of the identity of a drug is not an element of 21 U.S.C. § 841(a). United States v. Carrera, 259 F.3d 818, 830 (7th Cir.2001). Section 841(a) requires only that the defendant know that he possesses a controlled substance; it does not require that he know the type of controlled substance he possesses. United States v. Martinez, 301 F.3d 860, 865 (7th Cir.2002); Carrera, 259 F.3d at 830. The district court's jury instructions did not violate Apprendi.

AFFIRMED.