OPINION
PER CURIAM.Sgt. Robert Fitzgerald (U.S.Army, Ret.) has been charged in a federal indictment with abusive sexual contact with minors. The district court excluded certain evidence proffered by the government, specifically, expert testimony about the patterns of typical child molesters and lay testimony about Fitzgerald’s prior conduct. We affirm the decision to exclude the expert testimony; we vacate the decision to exclude the lay testimony about the defendant’s prior conduct, and we remand for further consideration of the admissibility of the lay testimony.
I.
Fitzgerald was an instructor for Junior ROTC (JROTC) in Chesterfield County, Virginia, public high schools between 1996 and 2001. His duties included the supervision of JROTC units while they participated in training exercises, competitions, and leadership courses at various U.S. military installations. The indictment charges Fitzgerald with two counts of abusive sexual contact with a minor, see 18 U.S.C. § 2244(a)(3), and three counts of abusive sexual contact through the use of fear, see 18 U.S.C. § 2244(a)(2), all allegedly occurring on federal property. Fitzgerald is charged under statutes that punish “intentional touching, either directly or through the clothing, of the genitalia, anus, groin, breast, inner thigh, or buttocks.” Id. § 2246(3). The contact must be committed with “an intent to abuse, humiliate, harass, degrade, or arouse or gratify the sexual desire of any person.” Id.
Count One (abusive sexual contact of a minor) and Count Two (abusive sexual contact through use of fear) charge that Fitzgerald touched the buttocks of ADW, a fifteen-year-old JROTC student, through her clothing. ADW alleges that Fitzgerald slapped her on the buttocks as she was getting off a bus. Count Three (abusive sexual contact of a minor) charges that Fitzgerald touched one of ADW’s breasts through her clothing. ADW alleges that Fitzgerald rubbed her shoulders and moved his hands down her chest until they touched the top of her breasts and that he squeezed the side of one of her breasts and said “excuse me.” Count Four (abusive sexual contact through use of fear) charges that Fitzgerald again touched ADW’s breast when he put his arm around her after she had been in a water fight. Count Five (abusive sexual contact through use of fear) charges that Fitzgerald rubbed the inner thigh and genitals of TMS, a seventeen-year-old JROTC member, through her clothing. According to TMS, the two were sitting in Fitzgerald’s truck when he told her to come to him if she wanted to have sex with an older black man; he then *860moved his hand up her leg and touched her genitals with the side of his hand.
In pretrial discovery the government listed a psychologist, Anthony J. Pinizzotto, Ph.D., as an expert witness. Pinizzotto works for the FBI as a clinical forensic psychologist. He was to testify about the methodology and behavior of child molesters. He would offer the opinion that a child molester often begins by engaging in seemingly innocuous behavior to gain a child’s trust and then moves to border-line behavior to test whether the child is receptive or suspicious. Pinizzotto was originally slated to offer the additional opinion that Fitzgerald’s actions were consistent with those of a typical child molester. The district court granted Fitzgerald’s motion in limine to exclude Pinizzotto’s testimony, concluding that it did not meet the reliability requirements for admission as expert testimony and that it would not assist the trier of fact. On appeal the government challenges only the district court’s exclusion of Pinizzotto’s expert testimony about the methodology and behavior of child molesters; it does not challenge the court’s exclusion of the testimony that Fitzgerald’s conduct fit the typical pattern of a child molester.
The government also gave notice that it intended to call seven other female JROTC students who would testify about inappropriate sexual contact or comments by Fitzgerald. The government argued to the district court that this testimony was admissible both as evidence of prior sexual assaults, Fed.R.Evid. 413, and as evidence of prior bad acts, Fed.R.Evid. 404(b). The district court granted Fitzgerald’s motion to exclude on the ground that the testimony was not admissible for the “limited purposes” permitted by Rule 413; the court did not address Rule 404(b).
On appeal the government divides the seven other female JROTC students into two groups, claiming that the testimony of two is admissible under Rule 413 and that the testimony of the remaining five is admissible under Rule 404(b). The two girls whose testimony the government seeks to introduce under Rule 413 allege that Fitzgerald touched them inappropriately. The first girl, JLE, would testify that Fitzgerald made inappropriate comments to her and that he touched her buttocks through her clothing at school. The second girl, HMM, would testify that Fitzgerald approached her before a competition ostensibly to straighten her name tag and said that he was “not going to do anything funny.” When he adjusted the tag, however, he brushed her breast with his hand. The government argues that Fitzgerald’s conduct toward these two girls would have been punishable under § 2244 if he had acted on federal property and that testimony about this conduct is admissible as evidence of prior sexual assaults under Rule 413. The five remaining female students all would testify that Fitzgerald directed sexually suggestive comments, questions, and gestures toward them, and two would also testify that he gave them extended hugs. This testimony, the government argues, is about prior bad acts of the defendant that is admissible under Rule 404(b).
The government has perfected an interlocutory appeal from the district court’s order excluding the testimony of the psychologist and the seven other female JROTC students. See 18 U.S.C. § 3731 (allowing an interlocutory appeal of a district court’s order excluding evidence “if the United States attorney certifies ... that the appeal is not taken for purpose of delay and that the evidence is a substantial proof of a fact material in the proceeding.”).
II.
We turn first to the district court’s exclusion of the government’s expert wit*861ness, Dr. Pinizzotto, who is a psychologist. The government would use Dr. Pinizzotto’s testimony about the methodology and behavior of child molesters to assist in proving Fitzgerald’s intent to molest the alleged victims. Dr. Pinizzotto would testify that a child molester typically begins by befriending the child in order to gain the child’s trust. The molester then engages in borderline behavior to test whether the child is receptive or suspicious. We review the district court’s order excluding this expert testimony for abuse of discretion. Gen. Elec. Co. v. Joiner, 522 U.S. 136, 143, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997). Expert testimony is admissible if it is reliable and “will assist the trier of fact to understand the evidence or to determine a fact in issue.” Fed.R.Evid. 702. The district court excluded Dr. Pinizzotto’s expert testimony because it did not satisfy the reliability factors spelled out by Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), and because it would not be helpful to the trier of fact. For the reasons that follow, we affirm this ruling.
We start with the reliability issue. Daubert sets forth the following non-exclusive checklist for assessing the reliability of expert testimony: (1) whether the expert’s theory can be or has been tested; (2) whether the theory has withstood peer review and publication; (3) whether there is a known or potential rate of error; (4) whether standards exist for the application of the theory; and (5) whether the theory has been generally accepted by the relevant scientific community. See Daubert, 509 U.S. at 593-94; United States v. Crisp, 324 F.3d 261, 265-66 (4th Cir.2003). In an effort to establish that Pinizzotto’s testimony would be rehable, the government provided the district court with an affidavit from Pinizzotto, a list of his publications, and one of his articles. Fitzgerald contested reliability in his motion in limine. The record does not indicate how the district court analyzed each of the Daubert factors, but the record ahows us to review the factors.
First, testing. Pinizzotto states in his affidavit that certain “testing, evaluation and assessment methods establish behavioral patterns to target and seduce child victims.” According to Pinizzotto, this testing has produced a body of data that corroborates and forms a basis for his opinions. He cites one article that discusses the application of the research to teachers and other school personnel who have molested children. However, Pinizzotto’s affidavit does not provide any indication of how his theories have been tested or how the testing supports the conclusions he would offer. At bottom, we have little more than assertions about testing, and this is not enough to establish that Pinizzotto’s theory has been adequately tested.
Second, peer review and publication. Pinizzotto’s affidavit asserts that his theories have been published and subjected to peer review. The publication list he provides shows that some of his work has been published, but there is no indication of what, if any, of his work has been peer reviewed. Pinizzotto’s affidavit also cites published articles and books written by others on the techniques of child molesters, but there is no indication whether the works of these other authors have been peer reviewed either. We give Pinizzotto some credit for publication, but we simply do not know whether his work has been subjected to peer review.
Third, known or potential rate of error. Pinizzotto’s materials do not mention error rate, so this factor is not satisfied.
Fourth, standards for application of the theory. Pinizzotto states that the studies he cites use “standards and controls appropriate for and approved for scientifically valid psychological studies,” but he does *862not say what these standards and controls are or whether he abides by them.
Fifth, acceptance in the scientific community. Pinizzotto offers the conclusion that his opinions “are generally accepted within the relevant scientific community.” He does not, however, provide sufficient detail for this statement to be accepted.
When we measure the record against the Daubert factors, as we just have, we cannot say that the district court abused its discretion when it concluded that the government “faded to prove in any way that [Pinizzotto’s] alleged expert testimony would pass [Daubert’s] reliability test.”
The district court also concluded that Dr. Pinizzotto’s testimony would not be helpful to the trier of fact. See Fed. R.Evid. 702. As the government points out, we have held that Rule 702 allows the admission of the expert testimony of an experienced narcotics officer who describes the methodology and tools of drug dealers. The officer typically offers an opinion that a particular defendant’s activity and tools are consistent with drug distribution, not personal use; such an opinion suggests that the government’s evidence establishes intent to distribute. See, e.g., United States v. Hopkins, 310 F.3d 145, 150-51 (4th Cir.2002); United States v. Gastiaburo, 16 F.3d 582, 588-89 (4th Cir.1994). The drug distribution cases and this case present very different issues for a jury. The average juror is not likely to be familiar with how drug dealers operate. Thus, a narcotics officer’s specialized knowledge can be useful to a juror in understanding the evidence in a drug case. See Fed.R.Evid. 702. On the other hand, the basic question in this case is whether the defendant’s comments and conduct amount to abusive sexual contact with minors. This appears to be a question that an average juror can decide without the assistance of expert testimony about the methodology of child molesters. An average juror, in other words, can figure out from the testimony of lay witnesses whether a comment or touch is innocent or evidence of sexual abuse.
The government relies heavily on the Seventh Circuit’s decision in United States v. Romero, 189 F.3d 576 (7th Cir.1999), to argue for the admissibility of Pinizzotto’s testimony. In Romero the court concluded that it was not an abuse of discretion to admit the testimony of an expert who testified about the methods of “preferential child molesters,” that is, molesters who definitely prefer sexual contact with children and methodically pursue that contact. Id. at 583. The intent element in Romero looks markedly different from the intent element the government must prove in this case. In Romero the government had to prove that the defendant traveled (and convinced a minor to travel) across state lines with the intent to engage in sexual activity. No sexual contact was alleged. Id. at 581; see also United States v. Long, 328 F.3d 655, 659-60 (D.C.Cir.2003). The court concluded in Romero that expert testimony would help the jury interpret the defendant’s seemingly innocent actions, much of which involved on-line discussions with young boys about UFOs, extraterrestrials, and the occult. Romero, 189 F.3d at 584-85. But in Fitzgerald’s case, the question is whether when he touched ADW and TMS, he intended to “abuse, humiliate, harass, degrade, or arouse or gratify the sexual desire of any person.” 18 U.S.C. § 2246(3). Pinizzotto’s testimony on the methodology and behavior of child molesters would not be as helpful to the jury as it was in Romero in determining intent. As the district court here noted, the average juror can understand whether a touch is innocent or abusive from lay testimony about the events in question. We cannot say, therefore, that the district court abused its discretion in *863concluding that Pinizzotto’s expert testimony would not be helpful to the trier of fact.
Because the district court did not abuse its discretion by excluding the government’s proffered expert testimony, both because it did not meet Daubert’s reliability requirements and because it would not be helpful to the trier of fact, we affirm this ruling.
III.
We turn next to the district court’s decision to exclude the testimony of seven other female JROTC students, who allege that they were also targets of Fitzgerald’s inappropriate conduct. Again, we review the district court’s decisions for abuse of discretion, and we note that an error of law amounts to an abuse of discretion. United States v. Stitt, 250 F.3d 878, 896 (4th Cir.2001).
A.
JLE and HMM, two female JROTC students, would testify that Fitzgerald touched them inappropriately when they were not at U.S. military facilities. The government seeks to admit this testimony under Rule 413, which governs the admission of prior acts of sexual assault by a defendant. Although the specific basis for the district court’s decision to exclude this testimony is unclear, it appears that the court believed conduct falling within Rule 413’s parameters is admissible only for limited purposes not shown by the government in this case. See J.A. 137 (“Rule 413 permits such evidence for limited purposes, and ... the United States has failed to show any of these limited purposes.”).
A broader analysis of the Rules of Evidence is required. While Rule 404(b) limits the purposes for which prior bad act evidence may be used, Rule 413 explicitly states that evidence of prior acts of sexual assault is admissible for “any matter to which it is relevant.” This does not mean that any evidence fitting within the scope of Rule 413 is automatically admissible. Evidence that is relevant under Rule 413 is also subject to Rule 403.
Evidence offered under Rule 413 must satisfy three elements: (1) the defendant must be accused of an offense of sexual assault as defined by Rule 413(d); (2) the evidence must pertain to the defendant’s commission of another sexual assault offense; (3) and the evidence must be relevant. Fed.R.Evid. 413(a); Doe v. Glanzer, 232 F.3d 1258, 1268 (9th Cir.2000); United States v. Guardia, 135 F.3d 1326, 1328 (10th Cir.1998). Only the second element is in contention here. In applying the second element of Rule 413, the court must determine whether a reasonable jury could conclude that the defendant committed a prior sexual assault offense. See Johnson v. Elk Lake Sch. Dist., 283 F.3d 138, 154-55 (3d Cir.2002); United States v. Enjady, 134 F.3d 1427, 1433 (10th Cir.1998); see also Huddleston v. United States, 485 U.S. 681, 690, 108 S.Ct. 1496, 99 L.Ed.2d 771 (1988). The defendant does not have to have been convicted of, or even charged with, the prior act. Johnson, 283 F.3d at 151-52 (discussing the legislative history). This, of course, means that the district court must also determine whether a reasonable jury could conclude that the defendant acted with the requisite intent when he committed the prior act. Here, in other words, the government must present sufficient evidence to allow a reasonable jury to conclude that when Fitzgerald touched JLE and HMM, he did so with the intent to “abuse, humiliate, harass, degrade, or arouse or gratify the sexual desire of any person” for the evidence to be admissible under Rule 413. 18 U.S.C. § 2246(3).
Even if the district court concludes that the conduct is admissible under Rule *864413, it must still consider whether the probative value of the evidence is “substantially outweighed” by the risk of prejudice, confusion, or undue delay. Fed. R.Evid. 403. Johnson, 283 F.3d at 155; Doe, 232 F.3d at 1268-69; Guardia, 135 F.3d at 1330; United States v. Sumner, 119 F.3d 658, 661 (8th Cir.1997); United States v. Larson, 112 F.3d 600, 604-05 (2d Cir.1997). As part of its Rule 403 analysis, the court should consider factors that affect the probative value of the proffered evidence, including the similarity of the prior acts to the act charged, the closeness in time of the prior acts to the charged conduct, the frequency of the prior acts, the presence or absence of intervening events, and the need for additional testimony to explain the prior acts. Blind-Doan v. Sanders, 291 F.3d 1079, 1082 (9th Cir.2002); Guardia, 135 F.3d at 1331; see also Johnson, 283 F.3d at 155-56 (listing factors). Because the district court is responsible for screening evidence for its admissibility under the rules in the first instance, see Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 592-93, 113 S. Ct. 2786, 125 L.Ed.2d 469 (1993), we remand for the district court to consider the evidence proffered under Rule 413 in accordance with the framework we have just described.
B.
The district court excluded the testimony of five more JROTC students who allege that Fitzgerald directed inappropriate comments, questions, and gestures toward them. The government proffers this testimony as evidence of prior bad acts under Rule 404(b). Evidence of prior bad acts is admissible under 404(b) if it is (1) relevant for a purpose other than propensity (such as to prove intent, for example), (2) probative of something essential to the case, and (3) rehable. United States v. Queen, 132 F.3d 991, 997 (4th Cir.1997). Because of the risk that the jury will use prior bad act evidence to decide the case based on the defendant’s propensity to commit illegal acts, a district court should exercise “greater care” in determining the admissibility of this evidence. Id. at 995. The more similar the prior acts are to the acts at issue, the more likely they are to be relevant. Id. at 996. Moreover, “similar act evidence is relevant only if the jury can reasonably conclude that the act occurred and that the defendant was the actor.” Huddleston, 485 U.S. at 685. The court must therefore determine whether there is “sufficient evidence to support a finding by the jury that the defendant committed the similar act.” Id. at 685. Finally, even if the evidence is admissible under Rule 404(b), it is still subject to the balancing required by Rule 403 to ensure that the probative value is not substantially outweighed by the risk of confusion, prejudice, or undue delay. See Queen, 132 F.3d at 997-98; see also Fed. R.Evid. 403. The district court has not analyzed whether the testimony of the five other girls is admissible under Rule 404(b) or excludable under Rule 403. We remand for that analysis.
IV.
We affirm the district court’s decision to exclude the expert testimony of the psychologist, Anthony J. Pinizzotto, Ph.D. We vacate the court’s decision to exclude the testimony of the seven female JROTC students who would testify about noncharged acts allegedly committed by the defendant; on remand the court will reconsider the admissibility of that testimony.
AFFIRMED IN PART, VACATED IN PART, AND REMANDED.