Chief Justice, BALES, opinion of the Court.

¶ 1 The State's evidence supporting the conviction of Joseph Javier Romero included testimony by a firearms examiner, based on a toolmark comparison, that a certain pistol had fired six shell casings found at the murder scene. We consider whether the trial court abused its discretion by precluding Romero from offering expert testimony that firearms examiners use subjective rather than scientifically rigorous methods in drawing conclusions from indentations on shell casings. Because Romero's expert witness was qualified and his testimony would have been helpful to the jury in understanding the evidence, the trial court erred in excluding the testimony.

I.

¶ 2 In June 2000, a man was killed by two gunshots. Although witnesses did not see the shooting, they heard gunshots and saw two or three men flee in a dark Ford Ranger or Mazda pickup truck. Police found six spent .40-caliber shell casings and bullet fragments at the murder scene. A cell phone was also found next to the victim's body.

¶ 3 Nearly one month later, police officers stopped Romero for reasons unrelated to the murder. He possessed the magazine for a.40-caliber Glock pistol. The officers subsequently found a .40-caliber Glock pistol without its magazine along the path Romero had traveled just before encountering them. Police retained the pistol and the magazine.

¶ 4 Seven years later, a "cold case'" investigative unit inspected the cell phone and traced it to Robert E. and, through him, to Romero. Robert E. told police that, while a college student in 2000, he had known a person named "Joe" who supplied him drugs and sometimes borrowed Robert E.'s black Ford Ranger. Robert E. recalled that he had loaned his pickup truck to Joe in the summer of 2000, possibly June, and Joe had kept it longer than expected.

¶ 5 Frank Powell, a police firearms expert, examined the six .40-caliber shell casings found at the murder scene and concluded that they were all fired from the same gun. Later, Powell test-fired the Glock pistol recovered when the police stopped Romero in July 2000. Comparing indentations on the shell casings, Powell concluded that this pistol had fired the casings found at the murder scene.

¶ 6 Romero was indicted for first degree murder. The jury hung on the charge, resulting in a mistrial. Before his retrial, Romero moved to preclude Powell from testifying, arguing that the firearm examiner's conclusions lacked the reliability required by Arizona Rule of Evidence 702 and Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). At a hearing on this motion, the trial court reviewed a transcript of Powell's testimony at Romero's first trial and also considered testimony by Dr. Ralph Haber, a defense expert. Dr. Haber was not offered to testify whether Powell had correctly analyzed the toolmarks on the shell casings. Instead, Dr. Haber, based on his expertise in the broader field of experimental design, criticized the scientific reliability of drawing conclusions by comparing toolmarks.

¶ 7 The court denied Romero's motion, finding that Powell was qualified as an expert in the field of firearms examination and that his opinions resulted from reliable principles and methods. It also granted the State's motion to preclude Dr. Haber from testifying as a defense expert at the second trial, reasoning that Dr. Haber was not qualified as an expert in firearms identification and, alternatively, that his testimony would impermissibly invite the jury to revisit Daubert issues decided by the judge with regard to Powell's testimony.

¶ 8 At Romero's second trial, Powell testified that the shell casings from the murder scene "matche[d] very well" with the casings from test-firing the Glock pistol. He therefore concluded that the casings from the murder scene were fired from the Glock that the police found when they stopped Romero in July 2000. The jury acquitted Romero of first degree murder but convicted him of the lesser-included offense of second degree murder. The trial court sentenced Romero to a presumptive term of sixteen years' imprisonment.

¶ 9 In affirming Romero's conviction and sentence, a divided panel of the court of appeals held that the trial court had not abused its discretion in admitting Powell's testimony or in precluding Dr. Haber from testifying. State v. Romero, 236 Ariz. 451, 457-60 ¶¶ 18-32, 341 P.3d 493, 499-502 (App. 2014). Specially concurring, one judge concluded that the trial court had erred by precluding Dr. Haber's testimony, but that the error was harmless. Id. at 469 ¶ 69, 341 P.3d at 511 (Eckerstrom, J., concurring).

¶ 10 We granted Romero's petition for review solely with regard to the preclusion of Dr. Haber's testimony because Rule 702's standard for admitting expert testimony is a recurring issue of statewide importance. We have jurisdiction pursuant to Article 6, Section 5(3) of the Arizona Constitution and A.R.S. § 12-120.24.

II.

¶ 11 We review a trial court's exclusion of evidence for an abuse of discretion, and we review de novo the interpretation of the Rules of Evidence. A court abuses its discretion by committing an error of law. State v. Bernstein, 237 Ariz. 226, 228 ¶ 9, 349 P.3d 200, 202 (2015). As the proponent of Dr. Haber's expert testimony, Romero bears the burden of establishing its admissibility by a preponderance of the evidence. Id.

A.

¶ 12 Arizona Rule of Evidence 702 allows an expert witness to testify if, among other things, the witness is qualified and the expert's "scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence...." Trial courts serve as the "gatekeepers" of admissibility for expert testimony, with the aim of ensuring such testimony is reliable and helpful to the jury. Id. cmt. (2012).

¶ 13 The trial court here concluded that Dr. Haber was not qualified to testify as an expert in firearms identification. In affirming, the court of appeals noted that Dr. Haber, although having reviewed the literature on firearms identification, had not previously been retained as an expert on firearms identification, conducted a toolmark analysis, attempted to identify different firearms, or conducted research on firearms identification. 236 Ariz. at 458 ¶¶ 23-25, 341 P.3d at 500.

¶ 14 The issue, however, is not whether Dr. Haber was qualified as an expert in firearms identification, but instead whether he was qualified in the area of his proffered testimony—experimental design. Here, the trial court determined that Powell was qualified to offer an expert opinion that the shell casings were all fired from the same Glock. But Romero did not offer Dr. Haber as an expert in firearms identification to challenge whether Powell had correctly performed his analysis or formed his opinions. Instead, Dr. Haber's testimony was proffered to help the jury understand how the methods used by firearms examiners in performing toolmark analysis differ from the scientific methods generally employed in designing experiments.

¶ 15 Under Rule 702, when one party offers an expert in a particular field (here, the State's presentation of Powell as an expert in firearms identification) the opposing party is not restricted to challenging that expert by offering an expert from the same field or with the same qualifications. The trial court should not assess whether the opposing party's expert is as qualified as—or more convincing than—the other expert. Instead, the court should consider whether the proffered expert is qualified and will offer reliable testimony that is helpful to the jury. Cf. Bernstein, 237 Ariz. at 230 ¶ 18, 349 P.3d at 204 (noting that when the reliability of an expert's opinion is a close question, the court should allow the jury to exercise its fact-finding function in assessing the weight and credibility of the evidence).

¶ 16 The gist of Dr. Haber's proffered testimony was that the methods generally used in conventional toolmark analysis fall short of scientific standards for experimental design. Dr. Haber's testimony was therefore directed at the scientific weight that should be placed on the results of Powell's tests. Such questions of weight are emphatically the province of the jury to determine. E.g., State v. Lehr, 201 Ariz. 509, 517 ¶¶ 24-29, 38 P.3d 1172, 1180 (2002). The trial court erred by focusing on whether Dr. Haber was qualified as an expert in firearm identification rather than considering the proper scope of his proffered testimony—experimental design.

B.

¶ 17 We turn to whether Dr. Haber was qualified to opine on the experimental design of toolmark analysis generally, and how it contrasts with other experimental designs rooted in the scientific method. Under Rule 702, a witness may be qualified based on "knowledge, skill, experience, training, or education." For a witness to be qualified as an expert, he or she need only possess "skill and knowledge superior to that of [people] in general." State v. Girdler, 138 Ariz. 482, 490, 675 P.2d 1301, 1309 (1983) (quoting State v. Watson, 114 Ariz. 1, 12, 559 P.2d 121, 132 (1976)). Careful study may suffice to qualify an expert if it affords greater knowledge on a relevant issue than the jury possesses. State v. Macumber, 112 Ariz. 569, 570, 544 P.2d 1084, 1085 (1976).

¶ 18 Dr. Haber has a Ph.D. in experimental psychology from Stanford, which in turn qualified him to teach experimental design at Yale, the University of Illinois, and the University of Rochester for some twenty-one years. Based on his education and experience, he founded his own consulting business, through which he analyzes forensic science methods and makes himself available to testify about their consistency with accepted methods of scientific experimentation. He routinely conducts peer review for academic journals in many scientific or forensic fields—including firearm and toolmark analysis—regarding the experimental designs used to support the conclusions reached. He has authored a paper for the California Bar Association regarding "evidence in the criminal courts on firearms and handgun identification." Although he has been retained only once to testify about the methods used in toolmark analysis, Dr. Haber has studied and evaluated this issue for four years and "thoroughly familiarized" himself with the research, publications, and methodology for toolmark identification, including all publications from the Association for Toolmark and Firearm Examiners.

¶ 19 With respect to experimental design, and a comparison of the methods generally used by firearms examiners to the scientific method, Dr. Haber is qualified as an expert.

C.

¶ 20 Apart from Dr. Haber's qualifications, his testimony would not have been admissible unless it would have been helpful to the jury in understanding the evidence. Ariz. R. Evid. 702(a). The State presented Powell's testimony that the indentations on shell casings demonstrated that the Glock had fired all the shells, including those at the murder scene, and the State argued that the toolmark comparisons demonstrated a match to "a reasonable degree of scientific certainty." Dr. Haber's testimony would have been helpful to the jury in understanding how the toolmark analysis differed from general scientific methods and in evaluating the accuracy of Powell's conclusions regarding "scientific certainty."

¶ 21 The thrust of Dr. Haber's testimony was that the methods underlying toolmark analysis (here comparing indentations and other marks on shell casings) are not based on the scientific method, but instead reflect subjective determinations by the examiner conducting the analysis. Haber would have explained that unlike experts who use other forms of forensic analysis rooted in the scientific method, firearms examiners do not follow an accepted sequential method for evaluating characteristics of fired shell casings and comparing them to control subjects. By describing the methods used by toolmark examiners, Dr. Haber's testimony could have helped the jury assess how much weight to place on Powell's "scientific" conclusion that the shell casings at the murder scene could only have been fired from the Glock found by the police when they stopped Romero.

¶ 22 In affirming the exclusion of Dr. Haber's testimony, the court of appeals observed that Dr. Haber "could not describe the methods or protocols of a toolmark analysis." 236 Ariz. at 458 ¶ 25, 341 P.3d at 500. This misconstrues the point of Dr. Haber's testimony. One of his critiques of the methodology used by firearms examiners is that they do not employ identifiable, standardized protocols. Excluding testimony about the lack of such protocols because the proffered expert witness cannot identify them would transform a procedure's weaknesses into a shield from criticism. Such reasoning would undermine—rather than promote—Rule 702's purpose of helping the jury understand the evidence.

¶ 23 Moreover, the testimony would have been helpful even though Dr. Haber had not himself performed a toolmark analysis and would not have opined on the particular analysis performed by Powell. Expert witnesses may helpfully educate the fact-finder about general principles without considering the facts of a particular case. State v. Salazar-Mercado, 234 Ariz. 590, 593 ¶ 10, 325 P.3d 996, 999 (2014) (holding that "cold" expert testimony may be admitted under Rule 702). Dr. Haber's lack of experience in performing toolmark analyses and firearm identification experiments might have affected the weight a juror would give his testimony, but it did not bar its admission.

¶ 24 Dr. Haber's testimony was intended to highlight that the conclusions drawn by firearms examiners from toolmarks do not result from the application of articulable standards and lack typical safeguards of the scientific method such as independent verification by other examiners. Thus, Dr. Haber's testimony could have helped the jury to understand any deficiencies in the experimental design of toolmark analysis and to assess any suggestion that such analysis was "scientific." Cf. Salazar-Mercado, 234 Ariz. at 594 ¶ 15, 325 P.3d at 1000 (affirming admission of expert testimony about general behavior patterns of child sexual abuse victims because it "might have helped the jury to understand possible reasons for the delayed and inconsistent reporting in this case").

D.

¶ 25 As an alternative grounds for excluding Dr. Haber's testimony, the trial court ruled that it would impermissibly amount to a second Daubert hearing before the jury. Romero, 236 Ariz. at 457 ¶ 19, 341 P.3d at 499. Because the court of appeals affirmed the preclusion of Dr. Haber's testimony based on his lack of qualifications as a firearms examiner, it did not address this alternative grounds. Id. at n. 4.

¶ 26 The trial court reasoned that because it found Powell's methodology and conclusions sufficiently reliable to be admissible, the defense could not present expert testimony at trial to challenge the prior evidentiary ruling. We have rejected similar reasoning in interpreting Arizona's previous version of Rule 702. In Lehr, we held that a trial court, after ruling based on a pretrial hearing that testimony by the State's DNA experts was admissible, erred by precluding cross-examination of the experts at trial about protocols they had followed. 201 Ariz. at 517 ¶¶ 25-29, 38 P.3d at 1180. Although the trial court had considered the testimony to be elicited on cross-examination before ruling the State's expert testimony admissible, the defense was entitled to offer the same evidence at trial through cross-examination to challenge that testimony. Id. The trial court's contrary reasoning, we noted, "fail[ed] to recognize that very often the same proof used to establish admissibility also impacts weight and credibility." Id. ¶ 25. Consequently, blanket preclusion at trial of evidence presented at a pretrial hearing "infringe[s] upon the role of the jury and improperly insulate[s] the state's evidence from critique." Id. ¶ 29.

¶ 27 Our post-Daubert amendments to Rule 702 do not alter this aspect of Lehr. Instead, we have recognized that a trial court's admission of disputed expert testimony leaves to the fact-finder the role of assessing its weight and credibility. "Cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence." Bernstein, 237 Ariz. at 231 ¶ 22, 349 P.3d at 205 (quoting Ariz. R. Evid. 702 cmt. (2012)).

¶ 28 Here, the trial court's alternative ground for preclusion was an error of law. Assuming that Powell's methods and conclusions regarding the purported "match" between the Glock and the shell casings at the crime scene were sufficiently reliable to be admitted into evidence (we declined review on this issue), it does not follow that the weight and credibility of this evidence, once admitted, may not be challenged. See Lehr, 201 Ariz. at 517 ¶ 29, 38 P.3d at 1180 (noting jury's province to determine weight and credibility of expert testimony).

¶ 29 Our opinion, of course, does not suggest that a jury would necessarily credit Dr. Haber's testimony if it had been admitted. The State could have challenged that testimony before the jury, including by noting some of the points mentioned by the court of appeals, such as Dr. Haber's lack of experience in actually performing toolmark examinations or by questioning whether standards for experimental design in other forensic areas should apply to toolmarks. Such arguments, however, go to the weight rather than the admissibility of Dr. Haber's testimony.

¶ 30 The court's exclusion of Dr. Haber's testimony will not require reversal of Romero's conviction if the State can establish the error was harmless beyond a reasonable doubt. The specially concurring opinion concluded the error was harmless, 236 Ariz. at 469 ¶ 69, 341 P.3d at 511 (Eckerstrom, J., concurring), but the majority did not address this issue. We accordingly remand to the court of appeals so that court may consider in the first instance whether excluding Dr. Haber's testimony was harmless.

III.

¶ 31 We vacate paragraphs 19 through 32 of the opinion of the court of appeals and remand to that court to determine if the error in excluding Dr. Haber's testimony was harmless.