A Superior Court jury convicted the defendant, Wayne C. Tassone, of unarmed robbery, in violation of G. L. c. 265, § 19 (b); and assault and battery, in violation of G. L. c. 265, § 13A (a). The issue on appeal is whether an expert witness may offer an opinion that the deoxyribonucleic acid (DNA) profile generated from a known saliva sample of the defendant matched a DNA profile obtained from a swab taken from eyeglasses that were left at the scene of a robbery where the expert had no affiliation with the laboratory that conducted the DNA testing of the eyeglasses swab. We conclude that an opinion regarding the results of DNA testing is admissible only where the defendant has a meaningful opportunity to cross-examine the expert witness about the reliability of the underlying data produced by such testing. Here, the defendant was deprived of a meaningful opportunity for such cross-examination because the analysts who generated the DNA profiles through DNA testing did not testify at trial, and the expert witness who offered the opinion of a match had no affiliation with the laboratory that tested the crime scene sample. Because the defendant preserved his objection to the admission of the expert opinion and its admission was prejudicial, we vacate the defendant’s convictions and remand for a new trial.
Background. We summarize the evidence at trial. Robert Brodeur worked as an assistant manager at a small variety store in Pittsfield. On June 22, 2009, at approximately 5:30 p.m., the only customer in the store was a white male, with short blonde hair, wearing a black T-shirt, black shorts, and eyeglasses. The man went to the store counter to purchase juice. When Brodeur opened the door to the cash register, the man tried to reach into the register to take the money; when that failed, he pulled the register away from Brodeur and threw it on the floor. Brodeur jumped over the counter and attempted to push the man out the door, but the man pushed Brodeur to the floor. While Brodeur was on the floor, the man grabbed $350 in cash and fled. When the police responded to the scene, they observed several items on the floor, including a pair of eyeglasses that did not belong to Brodeur.
*393The defendant worked as a house painter for a company owned by Scott Moore. On the morning of June 22, Moore met the defendant at a credit union in Pittsfield and gave him $300 in cash, along with a ladder and some tools. The defendant was wearing a black shirt and black shorts.
On June 23, the defendant told his then-fiancée, Brenda Streit, that he had been on a “crack” cocaine binge and was entering a local inpatient facility to treat his drug problem. In a subsequent telephone conversation, Streit told the defendant that she was going to the police department to look at a pair of eyeglasses, and the defendant told her that she knew what she had to do. When Streit asked him if he had committed the robbery, the defendant replied, “What do you think?” He told her that he had committed the robbery because of his drug problem.
On June 24, Brodeur was shown a photographic array of eight photographs and picked out the photograph of the defendant as the person who committed the robbery. He initially was eighty per cent sure but, after he “kept looking at it,” said he was “99 per cent sure.”
Both Moore and Streit identified the eyeglasses found at the scene of the robbery as belonging to the defendant, noting that they looked like the defendant’s eyeglasses and had paint specks on them. The defendant, however, shared a number of nonprescription eyeglasses with his brother, Nathan Hunt, and Hunt had a pair that looked “just like” the eyeglasses found at the crime scene. Hunt was the same height and body shape as the defendant, and also had a drug problem with both heroin and crack cocaine. Although Hunt was not employed by Moore’s company, the defendant brought Hunt to the home whose exterior he was painting to assist with the paint work.
With respect to the DNA testing, a police officer used a buccal swab to obtain a saliva sample from the defendant, which was sent to the State police crime laboratory. At the defendant’s trial, no one from the State police crime laboratory testified regarding the analysis of that saliva sample to generate a known DNA profile of the defendant.
Lynne Sarty, a chemist in the State police crime laboratory, took a moistened swab and rubbed it on the areas of the eyeglasses that would likely come into contact with the wearer’s skin, such *394as the sides and the nose piece underneath the lenses. Tiffany Roy, another chemist at the State police crime laboratory, testified that the swab from the eyeglasses was sent to the Cellmark laboratory (Cellmark) in Texas for DNA testing, which generated a DNA profile from the swab, but no one from the Cell-mark laboratory testified regarding the actual testing of that swab or the generation of a DNA profile from that swab. Instead, Roy testified on direct examination that she conducted an “independent review” of the DNA profile from the swab and the known DNA profile of the defendant, and concluded that “the DNA profile generated from the known saliva sample from [the defendant] matched the major male [DNA] profile obtained from the swab of the eyeglasses.”
The defendant appealed from his convictions and the Appeals Court affirmed. Commonwealth v. Tassone, 83 Mass. App. Ct. 197 (2013). The Appeals Court concluded that the defendant’s challenge to the admissibility of Roy’s expert opinion “is controlled in all material respects” by the judgment of the United States Supreme Court in Williams v. Illinois, 132 S. Ct. 2221 (2012). Tassone, supra at 199. The court stated that, although the plurality’s reasoning in Williams might not “be relied upon in circumstances presented in some subsequent confrontation clause cases, where the facts before us are not materially distinguishable from those presented to the Supreme Court in Williams, it must follow from the judgment in that case, which turned on the Court’s conclusion that the Cellmark report . . . was admissible, that the report here is admissible as well.” Id. We granted the defendant’s application for further appellate review.
Discussion. 1. The Williams opinion. In Williams, 132 S. Ct. at 2229, the defendant was accused of abducting, raping, and robbing the alleged victim. The doctors who treated the victim took vaginal swabs for a sexual assault kit; a police detective collected the kit, and sent it to the Illinois State police laboratory. Id. The State police laboratory sent the swabs to Cellmark in Maryland for DNA testing, and Cellmark “sent back a report containing a male DNA profile produced from semen taken from those swabs.” Id. A forensic specialist at the State police laboratory then conducted a computer search to determine if the *395Cellmark profile matched any entries in the State DNA database. Id. The search revealed a match with a DNA profile produced by the State police laboratory from a sample of the defendant’s blood taken after he had been arrested on unrelated charges. Id.
At the bench trial of the defendant in Williams, the prosecutor called a forensic analyst at the State police laboratory, who testified that she developed a DNA profile from a known blood sample of the defendant. Id. The prosecutor did not call as a witness any employee of Cellmark who participated in developing the DNA profile taken from the semen on the vaginal swabs, or who knew the procedures and protocols used by Cellmark in developing a DNA profile. Id. at 2229-2230. Instead, the prosecutor called an expert witness in forensic biology and DNA analysis from the State police laboratory who “admitted she had not seen any of the calibrations or work that Cellmark had done in deducing a male DNA profile from the vaginal swabs,” but trusted Cellmark to do reliable work because it was an accredited laboratory. Id. at 2230. Over objection, this expert testified, based on her comparison of the two DNA profiles, that there was a “match” between the DNA profile found in semen from the vaginal swabs taken from the victim and the known DNA profile of the defendant, and that the probability of a random match was one in 8.7 quadrillion in the black population, one in 390 quadrillion in the white population, and one in 109 quadrillion in the Hispanic population. Id. at 2230, 2236. The Cellmark report was neither admitted in evidence nor shown or read to the judge; moreover, it was not identified as a source of any of the expert’s opinions. Id. at 2230.
In considering the admissibility of the expert’s testimony, five of the nine Supreme Court Justices concluded in Williams that the judge did not err in admitting this evidence, with the reasoning of the plurality opinion of Justice Alito, joined by three other Justices, differing from that of the concurring opinion authored by Justice Thomas. See id. at 2244; id. at 2252 (Breyer, J., concurring); id. at 2255 (Thomas, J., concurring in the judgment). The plurality opinion concluded that the evidence was admissible on two independent grounds. First, the plurality determined that the expert witness’s statement that the DNA profile generated by Cellmark that matched the defendant’s profile was “found in *396semen from the vaginal swabs” was not offered for the truth of the matter asserted. Id. at 2236-2237. Rather, the plurality stated that the expert witness “simply assumed” this to be true when she offered her opinion that there was a match between the two DNA profiles, and that the judge, who was the finder of fact at the bench trial, understood it to be nothing more than an assumption. Id. Because the confrontation clause of the Sixth Amendment to the United States Constitution does not bar the admission of testimonial statements unless they are offered for the truth of the matter asserted, the plurality concluded that there was no confrontation clause violation. Id. at 2235, 2240, quoting Crawford v. Washington, 541 U.S. 36, 59-60 n.9 (2004).
Second, the plurality concluded that even if the expert’s statement were admitted for its truth, there still would be no confrontation clause violation because Cellmark’s DNA report regarding the vaginal swab “plainly was not prepared for the primary purpose of accusing a targeted individual.” Williams, 132 S. Ct. at 2243. The plurality reasoned:
“Here, the primary purpose of the Cellmark report, viewed objectively, was not to accuse [the defendant] or to create evidence for use at trial. When the [Illinois State police laboratory] sent the sample to Cellmark, its primary purpose was to catch a dangerous rapist who was still at large, not to obtain evidence for use against [the defendant], who was neither in custody nor under suspicion at that time. Similarly, no one at Cellmark could have possibly known that the profile that it produced would turn out to inculpate [the defendant] — or for that matter, anyone else whose DNA profile was in a law enforcement database. Under these circumstances, there was no ‘prospect of fabrication’ and no incentive to produce anything other than a scientifically sound and reliable profile.”
Id. at 2243-2244, quoting Michigan v. Bryant, 131 S. Ct. 1143, 1157 (2011).
Neither of these independent grounds, however, was adopted by a fifth Justice. Justice Thomas rejected the plurality’s contention that the statement regarding Cellmark testing was not admitted for its truth, declaring that “there was no plausible reason for the introduction of Cellmark’s statements other than to *397establish their truth.” Williams, 132 S. Ct. at 2256 (Thomas, J., concurring in the judgment). He also rejected the plurality’s “primary purpose test” as lacking “any grounding in constitutional text, in history, or in logic.” Id. at 2262 (Thomas, J., concurring in the judgment). He concurred in the judgment “solely because Cellmark’s statements lacked the requisite ‘formality and solemnity’ to be considered ‘testimonial’ for the purposes of the Confrontation Clause.” Id. at 2255 (Thomas, J., concurring in the judgment).
The dissenting opinion of Justice Kagan, joined by three other Justices, concluded that the expert witness’s testimony that the victim’s vaginal swab contained DNA that matched the defendant’s DNA profile was offered for its truth and violated the confrontation clause, id. at 2270-2272 (Kagan, J., dissenting), noting that “the prosecution introduced the results of Cell-mark’s testing through an expert witness who had no idea how they were generated.” Id. at 2265 (Kagan, J., dissenting). The dissent also rejected the plurality’s “primary purpose test.” Id. at 2273 (Kagan, J., dissenting). The dissent asserted that, “in all except its disposition, [the plurality] opinion is a dissent: Five Justices specifically reject every aspect of its reasoning and every paragraph of its explication.” Id. at 2265 (Kagan, J., dissenting). The dissent also rejected Justice Thomas’s argument that the Cellmark report lacked the “formality and solemnity” to be “testimonial.” Id. at 2275-2277 (Kagan, J., dissenting). According to the dissent, the consequence of the Court’s judgment reached by five Justices is to leave “significant confusion in [its] wake.”1 Id. at 2277 (Kagan, J., dissenting).
2. Application of the Williams opinion to this case. We dis*398agree with the Appeals Court’s conclusion that the facts in the instant case “are not materially distinguishable from those presented to the Supreme Court in Williams.” Tassone, 83 Mass. App. Ct. at 199. Unlike in Williams, the fact finder in this case was a jury, not a judge, and the judge provided no limiting instructions to the jury regarding their consideration of Roy’s testimony. The plurality in Williams acknowledged that:
“The dissent’s argument would have force if [the defendant] had elected to have a jury trial. In that event, there would have been a danger of the jury’s taking [the expert witness’s] testimony as proof that the Cellmark profile was derived from the sample obtained from the victim’s vaginal swabs. Absent an evaluation of the risk of juror confusion and careful jury instructions, the testimony could not have gone to the jury.”
Williams, 132 S. Ct. at 2236. Here, the jury likely considered Roy’s testimony — that the DNA profile that matched the defendant’s known DNA profile was obtained by Cellmark from the swab of the eyeglasses — for its truth, rather than as an assumption applied to reach the expert’s conclusion. In addition, in contrast with Williams, the “primary purpose” of the Cell-mark report in the instant case was to accuse the defendant and create evidence for use at trial, because the DNA testing at Cellmark was conducted after Brodeur had identified the defendant from a photographic array as the person who committed the robbery. Where neither of the two independent grounds relied on by the plurality would apply to the facts in the instant case, we do not agree with the Appeals Court that the admissibility of *399Roy’s testimony “must follow from the judgment in that case.” Tassone, supra at 199. The more challenging question, given the “significant confusion” that has been left in the wake of the Williams decision, 132 S. Ct. at 2277 (Kagan, J., dissenting), is whether the United States Supreme Court would conclude that evidence of this type, admitted under these circumstances, would violate the confrontation clause. Fortunately, we need not resolve that question because, regardless of the answer, we conclude that Roy’s opinion was not admissible under our common law of evidence.
3. Our common law of evidence. We noted in Commonwealth v. Greineder, 464 Mass. 580, 593, cert. denied, 134 S. Ct. 166 (2013), that “our evidentiary rules afford a defendant more protection than the Sixth Amendment.” Our common law of evidence is more protective of confrontation rights in two ways. First, although the Sixth Amendment’s confrontation clause and Fed. R. Evid. 703 permit the prosecution to elicit on direct examination from an expert witness the underlying facts or data on which the expert’s opinion is based for the limited purpose of explaining the basis for the expert’s opinion, our common-law rules of evidence do not. Greineder, supra at 592-594. See Commonwealth v. Barbosa, 457 Mass. 773, 784-785 (2010), cert. denied, 131 S. Ct. 2441 (2011). Our evidentiary rules permit the facts or data underlying the opinion to be elicited only by the defendant on cross-examination and, where this door has been opened by the defendant, by the prosecution on redirect examination. Greineder, supra at 583-584. See Mass. G. Evid. § 705 (2014).
Second, our common law of evidence requires that the defendant have a meaningful opportunity to cross-examine the expert about her opinion and the reliability of the facts or data that underlie her opinion. See Greineder, 464 Mass. at 595; Barbosa, 457 Mass. at 786. Where the admissibility of expert testimony is challenged, the judge as gatekeeper must determine, among other foundational requirements, whether “the expert’s opinion is based on facts or data of a type reasonably relied on by experts to form opinions in the relevant field.” Barbosa, supra at 783, citing Department of Youth Servs. v. A Juvenile, 398 Mass. 516, 531 (1986). Where the gatekeeping judge *400determines that the expert’s opinion meets a minimum standard of reliability and admits the evidence, a defendant is entitled to a meaningful opportunity to cross-examine the expert as to the reliability of the underlying facts or data. See Greineder, supra.
Where an expert in a homicide case offers an opinion regarding the cause of death, a defendant will generally have a meaningful opportunity to cross-examine the expert witness regarding possible flaws in the opinion based on the underlying autopsy report and notes, and photographs taken during the autopsy, regardless of whether the expert conducted the autopsy or was employed in the medical examiner’s office that did. See, e.g., Commonwealth v. Nardi, 452 Mass. 379, 383, 390-391 (2008). But “with DNA analysis, the testing techniques are so reliable and the science so sound that fraud and errors in labeling or handling may be the only reasons why an opinion is flawed” (emphasis in original). Barbosa, 457 Mass. at 790, citing National Research Council, Strengthening Forensic Science in the United States: A Path Forward 130 (2009). Therefore, where a DNA expert offers an opinion regarding a DNA match, a meaningful opportunity for cross-examination means that a defendant must have the opportunity substantively to explore the “risk of evidence being mishandled or mislabeled, or of data being fabricated or manipulated, and . . . whether the expert’s opinion is vulnerable to these risks.” Barbosa, supra at 791.
In Greineder, 464 Mass. at 582, 598, we concluded that the defendant had a meaningful opportunity to cross-examine the prosecution’s DNA expert about the reliability of the data that formed the basis for her opinion that the defendant’s DNA profile matched the DNA profile found on a knife and gloves recovered from the crime scene, even though the expert did not personally conduct the DNA testing. But in that case the prosecution’s expert was the forensic laboratory director of Cellmark, where the DNA testing was conducted, and “was in a unique position to speak to the DNA testing process employed by Cellmark, including the chain of custody and evidence-handling protocols.” Id. at 596-597 (expert was “in a better position than an individual analyst to testify to the reliability and accuracy of Cellmark’s testing procedures”).
Similarly, in Barbosa, 457 Mass. at 779, 786, where the *401prosecution’s DNA expert testified that the victim was “included as being the possible source of the DNA extracted” from bloodstains on the defendant’s boot and pants even though another analyst in her laboratory had conducted the DNA testing, we concluded that “the defendant’s right of confrontation was not violated by the admission of [the expert’s] opinion, because he had a fair opportunity to confront [the expert] as to the reasonable basis for that opinion.” Id. at 786. There, the prosecution’s expert was knowledgeable of the protocols used within the laboratory and of the testing analyst’s work; the expert supervised and trained the analyst who conducted the DNA testing, reviewed the analyst’s work by looking at the worksheets and reports that the analyst generated during the testing process, and signed the analyst’s final report regarding the testing she performed after a full technical review. Id. at 782, 791. In sum, although the expert “did not have personal knowledge of how [the analyst] conducted the testing, she was familiar with the protocols [the analyst] was taught to employ, [the analyst’s] documentation of her work, and [the analyst’s] work history.” Id. at 791.
Here, in contrast, at the hearing on the defendant’s motion in limine where the judge denied the defendant’s challenge to the admissibility of Roy’s opinion, Roy testified that she was a chemist at the State police laboratory and that Cellmark, not her laboratory, had generated the DNA profile from the eyeglasses found at the scene of the crime. Roy was able to confirm from the electropherogram sent by Cellmark that it reflected the DNA profile reported by Cellmark, but she was not in a position to confirm that the DNA profile was from the eyeglasses swab; she knew only that Cellmark said that it was. Further, there was no evidence that Roy previously had been employed by Cellmark, had any personal knowledge of the chain-of-custody or evidence-handling protocols used by Cellmark, or had seen any of the worksheets generated by Cellmark during the testing.
Under these circumstances, we conclude that the defendant did not have a meaningful opportunity to cross-examine Roy regarding the laboratory work, procedures, or protocols performed by Cellmark, or as to the reliability of the Cellmark data on which her opinion of a “match” with the defendant’s *402DNA profile rests. Where an opinion that DNA found at the crime scene matches the defendant’s DNA profile is offered by an expert who is not affiliated with the laboratory where the testing of the crime scene DNA sample was conducted, and where no analyst from that laboratory testifies regarding that testing, a defendant effectively is denied the opportunity to explore through cross-examination whether the opinion is flawed. Because the defendant here had no meaningful opportunity for cross-examination, the admission of Roy’s opinion violated the right to confrontation provided by our common law of evidence. See Greineder, 464 Mass. at 595; Barbosa, 457 Mass. at 783-784.
The prosecution may not admit powerful evidence of a DNA match against a defendant and deny the defendant a meaningful opportunity to challenge the reliability of the facts or data on which the opinion rests by failing to call an expert witness affiliated with the laboratory that tested the sample connected to the crime scene. We know of no case before Williams where the prosecution attempted to offer an opinion of a DNA profile match without calling the DNA analyst who conducted the testing of the crime scene DNA or a knowledgeable employee from that laboratory. At least in State courts in Massachusetts, the Williams opinion does not open the door to such a practice. Regardless of whether the Supreme Court ultimately interprets the confrontation clause to permit the admission of such an opinion under circumstances that effectively deny the defendant any meaningful opportunity for cross-examination, its admission in our courts is barred by the right of confrontation in our common law of evidence. In other words, if the Commonwealth sends the crime scene DNA to Cellmark for analysis, and seeks to offer in evidence an opinion that the crime scene DNA matches the DNA of the defendant, it will need, at a minimum, to call an expert witness from Cellmark.2
4. Prejudicial error. The defendant objected to the admission *403of Roy’s opinion on confrontation grounds by filing a motion in limine. After denying the motion, the judge agreed that the defendant’s rights were saved regarding the objection. See Commonwealth v. Aviles, 461 Mass. 60, 66 (2011) (defendant relieved of need to object to evidence at trial where judge, after denying motion in limine, tells defendant that his rights are saved). Because we conclude that the admission of the opinion under these circumstances violated our common law of evidence, we must determine whether the error was nonprejudicial, that is whether “the error did not influence the jury, or had but very slight effect.” Commonwealth v. Christian, 430 Mass. 552, 563 (2000), quoting Commonwealth v. Flebotte, 417 Mass. 348, 353 (1994). Although we recognize that the non-DNA evidence against the defendant was strong, we cannot say with confidence that the admission of Roy’s opinion had “but very slight effect” on the jury. The primary issue at trial was identification and there was evidence that the defendant’s brother had the same height and body shape as the defendant, wore eyeglasses “just like” those of the defendant, and had a drug problem similar to that of the defendant.
Apart from the DNA evidence, the most compelling evidence against the defendant was the eyewitness identification by Bro*404deur, the admissions the defendant made to Streit, and the eyeglasses left at the scene. But without the DNA evidence that connected the defendant to the eyeglasses found at the scene, the jury might have questioned whether Brodeur could have reliably identified the defendant rather than his brother, whether Streit’s testimony was reliable because their engagement had ended, and whether the eyeglasses belonged to the defendant rather than his brother. The DNA evidence not only tied the defendant to the eyeglasses, but also likely diminished any doubts the jury might have had as to the reliability of Brodeur’s identification and Streit’s testimony regarding the defendant’s admissions. In these circumstances, we conclude that the admission of Roy’s opinion regarding the DNA evidence was prejudicial error.3
Conclusion. We reverse and vacate the defendant’s convictions, and remand for a new trial.
So ordered.
The dissent’s concern was well founded. “When a fragmented Court decides a case and no single rationale explaining the result enjoys the assent of five Justices, ‘the holding of the Court may be viewed as that position taken by those Members who concurred in the judgments on the narrowest grounds ....’” Marks v. United States, 430 U.S. 188, 193 (1977), quoting Gregg v. Georgia, 428 U.S. 153, 169 n.15 (1976). See McClure v. Secretary of the Commonwealth, 436 Mass. 614, 626 n.15 (2002) (applying rule of Marks). In the wake of the fractured plurality decision in Williams v. Illinois, 132 S. Ct. 2221 (2012), courts have struggled to find any such common denominator that may constitute its holding. See, e.g., United States v. James, 712 F.3d 79, 95-96 (2d Cir. 2013) (“Williams does not, as far as we can determine, using the Marks analytic approach, yield a single, useful holding *398relevant to the case before us. It is therefore for our purposes confined to the particular set of facts presented in that case”). See also United States v. Duron-Caldera, 737 F.3d 988, 994 n.4 (5th Cir. 2013) (“In Williams, there is no such common denominator between the plurality opinion and Justice Thomas’s concurring opinion. Neither of these opinions can be viewed as a logical subset of the other”). Unable to extract a governing standard from Williams, courts facing similar confrontation clause questions have applied both the plurality’s “primary purpose” and Justice Thomas’s “formality and solemnity” tests to the facts before it. See, e.g., State v. Deadwiller, 350 Wis. 2d 138, 160-165 (2013); State v. Medina, 232 Ariz. 391, 406-407 (2013), cert. denied, 134 S. Ct. 1309 (2014).
We also note that the opinion offered by Tiffany Roy — “[t]hat the [deoxyribonucleic acid (DNA)] profile generated from the known saliva sample from [the defendant] matched the major male profile obtained from the swab of the eyeglasses” — was not accompanied by an opinion regarding the mathematical probability that another person possessed the same DNA profile. We stated in Commonwealth v. Barbosa, 457 Mass. 773, 789 (2010), cert. *403denied, 131 S. Ct. 2441 (2011), quoting Human Genome Program, U.S. Dep’t of Energy, Genomics and its Impact on Science and Society: A 2008 Primer 3 (2008), that, where “[t]he human genome sequence is almost exactly the same (99.9%) in all people,” a match of the defendant’s DNA profile with the DNA found at a crime scene “says almost nothing about the likelihood that the defendant was present at the crime scene unless the jury learn from an expert” the mathematical probability that another person has this same DNA profile. Barbosa, supra. Roy’s opinion should not have been admitted without evidence regarding the estimated frequency of occurrence of the DNA profile. See Commonwealth v. Mattei, 455 Mass. 840, 842 (2010) (“expert testimony that DNA tests could not exclude the defendant as a potential source of DNA found at the crime scene, absent testimony regarding statistical findings explaining the import of such a result, was likely to confuse and mislead the jury such that any probative value of the test results was substantially outweighed by their prejudicial effect”). However, we conclude that defense counsel did not provide ineffective assistance for failure to object to the admission of the opinion on this ground, because the Cellmark report (which was not admitted in evidence) estimated the frequency of occurrence of the DNA profile among various populations as less than one in 1 quadrillion unrelated individuals, and Roy likely would have independently been able to reach this same conclusion.
Because we conclude that the admission of Roy’s opinion was prejudicial, we need not determine whether its admission violated the right of confrontation in art. 12 of the Massachusetts Declaration of Rights, and therefore should be reviewed under the standard applied to constitutional error: harmless beyond a reasonable doubt. See Commonwealth v. Vinnie, 428 Mass. 161, 163 (1998), quoting Chapman v. California, 386 U.S. 18, 23 (1967). Before the issuance of the Supreme Court’s opinion in Williams v. Illinois, 132 S. Ct. 2221 (2012), we declared that “the protection provided by art. 12 is coextensive with the guarantees of the Sixth Amendment [to the United States Constitution].” Barbosa, 457 Mass. at 780 n.7, quoting Commonwealth v. Nardi, 452 Mass. 379, 388 n.10 (2008). We leave open the question whether that remains true after Williams.