Judge RYAN delivered the opinion of the Court.
Appellee was convicted, contrary to his pleas, by a general court-martial composed of officer and enlisted members of one specification of aggravated sexual assault, one specification of burglary, and one specification of unlawful entry, in violation of Articles 120, 129, and 134, UCMJ, 10 U.S.C. §§ 920, 929, 934, United States v. Katso, 73 M.J. 630, 632 (A.F.Ct.Crim.App.2014). Appellee was sentenced to confinement for ten years, a dishonorable discharge, and forfeiture of all pay and'allowances. Id, The convening authority approved the sentence. Id. The United States Air Force Court of Criminal Appeals (CCA) set aside and dismissed the findings and sentence, holding that the testimony of a Government expert witness was based on a testimonial report written by an out-of-court declarant, thereby violating Ap-pellee’s right to confrontation under the Sixth Amendment of the United States Constitution. Id. at 638-40, 642.
The Judge Advocate General of the Air Force certified the following issue to this Court:
WHETHER THE AIR FORCE COURT OF' CRIMINAL APPEALS ERRED WHEN IT FOUND APPELLEE’S SIXTH AMENDMENT RIGHT TO CONFRONTATION WAS VIOLATED WHEN THE MILITARY JUDGE PERMITTED, [275]*275OVER DEFENSE OBJECTION, THE TESTIMONY OF THE GOVERNMENT’S DNA EXPERT, AND THAT THE ERROR WAS NOT HARMLESS.
This ease requires us to examine the application of the Sixth Amendment to testimony relating the results of forensic analysis that was the product of collaboration among a number of laboratory employees. When an expert’s knowledge and opinions are based in part on tests performed by others, what may the expert tell the factfinder without violating the defendant’s right to confrontation? To answer this question, we apply the frameworks developed by the Supreme Court and by this Court to a set of facts that neither court has considered.
We hold that the testimony of the Government’s forensic expert witness, David Davenport, did not violate Appellee’s right to confrontation. Unlike the experts in Bullcoming v. New Mexico, — U.S. -, 131 S.Ct. 2705, 2715, 180 L.Ed.2d 610 (2011), and United States v. Blazier (Blazier II), 69 M.J. 218, 226 (C.A.A.F.2010), Mr. Davenport’s personal knowledge regarding the derivation of the evidence at issue made him neither a “surrogate” expert, Bullcoming, 131 S.Ct. at 2715, nor a mere “conduit” for the testimonial statements of another. Blazier II, 69 M.J. at 225; see also Williams v. Illinois, — U.S. -, 132 S.Ct. 2221, 2241, 183 L.Ed.2d 89 (2012). Mr. Davenport conducted a thorough review of the entire case file, including the documents submitted with the evidence, the tests performed on the eviden-tiary samples, and -the quality control measures. He personally compared the DNA profiles from the evidentiary samples to the DNA profiles from the known samples, reran the statistical analysis, and formulated his own carefully considered conclusions. Much of the data underlying his opinion was not testimonial, and, assuming arguendo that the report prepared for his technical review was testimonial, Mr. Davenport did not act as a mere conduit for the report. See Memorandum from Robert Fisher, Forensic DNA Examiner, to Commander; Air Force Office of Special Investigations, Detachment 320 (Jan. 28, 2011) [hereinafter Final Report]. The military judge’s denial of Appellee’s motion to exclude the expert’s testimony was not an abuse of discretion, and the decision of the CCA is reversed.
I. FACTS
A Collecting and Analyzing the DNA Evidence
On the morning of December 11, 2010, Senior Airman (SrA) CA reported that she had been raped, and identified Appellee as the perpetrator. Agents from the Air Force Office of Special Investigations (AFOSI) promptly brought SrA CA to the hospital for an examination. A Sexual Assault Nurse Examiner (SANE) testified that she collected, among other items, vaginal, oral, and rectal swabs from SrA CA a blood sample, and debris from SrA CA’s clothing. The nurse examiner handed these samples to an AFOSI agent. Another SANE testified that she collected Appellee’s blood and saliva, obtained penile and scrotal swabs, and handed the samples to an AFOSI agent.'
AFOSI Special Agent (SA) Richard Blair testified at trial that he received the samples from the two agents who had been at the hospital. The samples were combined into two separate “sexual assault kits,” containing samples from Appellee and SrA CA respectively. SA Blair explained that the agents who received the samples prepared a set of documents to accompany each kit, which SA Blair reviewed. SA Blair sent this evidence and documents to the United States Army Criminal Investigation Laboratory (USA-CIL), enclosing a request form that described each piece of evidence and listed identifying numbers for the evidence.1
[276]*276B. Mr. Davenport’s Testimony on the Motion to Suppress
Robert Fisher, the USACIL employee responsible for the initial analysis of the sexual assault kits, was in Florida during the court-martial to be by his mother’s side while she underwent major surgery. The Government notified the defense that it would elicit testimony about the forensic analysis from Mr. Davenport, who conducted the technical review of Mr. Fisher’s analysis. At the time of the trial, Mr. Davenport had worked as a forensic DNA examiner at USACIL for more than six years. Appellee made a motion in limine to exclude Mr. Davenport’s testimony, arguing that such testimony would violate his right to confrontation. In an Article 39(a), UCMJ, 10 U.S.C. § 839(a) (2012), session, Mr. Davenport testified both about the steps that USACIL technicians follow to process evidence in cases of alleged sexual assault and his own role in reviewing and testing the evidence in this case.
i. USACIL’s Procedure for Processing Evidence
Testifying during the healing on the motion to suppress, Mr. Davenport described the path of sexual assault kits through USA-CIL. First, employees in the evidence processing section receive the evidence. They then scan and save an electronic copy of the forms accompanying the evidence into a “case file” on the laboratory’s computer system. The “case examiner” checks the evidence out of the evidence processing section, breaks the seal on the evidence, and checks the forms that accompanied the package against the evidence to ensure that the lab received all items reflected therein. The examiner then inventories the evidence, verifies that the evidence was properly sealed, and notes any irregularities with the evidence. The examiner is required to document any accidents or mistakes that occurred during the tests.
The case examiner performs a serological examination, which entails looking for traces of semen on the evidence collected from the victim. Various steps of this exam require the examiner to record his visual observations.2 The examiner then creates DNA profiles from the “evidentiary samples” — the samples identified by the serology exam and the swabs collected during the SANE’s examination of the suspect — and creates DNA profiles from the blood or saliva of the parties, the “known samples.” To create a DNA profile, the examiner must “purify” the DNA, extracting it from the sample; determine the quantity and type of DNA present using an instrument that generates a computer printout; copy portions of the DNA; and use another instrument to produce the machine-generated data that comprises the DNA profile.
The examiner compares the DNA profiles from the evidentiary samples to the DNA profiles from the known samples. Based on this comparison, the examiner determines whether any sample identified by the serology exam contains a DNA profile matching the DNA from the known sample of the suspect, and whether any swab collected from the suspect contains a DNA profile matching DNA from the known sample of the victim. For each match, the examiner calculates the probability that the DNA profile on the evidentiary sample would match an unrelated individual selected at random from the American population. The examiner then drafts a report summarizing these results.
[277]*277ii. Mr. Davenport’s Technical Review
As the “technical reviewer,” Mr. Davenport was required to verify Mr. Fisher’s results and approve the Final Report, which Mr. Fisher drafted. In the Article 39(a) session, Mr. Davenport outlined the steps he took to review Mr. Fisher’s work. According to Mr. Davenport, his review focused on the items in the ease file, which contained, inter alia, the request for analysis and forms submitted by AFOSI, Mr. Fisher’s handwritten notes, records of the quality control measures used during testing, all printouts generated during the testing process,' the raw DNA profile data, and the Final Report. Mr. Davenport neither handled the initial evidence submitted nor observed Mr. Fisher’s testing procedures.
Mr. Davenport verified that Mr. Fisher followed protocol and properly documented each step, and that the protocol utilized by USACIL is widely accepted in the field of forensic DNA analysis. Mr. Davenport testified that USACIL procedures require many quality control measures, such as running positive and negative controls, recording the lot numbers of the chemicals used, processing an unrelated known DNA sample along with the samples at issue, and processing test tubes that contain reagent but not DNA Mr. Davenport testified that he was able to determine, based on the contents of the case file, that Mr. Fisher took these measures.
Mr. Davenport explained that, since Mr. Fisher was required to document accidents or mistakes, he could determine whether any accidents occurred or whether Mr. Fisher made any mistakes by reviewing Mr. Fisher’s notes. Additionally, he would have been able to catch undocumented mistakes by checking for irregularities in the results. For example, Mr. Davenport checked that the quality control sample produced the expected results and that known samples produced correctly gendered profiles. Logically inconsistent results — such as a complete male profile in non-semen DNA taken from the victim or a complete female profile generated from DNA supposedly extracted from semen — could signify a mix-up. Additionally, the presence of a DNA profile n'ot matching a known sample could indicate contamination. These objective measures enable the technical reviewer to determine that cross-contamination did not occur.
Mr. Davenport independently compared the DNA profiles of the evidentiary and known samples to verify the matches. This involved processing the machine-generated raw profile data using a computer program and interpreting the profiles to detect matches between the samples.. Mr. Davenport then recalculated the probability of a match between the DNA profiles for each matching evidentiary sample and an individual selected at random from the American population. Based on review and interpretation of all of the above, Mr. Davenport determined that he agreed with Mr. Fisher’s results and initialed the Final Report, allowing the report to progress to the next stage of the review process.
iii. Military Judge’s Ruling on Motion to Exclude Testimony
Relying on Mr. Davenport’s’motions testimony detailing his knowledge of and involvement in the testing process, the military judge concluded that “Mr. Davenport’s opinions will be based on his training and experience, and his review of the entire case,” and denied Appellee’s motion. The military judge found that Mr. Davenport analyzed the raw data “to ensure he reached the same results and conclusions as Mr. Fisher had,” and noted that Mr. Davenport explained he had “review[ed] the case from beginning to end.” The ruling permitted Mr. Davenport to state his “opinion concerning the reliability of testing procedures used in this case, the findings/results in this case and the frequency statistics” only “[s]o long as Mr. Davenport does not become a conduit of inadmissible testimonial hearsay.”
C. Mr. Davenport’s Courts Martial Testimony
Mr. Davenport testified before members as an expert, providing his independent opinion on the results of the DNA analysis. Mr. Davenport told members that, as the technical reviewer, he based his opinions on his review of the case file. The Final Report was not admitted into evidence. Mr. Daven[278]*278port only referenced the Final Report to note that he reviewed Mr. Fisher’s interpretation of the results and checked the Final Report against the documents submitted by AFOSI to make sure that the report properly listed and identified the items submitted as evidence.3 Based on his review of the entire case file, Mr. Davenport testified that:
1. The evidence collected from SrA CA and Appellee was tested “per protocol,”
2. The evidence was received in a sealed condition,
3. The evidence was inventoried properly,
4. The known samples were analyzed properly,
5. DNA profiles were generated “from the known blood of [SrA CA] and [Ap-pellee],”
6. The swabs collected from SrA CA contained semen,
7. DNA consistent with SrA CA and Ap-pellee was found on the rectal swabs from SrA CA,
8. Unidentifiable male DNA was found on SrA CA’s vaginal swab, and
9. DNA consistent with SrA CA and Ap-pellee was found on Appellee’s penile and scrotal swabs.
Mr. Davenport also testified to the likelihood that the recovered DNA profiles would match other individuals. On cross-examination, defense counsel clarified briefly that Mr. Davenport did not conduct the initial tests that produced the DNA profiles. Defense counsel successfully got Mr. Davenport to concede that the DNA analysis did not reveal anything about the nature of the sexual contact.
II.CCA DECISION
The CCA’s analysis focused on the relationship between Mr. Davenport’s trial testimony and the Final Report, but did not directly address either the testimony on the motion in limine or the military judge’s findings or conclusions on that motion. Katso, 73 M.J. at 637-40. The CCA held that the Final Report was testimonial, and determined that Mr. Davenport’s trial testimony improperly repeated information from the report. Id. at 638-40. Specifically, it held that Mr. Davenport repeated testimonial hearsay when he identified Appellee as the source of the DNA found on samples collected from SrA CA, acting as a “conduit” for this information. Id. at 639-40 (quoting Mazier II, 69 M.J. at 225). The CCA found as a matter of fact that “the record of trial does not definitively establish that Mr. Davenport had first-hand knowledge as to whom the known DNA sample or its corresponding profile belonged” and was “able to identify [Appellee] by name only by repeating the testimonial statement contained in Mr. Fisher’s report that directly linked [Appellee] to the generated DNA profile.” Id. at 638-39. Because the Government “failed to demonstrate that the DNA evidence played an insignificant role” in the case, the CCA was not convinced beyond a reasonable doubt that Mi’. Davenport’s presentation of the evidence was harmless. Id. at 641.
III.DISCUSSION
The Sixth Amendment prohibits the admission of “testimonial statements of a witness who did not appear at trial,” unless the witness is “unavailable to testify, and the defendant had had a prior opportunity for cross-examination.” Crawford v. Washington, 541 U.S. 36, 53-54, 124 S.Ct. 1354, 158 L.Ed.2d 177 (2004). Whether evidence is testimonial hearsay is a question of law reviewed de novo. United States v. Tearman, 12. M.J. 54, 58 (C.A.A.F.2013). We review the military judge’s ruling on a motion to exclude evidence for an abuse of discretion, “eonsiderfing] the evidence in the light most favorable to the prevailing party.”4 Reister, [279]*27944 M.J. at 413 (internal quotation marks omitted). Additionally, this Court will not overturn- the CCA’s factual findings unless they are clearly erroneous or unsupported by the record. United States v. Teffeau, 58 M.J. 62, 66-67 (C.A.A.F.2003) (citing United States v. Tollinchi, 54 M.J. 80, 82 (C.A.A.F.2000)).
A
“[T]he principal evil at which the Confrontation Clause was directed was the civil-law mode of criminal procedure, and particularly its use of ex parte examinations as evidence against the accused.” Crawford, 541 U.S. at 50, 124 S.Ct. 1354. The Confrontation Clause thus protects defendants by excluding the introduction of “hearsay” that is “testimonial,” the equivalent of an ex parte examination. Id. at 51, 124 S.Ct. 1354. Kel-evant to this ease, determining whether an expert witness’s testimony has violated the Confrontation Clause requires asking two questions: First, did the expert’s testimony rely in some way on out-of-court statements that were themselves testimonial? Id. at 51-52, 124 S.Ct. 1354. Second, if so, was the expert’s testimony nonetheless admissible because he reached his own conclusions based on knowledge of the underlying data and facts, such that the expert himself, not the outiof-court declarant, was the “witness[ ] against [Appellee]” under the Sixth Amendment? U.S. Const, amend. VI; see Blazier II, 69 M.J. at 224-25; Crawford, 541 U.S. at 44, 124 S.Ct. 1354. We turn to these two questions in order.5
B.
This Court has already delineated the boundary between testimonial and nontesti-monial statements in detail. “[A] statement is testimonial if ‘made under circumstances which would lead an objective witness reasonably to believe that the statement would be available for use at a later trial.’ ” United States v. Sweeney, 70 M.J. 296, 301 (C.A.A.F.2011) (quoting United States v. Blazier (Blazier I), 68 M.J. 439, 442 (C.A.A.F.2010)); see Crawford, 541 U.S. at 51-52, 124 S.Ct. 1354; Tearman, 72 M.J. at 58. In making this determination, this Court has asked whether it would “be reasonably foreseeable to an objective person that the purpose of any individual statement ... is evidentiary,” considering the formality of the statement as well as the knowledge of the declarant. Tearman, 72 M.J. at 58 (citation and internal quotation marks omitted); compare id., 72 M.J. at 59-61 (chain of custody documents and internal review worksheets were not testimonial, in part because an objective witness would reasonably believe that the documents were filled out for “internal control, not to create evidence” and because they “lack[ed] any indicia of formality or solemnity”), with Sweeney, 70 M.J. at 299, 304 (a signed memorandum reporting the results of a drug and a signed, “formal, affidavit-like” document certifying the integrity of the sample and compliance with protocol were testimonial), and Blazier I, 68 M.J. at 440, 443 (signed declarations served an “evidentiary purpose” because they “summarize[d] and clearly set forth [an] ‘accusation,’ ” and were generated in response to a command request).
As detailed more fully below, many of the out-of-court data and “statements” relied upon by Mr. Davenport in reaching his conclusion were not testimonial. The case file— containing the AFOSI documents, the computer-generated raw data, and Mr. Fisher’s handwritten notes, including his documentation of the conditions of the samples upon arrival and quality control measures — is not in the record. However, Mr. Davenport’s testimony makes clear that he reviewed all [280]*280such documents. Nothing suggests that the AFOSI documents, which appear to primarily serve a chain of custody function, see supra note 1, were testimonial, or that the computer-generated raw data was either a statement or testimonial. See Teaman, 72 M.J. at 61; Sweeney, 70 M.J. at 305. Nor is there any indication that Mr. Fisher’s notes or his other lab results that underlay the Final Report were signed, certified anything, bore indicia of formality, or that Mr. Fisher expected them to be used at trial.
Moreover, the CCA’s “finding” that Mr. Davenport was “able to identify [Appellee] by name only by repeating the testimonial statement contained in Mr. Fisher’s report that directly linked [Appellee] to the generated DNA profile,” Katso, 73 M.J. at 639 (emphasis added), was clearly erroneous and is unsupported by the record. See Teffeau, 58 M.J. at 66-67; Kitts, 43 M.J. at 28; see also United States v. Piolunek, 74 M.J. 107, 110 n. 3 (C.A.A.F.2015). Rather, the record indicates that Mr. Davenport learned the names of the parties the same way Mr. Fisher did— through the underlying data in the case file, including the forms submitted by AFOSI. See supra note 1. The Confrontation Clause does not require either Mr. Fisher or Mr. Davenport to personally shadow the evidence from its collection to USACIL in order to opine that it is what it purports to be and was collected from the persons indicated on the forms. Furthermore, both Mr. Fisher and Mr. Davenport were in a position to testify as to whether lab procedures were followed in this regard.
In sum, Mr. Davenport’s statements regarding proper testing, receipt, inventory, and analysis of the evidence, as well as his identification of the parties, relied on nontes-timonial items in the ease file.6 This testimony was therefore admissible.
The Final Report, which served as a consolidated and conelusory summary of Mr. Fisher’s analysis, presents a more complicated problem, since, while it does not contain a formal certification, the record indicates that Mr. Fisher knew Appellee was a suspect in a sexual assault and that the Final Report would be “made official and sent to the agent in the case.” This problem is best resolved by assuming, arguendo, that the Final Report itself was testimonial, though based on evidence that was not testimonial, and that Mr. Davenport’s testimony regarding the results of the serology exam and the DNA analysis may have relied in part on the Final Report (since he had to conduct the technical review of it) as well as the evidence that underlay it. From there, we proceed to the question whether, given those assumptions, Mr. Davenport’s opinion was admissible despite its partial reliance on testimonial hearsay that was not itself introduced or repeated at trial.
C.
To determine whether the portions of Mr. Davenport’s testimony that may have been based in part on testimonial hearsay should have been admitted, we apply this Court’s precedent, from Blazier I through Tearman. Under that line of cases, we ask whether Mr. Davenport had sufficient personal knowledge to reach an independent conclusion as to the object of his testimony and his expert opinion. Blazier II, 69 M.J. at 224-25. Framed another way, this Court queries whether Mr. Davenport was a “witness[ ] against” Appel-lee, the type of declarant Appellee had the constitutional right to cross-examine, or a mere “conduit” for another “witness[ ],” namely, Mr. Fisher. U.S. Const. amend. VI; Blazier II, 69 M.J. at 225; see Crawford, 541 U.S. at 44, 124 S.Ct. 1354 (recounting Sir Walter Raleigh’s demand that Lord Cobham, who had implicated Raleigh in a treason plot, be compelled to appear in person at trial to testify against him). We review Supreme Court precedent and undertake a highly fact-specific inquiry to determine that Mr. Davenport was a “witness[ ]” rather than a “conduit.”
[281]*2811,
The Supreme Court has repeatedly applied the Confrontation Clause in the context of expert forensic analysis to determine whether evidence admitted at trial repeated testimonial hearsay, but it has not faced a situation identical to the one before us. Mr. Davenport, an expert with detailed knowledge of the results he presented, delivered testimony that does not easily fit into the Supreme Court’s framework. However, reviewing those Confrontation Clause cases most directly related to the facts of this case convinces us that no Supreme Court precedent bars the application of the principles established in ¡¡lazier II or warrants concluding that Mr. Davenport’s testimony simply repeated testimonial hearsay by an out-of-court declarant. We discuss each case in turn.
In Melendez-Diaz v. Massachusetts, 557 U.S. 305, 308, 329, 129 S.Ct. 2527, 174 L.Ed.2d 314 (2009), the Supreme Court held that the petitioner’s right to confrontation was violated when the prosecution submitted “certificates of analysis” into evidence without relying on the testimony of an expert witness. In Appellee’s case, the Government did not introduce the Final Report. In Melendez-Diaz, the Supreme Court held that the analysts who prepared the certificates were “witnesses,” 557 U.S. at 311, 129 S.Ct. 2527 (internal quotation marks omitted), who, through the certificates, made only a “bare-bones statement” and did not provide the petitioner the opportunity to learn about the tests performed or the analysts’ ability to interpret those tests. Id. at 320, 129 S.Ct. 2527. In this case, Mr. Davenport’s testimony at trial and knowledge of the underlying facts provided Appellee ample opportunity to ascertain “what tests [Mr. Fisher] performed, whether those tests were routine,” and whether Mi'. Davenport had the requisite “judgment” and “skills” to interpret the results. See id.
In Bullcoming, 131 S.Ct. at 2711-12, the trial court allowed the state to introduce a certified report attesting to the petitioner’s blood alcohol content through the testimony of an expert witness who had no knowledge about the analysis at all. The Supreme Court reversed, holding that the “surrogate testimony” of the expert, “who had neither observed nor reviewed [the] analysis,” could not “convey what [the scientist] knew or observed about the events his certification concerned.” Id. at 2712, 2715. Several concerns with the testimony led the Supreme Court to conclude that the expert had provided “surrogate” testimony: the expert could not describe “the particular test and testing process [the analyst] employed,” id. at 2715; could not “expose any lapses or lies on the ... analyst’s part,” id.; nor could he explain why the analyst “had been placed on unpaid leave.” Id. Moreover, the State did not “assert that ... [the expert] had any independent opinion concerning [the results].” Id. at 2716 (internal quotation marks omitted).
Here, by contrast, no certified report was introduced; Mr. Davenport described, based on his personal knowledge, the tests and testing processes used, and the means for discerning protocol lapses. Unlike the analyst in Bullcoming, 131 S.Ct. at 2715, Mr. Fisher had not been placed on unpaid leave for unexplained reasons; rather, Mr. Fisher missed the trial in order to be with his ill mother, circumstances that indicate neither incompetence on Mr. Fisher’s part nor a ploy by the Government to gain a tactical advantage. Mr. Davenport confirmed that he had formed and was testifying to his “independent opinion,” providing opportunity for cross-examination. Moreover, Mr. Davenport’s extensive review likely places this ease well within Justice Sotomayor’s hypothetical. See id. at 2722 (Sotomayor, J., concurring in part) (the testimony might have been admissible if the expert had some “degree of involvement” in the testing process, as when “the person testifying is a supervisor, reviewer, or someone else with a personal, albeit limited, connection to the test at issue”).
In Williams, 132 S.Ct. at 2230, the expert witness testified that she compared a DNA profile known to be from the petitioner to a DNA profile from a swab from the victim. The scientist who generated the known profile also testified. Id. at 2229. However, the [282]*282DNA profile from the victim’s swab was produced by an outside laboratory, Cellmark, and nobody from Cellmark testified at trial. Id. The expert “trusted Cellmark to do reliable work because it was an accredited lab,” but “had not seen any of the calibrations or work that Cellmark had done.” Id. at 2230. The plurality decided that the testimony regarding the underlying report was not presented for the “truth of the matter asserted,” but rather as a basis for the expert’s opinion. Id. at 2236-37, 2240. Five Justices disagreed with this conclusion. Id. at 2258-59 (Thomas, J., concurring in the judgment); id. at 2268-69 (Kagan, J., dissenting, in which Scalia, Ginsburg, and Sotomayor, JJ., joined).
In this ease, Mr. Davenport saw all of the calibrations and work underlying the tests, and his close scrutiny and analysis of the results, comparison of the DNA profiles, and rerunning of the statistical analysis differed remarkably from the Williams expert’s bald reliance on the Cellmark report, and seems to satisfy the concerns expressed by Justices Thomas and Kagan. See id. at 2258 (Thomas, J., concurring in the judgment) (explaining that the value of the expert’s testimony “depended on the truth” of the ouCof-court statement on which the expert relied); id. at 2270 (Kagan, J., dissenting, in which Scalia, Ginsburg, and Sotomayor, JJ., joined) (“[The expert] became just like the surrogate witness in Bullcoming — a person knowing nothing about ‘the particular test and testing process,’ but vouching for them regardless.” (quoting Bullcoming, 131 S.Ct. at 2715)). Whatever the differences between Williams and this case may be, the lack of majority support in Williams “for any point but the result” means that Williams does not alter “this Court’s Confrontation Clause jurisprudence.” Tearman, 72 M.J. at 59 n. 6 (dictum).
Thus, although the Supreme Court has not provided a workable majority rule that would resolve this ease, the. Court’s precedent does not dictate the conclusion that Appellee lacked the opportunity to confront a witness against him.
ii.
In the absence of clear guidance from the Supreme Court, we are bound, within the constraints discernible from controlling precedent, to provide a clear rule for the military justice system. Fortunately, we already have a rule. This Court’s precedent makes clear that even when an expert relies in part upon “statements” by an out-of-court declar-ant, the admissibility of the expert’s opinion hinges on the degree of independent analysis the expert undertook in order to arrive at that opinion. Blazier II, 69 M.J. at 224-25.
On the one hand, experts may not “act as a conduit for repeating testimonial hearsay,” id. at 225, circumventing the Sixth Amendment by acting as a “transmitter” instead of communicating an “independent judgment.” Id. (quoting United States v. Ayala, 601 F.3d 256, 275 (4th Cir.2010)). For this reason, the testimony of an expert witness who repeated statements in inadmissible cover memoranda violated the Confrontation Clause. Id. at 226. The witness should have “proffer[ed] a proper expert opinion based on machine-generated data and calibration charts, his knowledge, education, and experience and his review of the drug testing reports alone.” Id.-, see also Sweeney, 70 M. J. at 304 (an expert’s testimony that a document “showed the presence of cocaine and codeine” was erroneously admitted). On the other hand, “[a]n expert witness need not necessarily have personally performed a forensic test in order to review and interpret the results and data of that test.” Blazier II, 69 M.J. at 224-25; cf. M.R.E. 703 (“[F]aets or data ... upon which an expert bases an opinion .... need not be admissible.”). Experts may “review and rely upon the work of others, including laboratory testing conducted by others, so long as they reach them own opinions in conformance with evidentiary rules regarding expert opinions.” Blazier II, 69 M.J. at 224. That is precisely what happened here.
This rule is not inconsistent with Williams or the precedent applied by the Williams plurality or dissenters. See supra Part III. C.i. Moreover, as Justice Kagan noted in her dissent, the “clear rule” established in [283]*283prior cases is “clear no longer,” since “[t]he five Justices who controlled] the outcome of [Williams] agree[d] on very little,” and “left significant confusion in their wake” Williams, 132 S.Ct. at 2277 (Kagan, J., dissenting, in which Scalia, Ginsburg, and Soto-mayor, JJ., joined). Nonetheless, two things are clear: first, none of the decisions of the Supreme Court have purported to jettison expert testimony in toto; and, second, neither have they suggested that each individual who touched the evidence or was involved in its analysis must testify.
And, since Williams, other courts have also focused on the extent to which an expert formed an independent opinion to determine whether the testimony was permissible in light of Crawford, as we did in Blazier II. See, e.g., United States v. Vera, 770 F.3d 1232, 1239-40 (9th Cir.2014) (an expert’s opinion regarding a gang’s control over narcotics trafficking was admissible because the combination of the expert’s knowledge, even if gleaned from testimonial statements, and his own observations turned his opinion into “an original product that could have {peen tested through cross-examination” (internal quotation marks omitted)); Leger v. State, 291 Ga. 584, 732 S.E.2d 53, 60 (2012) (a laboratory supervisor’s testimony was admissible because her involvement in the testing gave her a “significant personal connection to the test”); State v. Roach, 219 N.J. 58, 95 A.3d 683, 688, 697 (2014) (an expert was permitted to testify regarding the comparison between a DNA profile she generated and a profile generated by another analyst because she “used her scientific expertise and knowledge to independently review and analyze” the analyst’s data, and “satisfped] herself of the reliability of the results”); State v. Lopez, 45 A.3d 1, 13-14 (R.I.2012) (an expert’s opinion was admissible because he was “integrally involved” in the testing process, formulating his own conclusions rather than “act[ing] as a conduit of the opinions of, or parrot[ing] the data produced by, otherfs]”).7
iii.
Even if Mr. Davenport’s in-court statements that semen or DNA were found on the evidentiary swabs and that certain DNA samples matched each other were based in part on the Final Report, they were admissible. Mr. Davenport performed an extensive independent review of the case file, upon which the Final Report was based, during which he determined that Mr. Fisher took the prescribed quality control measures, that no accidents occurred, and that the results were logically consistent. He compared the ECD numbers on the Final Report to the numeric identifiers found elsewhere in the case file to cheek that Mr. Fisher had analyzed the correct samples. He reanalyzed the DNA profile data that Mr. Fisher generated to verify the matches that Mr. Fisher reported and recalculated the frequency statistics. This extensive review process, explored in full before the military judge during the hearing on the motion in limine, allowed Mr. Davenport to “satisfy [him]self of the reliability of the results.” See Roach, [284]*28495 A.3d at 697. In sum, Mr. Davenport presented his own expert opinion at trial, which he formed as a result of his independent review, and clearly conveyed the basis for his conclusions during the hearing on the motion in limine.
That Mr. Davenport did not himself perform aspects of the tests “goes to the weight, rather than to the admissibility” of his opinion. Blazier II, 69 M.J. at 225. And given defense counsel’s limited cross-examination of Mr. Davenport at trial, we decline to assume that they believed that there were grounds to attack the tests he did not personally perform.
D.
We conclude that this case does not' implicate the concern described in Crawford, as Appellee was not deprived of the opportunity to question and confront an opposing witness. 541 U.S. at 50-51, 124 S.Ct. 1354. Mr. Davenport’s conclusions regarding the presence of semen and identification of DNA were his own. Even if those conclusions may have derived in part from the Final Report, Mr. Davenport’s reliance on other, nontestimonial factual bases — which also served as the foundation for the Final Report — allowed him to render his own opinion. The witness against Appellee was not Mr. Fisher or the Final Report, but Mr. Davenport, who appeared in person at trial. Appellee hád the opportunity to cross-examine Mr. Davenport about his review of the case file and his expert opinion, and, generally, to “subject [the testimony] to adversarial testing.” Id. at 43, 124 S.Ct. 1354.
Having thus parsed Mi*. Davenport’s testimony, we conclude that it was admissible. Therefore, the military judge did not abuse his discretion in denying Appellee’s motion in limine, nor did Mr. Davenport’s trial testimony violate Appellee’s right to confrontation.
IV. DECISION
The decision of the United States Air Force Court of Criminal Appeals is reversed.8 The record is returned to the Judge Advocate General of the Air Force for remand to the CCA for further proceedings under Article 66, UCMJ, 10 U.S.C. § 866 (2012).