OPINION ON APPELLANT’S PETITION FOR DISCRETIONARY REVIEW
BENAVIDES, Judge.A jury convicted appellant of unlawful delivery of a controlled substance. The court found two enhancement paragraphs true and assessed punishment at confinement for thirty-five (35) years. The Court of Appeals affirmed the conviction. Vargas v. State, 781 S.W.2d 356 (Tex.App.— Houston [1st] 1989). We granted appellant’s petition for discretionary review to consider whether the State’s use of its peremptory challenges violated appellant’s right of due process under Batson v. Kentucky, 476 U.S. 79, 106 S.Ct. 1712, 90 L.Ed.2d 69 (1986).
In Batson the United States Supreme Court reaffirmed that the State’s purposeful use of peremptory challenges in a racially discriminatory manner violates the equal protection clause of the fourteenth amendment. The Court outlined a new method of showing such discrimination. Once a defendant makes a prima facie showing of the State’s discriminatory use of peremptory challenges, the burden shifts to the State to give a neutral explanation for the peremptory challenges. The Supreme Court emphasized that this explanation must be clear and reasonably specific. General assertions that a prosecutor’s reasons are not discriminatory or that the challenged venireperson would be partial to the defendant because of their shared race are not sufficient to rebut a prima facie case.
The record reflects that after the jurors were selected but before they were sworn, defense counsel objected to the State’s use of its peremptory challenges to strike five of six blacks on the venire. At the trial court’s request the prosecutor proffered his response to defense counsel’s objection setting forth the reasons for his strikes. The prosecutor did so.
The Court of Appeals characterized the prosecutor’s reasons for his strikes as “racially neutral, albeit weak.” Citing Keeton v. State, 749 S.W.2d 861 (Tex.Cr.App.1988), the court examined the evidence in the light most favorable to the trial court’s rulings and upheld the ruling because it was supported by the record. The court noted that other than his initial objection, appellant did nothing further to prove his allegations that the prosecutor had exercised his peremptory strikes in a racially discriminatory manner and did not show that the prosecutor’s neutral explanations were merely pretextual. The Court of Appeals refused to consider appellant’s argument, made for the first time on appeal, that a comparison of the white venireper-sons that the State did not strike with the black venirepersons that the State struck showed some white venirepersons with characteristics similar to the black venire-persons.
Appellant contends that the Court of Appeals erred in holding that footnote 6A in Tompkins v. State, 774 S.W.2d 195, 202 (Tex.Cr.App.1987), precluded a comparison analysis between white venirepersons who were not struck and black venirepersons who were struck. Appellant also contends that even if the comparison analysis is not utilized, the prosecutor’s stated reasons are racially discriminatory under so-called objective standards such as those discussed in Keeton. It is to this latter contention we turn first.
As the Court of Appeals stated, in Kee-ton a majority of this Court held that an appellate court should review a Batson issue by viewing the evidence in the light most favorable to the trial judge’s rulings and determining if those rulings are supported by the record. A plurality of this Court sought to further define the applicable appellate standard of review in Batson cases. In Whitsey v. State, 796 S.W.2d 707 (Tex.Cr.App.1990) (opinion on rehearing), the plurality adopted the “clearly erroneous standard” as an extension of the “supported by the record” standard. Id. at *554712.1 The plurality also stated that the “supported by the record” standard “is actually an analytical tool used in determining whether a trial judge’s findings of fact are clearly erroneous or should be accorded great deference.” Id. at 740. The analysis is essentially the same under any of these standards.
In Hernandez v. New York, — U.S. -, 111 S.Ct. 1859, 114 L.Ed.2d 395 (delivered May 28, 1991), a majority of the United States Supreme Court held that Batson claims, whether from federal or state courts, are to be analyzed under a “clear error standard of review.” Id. at -, 111 S.Ct. at p. 1871. The Court emphasized the factual nature of Batson inquiries stating:
Batson’s treatment of intent to discriminate as a pure issue of fact, subject to review under a deferential standard, accords with our treatment of that issue in other equal protection cases, [citations omitted]
[[Image here]]
Deference to trial court findings on the issue of discriminatory intent makes particular sense in this context because, as we noted in Batson, the finding will ‘largely turn on evaluation of credibility.’ 476 U.S., at 98, n. 21 [106 S.Ct. at 1724, n. 21]. In the typical peremptory challenge inquiry, the decisive question will be whether counsel’s race-neutral explanation for a peremptory challenge should be believed. There will seldom be much evidence bearing on that issue, and the best evidence often will be the demeanor of the attorney who exercises the challenge. As with the state of mind of a juror, evaluation of the prosecutor’s state of mind based on demeanor and credibility lies ‘peculiarly within a trial judge’s province.’ [citation omitted]
Id. at -, 111 S.Ct. at p. 1869. The Court discounted a claim that caselaw indicated an appellate court should make an independent review of the facts involved in a Bat-son contention. The Court noted that proper deference was shown to state court factual determinations when a reviewing court was not left with a “ ‘definite and firm conviction that a mistake has been committed.’ United States v. United States Gypsum Co., supra, [333 U.S. 364] at 395 [68 S.Ct. 525, at 542, 92 L.Ed. 746 (1948)].” Id. at p. -, 111 S.Ct. at p. 1871.
We apply this “clear error standard of review” as explained in Hernandez. We apply this standard by reviewing the record, including the voir dire and the racial makeup of the venire, the prosecutor’s neutral explanations, and appellant’s rebuttal and impeaching evidence.
Appellant’s contention that the objective factors discussed in Keeton control the analysis is not correct. These factors certainly may be considered in evaluating the trial judge’s overruling of a Batson claim, but they are not determinative. The overriding standard is still whether the trial judge’s decision was supported by the record so that it is not clearly erroneous. Appellant relies upon three of these objective factors to contend that the prosecutor’s reasons for his strikes were not shown to be racially neutral. Appellant argues that the prosecutor’s strikes of Clark, Porter and Martin demonstrate an assertion of a group bias without showing that the group trait applies to the challenged juror; that there was no examination or perfunctory examination of the challenged venirepersons; and that the given explanation had no rational relationship to the case on trial. Appellant did not cross-examine or impeach the prosecutor to show that the prosecutor’s “assumptions” that a venireperson who was a legal assistant, a venireperson with a previous arrest, and one who had a relative convicted of public intoxication, would not favor the State in any given case.
In Whitsey this Court discussed with approval several federal cases involving prosecutors’ explanations based upon age, eye contact, body language, and that the veni-repersons were “ ‘young, single, and without children or a ‘substantial stake in the community’ and that each appeared inat*555tentive....”’ Whitsey, supra, at 725 (quoting United States v. Lance, 853 F.2d 1177 (5th Cir.1988). This Court noted that the courts of appeals in those cases deferred to each trial judge’s findings because the prosecutor partially based his neutral explanations on nonquantifiable characteristics and the trial judge personally observed the proceedings. Likewise, in the instant case the trial judge observed the proceedings, including that of the six blacks on the venire, one served on the jury and the prosecutor struck the other five for the following reasons:
Veniremember Gordon — No. 4
No. 4, Your Honor, she stated that in response to my question she needed more than one eyewitness. She was brought up to the bench and upon further questioning she told you that she could be a fair and impartial witness; she thinks she could follow the law.
I’m not convinced of that fact and therefore, I struck No. 4.
Veniremember Clark — No. 16
No. 16 is Ms. Lennie Clark. She’s a legal assistant, Your Honor. And I’m not making a general habit to strike people in the legal community. It’s my experience that they have been my toughest. So therefore, I struck; and also, she sat somewhat unattentively.
Veniremember Guitry — No. 22
No. 22 told me during my voir dire examination he had a nephew killed by police officers and he had a real bad experience. He didn’t think he would block that from his mind. At the bench he told this Court that he didn’t have faith in the D.A.’s Office and therefore, Your Honor, I struck him.2
Veniremember Porter — No. 28
No. 28 has a family member that was convicted of a PI. I struck her for those reasons. I struck 26 which is a white male Caucasian because he had a cousin that was in drugs. I struck No. 7 which is a white male Caucasian because he was arrested for PI ...
Veniremember Martin — No. 36
... [A]nd then on No. 36, Your Honor, he was arrested for DWI and we didn’t even reach him to start off with.
While not all of these reasons are nonquan-tifiable, the prosecutor’s assumption that these venirepersons would not be preferred State’s jurors is not facially unreasonable. Gordon’s initial lack of belief in a case with one eyewitness is a plausible reason to strike her as not so favorable to the State, even though she later said she could follow the law. The State might reasonably not want her on the jury in a case involving two eyewitnesses who were both assigned to the much-maligned Southeast Tactical Response Unit of the Houston Police Department. If the credibility of one was suspect, so too might be the other.
In reference to his strike of Clark, the prosecutor stated as one reason that in his experience people from the legal community were “tough” on the State. Past experience is somewhat nonquantifiable.
In justifying his strike of Porter, who had a family member that was convicted of public intoxication, the prosecutor stated that he had also struck two white venire-persons for what he considered to be similar reasons. One of the white venireper-sons had a cousin who had been arrested for an offense involving drugs and the other venireperson had been arrested for public intoxication. The prosecutor’s limited questioning of these venirepersons was the same as that of Porter.
The prosecutor also struck Martin because he had been arrested for driving while intoxicated.3 This action is consistent with and similar to his strikes of Porter and the mentioned white venirepersons, who had themselves been involved with the criminal justice system or had relatives who had been so involved.
*556While these reasons would seem stronger had the prosecutor individually questioned all of the stricken venirepersons, the reasons are not rendered racially impermissible simply because he did not do so. Defense counsel did not cross-examine the prosecutor or impeach him on any of his reasons, even when the prosecutor also mentioned white venirepersons whom he claimed he struck for similar reasons as he did venireperson Porter. Defense counsel did not present the trial judge with any comparison, general or detailed, of unchallenged white venirepersons and the black venirepersons who were struck. There is nothing to show that the prosecutor’s stated reasons were pretextual. These reasons are racially neutral on their face. As the plurality noted in Whitsey:
“If the district court's account of the evidence is plausible in light of the record viewed in its entirety, the court of appeals may not reverse it even though convinced that had it been sitting as the trier of fact, it would have weighed the evidence differently. Where there are two permissible views of the evidence, the fact-finder’s choice between them cannot be clearly erroneous.”
Whitsey, supra at 722 (citations omitted), quoting Anderson v. Bessemer City, 470 U.S. 564, 573-574, 105 S.Ct. 1504, 1511, 84 L.Ed.2d 518, (1985). See also Hernandez, supra. The trial court implicitly found the prosecutor’s explanations to be “race-neutral,” and there is nothing before us to suggest otherwise. We overrule appellant’s ground for review.
This brings us to appellant’s second contention, that a comparison analysis of unchallenged white venirepersons to the black venirepersons that were struck demonstrates that the prosecutor’s reasons were impermissibly based upon race. The Court of Appeals relied upon footnote 6A in Tompkins to hold that this rebuttal evidence presented by appellant for the first time on appeal would not be considered. Appellant attempts to distinguish Tompkins by arguing that it was a capital murder case involving lengthy individual examinations of venirepersons. Nothing in Tompkins indicates that such distinction makes any difference. Appellant argues that a court must analyze, not merely accept, the State’s justification; a comparison analysis is critical to a fair determination.
In Young v. State, 826 S.W.2d 141 (Tex.Cr.App.1991) we held that a defendant may, on appeal, rely upon a “comparative analysis” of venirepersons showing disparate treatment between black and white venirepersons to support his argument that the prosecutor’s reasons for using peremptory strikes were pretexts. To have such evidence considered, a defendant is not required to request that the trial judge make his finding upon a comparison analysis nor is he obliged to offer in evidence testimony from jury selection supporting such an analysis. We stated that this type of analysis is significant in order to give an appellate court a more accurate picture of the voir dire than a cold record.4 However, we also specifically stated that a defendant or the State may argue “what is in evidence from the voir dire and the Batson hearing....” Id. 826 S.W.2d at 146. We noted that the trial court as supervisor of the voir dire is in a position to readily perceive discrepancies during the jury selection process. Young, supra, 145.
The comparison “evidence” upon which appellant relies is at least partly based upon the juror information cards, which were not mentioned or offered into evidence by either side. It is not proper for an appellate court reviewing a trial court’s decision on a matter to rely upon information that was not admitted as evidence at the Batson hearing in the trial court. Second, to allow an appellate court to rely upon such information undermines the standard of review for a Batson issue which gives great deference to the trial court. See Hernandez, supra; Keeton, supra; and Whitsey, supra. It would be incongruous to give great deference to a trial court’s ruling if it is “supported by the record,” and yet reverse that ruling on the *557basis of comparison information that was not in evidence nor presented to the trial court at the time of the hearing and was not part of the record upon which the trial court based its ruling. Moreover, while Young can be said to stand for the proposition that discrepancies between stated “race neutral” explanations and jury answers during voir dire may be readily determined by the trial judge supervising the voir dire, discrepancies between the “race neutral” explanations and the answers in jury information cards generally furnished to and completed by prospective jurors pri- or to the voir dire are not readily determinable.
A defendant must present impeaching evidence to the trial court if he intends to rely on it later. An appellate court may not reverse a trial court’s finding based upon information that was not introduced into evidence or elicited before the trial judge during the voir dire.
We hold that where, as in this case, the basis of the comparison evidence is not in evidence presented to the trial court, either at the voir dire or at the subsequent Bat-son hearing, an appellate court may not consider it in evaluating a Batson claim. It is apparent that the juror information cards in this case fall into that category.
However, our examination of the appellate record also reveals that some evidence of the kind relied upon by Appellant for comparison with the racially neutral explanations given by the prosecutor does appear in the record of jury selection. For example, it is clear from the voir dire examination that veniremember Bue, like venire-member Clark, was a paralegal by profession. Unlike Clark, however, Bue was not struck from jury service by the prosecuting attorney, even though the prosecutor assigned employment in the legal profession as the basis for using a peremptory challenge against Clark. Under our recent holding in Young Appellant is entitled to rely upon this evidence, and any other of similar tenor or effect from voir dire, when arguing before the appellate court that certain veniremembers were in fact excluded by the prosecuting attorney on the basis of their race. Because it appears that the lower court also refused, on authority of footnote 6A in Tompkins, to consider this evidence, it is our opinion that its decision was reached in a manner contrary to law.
Accordingly, the judgment of the Court of Appeals is reversed and this cause is remanded there for reconsideration of Appellant’s Batson complaint in a manner not inconsistent with this opinion.
. A plurality opinion is of limited precedential value. See Farris v. State, 819 S.W.2d 490, 501, n. 3 (Tex.Cr.App.1990).
. On appeal appellant states that he does not challenge the State’s strike of Veniremember Guitry.
. Given our disposition of this reason for striking Martin, we do not reach the other justification given by the prosecutor that Martin was not reached on the jury list.
. We caution that such review does not include an independent appellate review of fact findings, which in the Batson context is best left to the trial court because such issues largely turn on evaluation of credibility. Hernandez, supra.