The opinion of the court was delivered by
Six, J.:This is a peremptory challenge jury selection case, alleging the State’s strikes were not “race neutral.” Arthur Walston was convicted of involuntary manslaughter in the shooting death of a 12-year-old girl. He alleges that the State’s use of peremptoiy challenges to strike African-American panelists from the venire was motivated by race, violating the 14th Amendment Equal Protection Clause and Batson v. Kentucky, 476 U.S. 79, 90 L. Ed. 2d 69, 106 S. Ct. 1712 (1986).
The Court of Appeals, in an unpublished opinion filed May 6, 1994, affirmed the district court’s finding that the State’s race-neutral justifications for the strikes were satisfactory. We granted Walston’s petition for review. See Rule 8.03 (1993 Kan. Ct. R. Annot. 44.)
The issue is whether the district court abused its discretion in finding the State’s explanations for each of its four strikes to be race neutral and therefore permissible. The third and final step of the Batson analysis is the only step at issue in the instant appeal. We affirm the Court of Appeals and the district court.
The Standard of Review
The appellate standard of review applicable to a trial court’s ruling that the State did or did not act with discriminatory purpose in exercising a peremptory challenge under Batson is deferential to the trial court, regardless of whether the standard is phrased as “abuse of discretion” or “clearly erroneous.” See, e.g., State v. *374Kingsley, 252 Kan. 761, 772, 851 P.2d 370 (1993) (appellate review of a trial court’s acceptance of the State’s reasons as racially neutral for removal of juror “is on the basis of abuse of discretion’’). We endorse abuse of discretion as the phrase to describe the deferential standard to be applied. Judicial discretion is abused only when exercised in an arbitrary, fanciful, or unreasonable manner, or in other words, when no reasonable person would take the view adopted by the trial court. State v. Wagner, 248 Kan. 240, 242, 807 P.2d 139 (1991).
Facts
Because jury selection is the only issue on review, the facts underlying Walston’s conviction are of minor significance. However, the facts provide background for the stated justifications for striking certain African-American panelists from the venire. The Court of Appeals summarized:
“On the morning of March 5, 1992, defendant reported to the Kansas City, Kansas police that his car had been stolen from his mother’s residence. After reporting the theft to the police, defendant began looking for his car. He contacted several friends and subsequently located his car parked in front of a house. He returned to his mother’s house and notified the police that he had located his car.
“After defendant returned to where he had last seen his car, he discovered it was gone. He, however, later spotted his car going north on 34th Street in Kansas City, Kansas. He gave chase, attempting to stop the driver by flagging the car down; however, the driver failed to stop and appeared to speed up. Defendant then stopped his car, took out a .38 caliber handgun, and fired a shot at the car. The bullet shattered the rear window, and struck and killed a 12-year-old passenger.
“In a panic, defendant returned to his mother’s house, concealed the weapon, and returned to the scene of the shooting. Upon his return, he was arrested by the police. Defendant was charged, tried before a jury, and found guilty of involuntary manslaughter.”
Walston is African-American. At the conclusion of voir dire, the State used four of its six peremptory challenges to strike African-Americans from the venire. Walston’s attorney objected on the basis the challenges were racially motivated.
Walston’s objection was discussed in chambers. The trial judge considered whether Walston had satisfied the first part of the *375Batson test, i.e., whether he had established a prima facie case of discriminatory intent. The State argued that diere was no evidence of systematic discrimination. The State suggested that it had no motive to exclude African-Americans because the victim and “half” of the State’s witnesses were of that race. The trial judge found an inference of discriminatory intent. The judge noted that there were, in his estimation, 7 African-Americans out of 24 venirepersons before the State’s peremptory strikes. Consequently, the trial judge requested that the State provide race-neutral reasons for its strikes.
Explaining its peremptory challenges of the four African-American venirepersons, the State said it struck:
V.T. “because she said [in voir dire] that she had an adult son who had been convicted of drug offenses and she thought he had been treated unfairly”;
T-T. because he and his older brother knew “one of the Reverends who apparently the defense intends to call as a character witness”;
S.E. because (1) she was married and had no children, (2) her body language suggested a sympathy for the defense (according to the prosecutor, S.E. “woke up when [defense counsel] was talking,” was “nodding her head continually through [defense counsel’s] voir dire, leaning forward,” and “seemed to pay careful attention to what [defense counsel] was saying”), and (3) she may have lacked experience driving a vehicle; and
C.R. because (1) her body language, like S.E.’s, suggested a sympathy for the defense (the prosecutor described C.R.’s and S.E.’s body language concurrently in saying that they “woke up when [defense counsel] was talking” and “seemed to pay careful attention to what [defense counsel] was saying,”) and (2) she had no children.
The record of the voir dire proceedings supports each of the State’s articulated reasons for striking the four jurors in question, with two exceptions. First, the State’s body-language justifications for striking S.E. and C.R. are not capable of being reviewed. Prospective jurors’ nods, leans, winks, smiles, or scowls do not show up on the record, unless counsel expressly makes a note of them. *376For this reason, we have held that while body language may be a “valid and neutral reason” for striking a juror, a trial judge “must be particularly sensitive when body language, alone, is advanced as a reason for striking a juror.” State v. Hood, 245 Kan. 367, 374, 780 P.2d 160 (1989). The State, in the case at bar, did not rely exclusively on body-language justifications for any challenged juror, nor did it assert body-language justifications for every stricken African-American. Either approach by the State might cause a trial court to view body-language justifications with heightened suspicion.
Second, the State’s driving-related justification for striking S.E. is questionable based on the transcript of the voir dire proceedings. The State based its uncertainty about S.E.’s driving experience on the following exchange:
“[Prosecutor]: How many people drive on 1-635? (Several responses)
Maybe I should say how many people have never driven on 1-635? Okay, only one person. Is it Ms. — “[Prosecutor]:
[S.E.] “[S.E.]:
—[S.E.] You’ve never been on 1-635? “[Prosecutor]:
I don’t know. “[S.E.]:
Have your ever — have you been in that area ever? “[Prosecutor]:
No. “[S.E.]:
So you’ve been in a car. Are you familiar with 1-635, then? “[Prosecutor]:
“[S.E.]: Uh-huh.”
This exchange does not establish that S.E. lacks driving experience; it only proves she was uncertain or confused about the whereabouts of 1-635 and whether she had ever been on it. The State on appeal says that the prosecutor’s concern about S.E.’s driving experience was tied to the fact that a car was involved in the facts of the crime. The prosecutor did not want a juror who was preoccupied with facts related to the car, “such as whether the driver could have avoided the shooting.” The prosecutor’s appellate rationale was not shared with the trial judge; consequently, we are reluctant to consider it in reviewing whether the trial court abused its discretion.
The State’s race-neutral reasons for striking the four African-Americans were not discussed individually. The trial judge gave *377a blanket approval, stating: “[H]aving made the inquiry as to the motives, I find that these are race-neutral reasons and, therefore, I’m going to impanel the jury as selected by the — both of you.”
Walston’s Contentions
Walston contends that the trial court erred because it failed to properly compare the African-American panelists who were struck with the whites who were not struck. He complains that the trial judge made no determination of whether the characteristics the State identified in explaining its strikes of the African-Americans also existed in the unchallenged whites. Such a comparison is required, Walston contends, under State v. Belnavis, 246 Kan. 309, 312, 787 P.2d 1172 (1990).
Syllabus ¶ 2 of Belnavis states: ‘Where a prosecutor’s explanation for peremptory challenges relies on characteristics of persons struck and those same characteristics are present in individuals not challenged, the explanation is not race neutral.” See United States v. Wilson, 853 F.2d 606, 610 (8th Cir. 1988).
In his petition for review, Walston concedes that the Court of Appeals engaged in a Belnavis-type comparative analysis and upheld the peremptory challenges. He argues, however, that the trial court’s failure to compare challenged and unchallenged panelists was incurable constitutional error. Walston asserts that the trial court disregarded step three of Belnavis. His arguments, however, are not supported by Batson. Moreover, his characterization of “step three of Belnavis” is inconsistent with the third step in the Batson analysis.
The Batson Analysis
Batson held that “the Equal Protection Clause forbids the prosecutor to challenge potential jurors solely on account of their race or on the assumption that black jurors as a group will be unable impartially to consider the State’s case against a black defendant.” 476 U.S. at 89. Batson outlined a three-step process for evaluating claims of discrimination in jury selection that “permits prompt rulings on objections to peremptory challenges without substantial disruption of the jury selection process.” Hernandez v. New York, 500 U.S. 352, 358, 114 L. Ed. 2d 395, 111 S. Ct. 1859 (1991). *378The three-step analysis outlined in Batson and Hernandez was set forth in State v. Poole, 252 Kan. 108, 843 P.2d 689 (1992):
“ ‘First, the defendant must make a prima facie showing that the prosecutor has exercised peremptory challenges on the basis of race. [Batson, 476 U.S. at 96-97.] Second, if die requisite showing has been made, the burden shifts to the prosecutor to articulate a race-neutral explanation for striking die jurors in question. [476 U.S. at 97-98.] Finally, the trial court must determine whether the defendant has carried his burden of proving purposeful discrimination. [476 U.S. at 98.]’ ” 252 Kan. at 110 (quoting Hernandez, 500 U.S. at 358-59). (Emphasis added.)
Discussion
The trial court in the case at bar conducted all three steps of the Batson analysis. First, upon Walston’s objection, the judge stated to the prosecution: “You used four of your strikes on African-Americans. Mr. Walston is an African-American. I think he’s satisfied the first part of the test and I’d like to see if you have a race neutral reason for your strikes.” Second, the State responded by explaining its strikes. Third, after hearing the State’s explanations, the trial court made the race-neutral finding. Walston had not “carried his burden of proving purposeful discrimination.” Poole, 252 Kan. at 110 (quoting Hernandez, 500 U.S. at 359.)
After hearing the State’s explanations, and having observed the entire voir dire process, the trial court, under Batson, is to make a purely factual determination: Has the prosecution purposefully discriminated by exercising peremptory challenges against persons solely on account of their race? Batson and Hernandez make this point clear:
“As in any equal protection case, die ‘burden is, of course,’ on the defendant who alleges discriminatory selection ... ‘to prove die existence of purposeful discrimination.’ [Citations omitted.] In deciding if the defendant has carried his burden of persuasion, a court must undertake a ‘sensitive inquiry into such circumstantial and direct evidence of intent as may be available.’ [Citation omitted.]” Batson, 476 U.S. at 93 (Emphasis added.)
The Supreme Court noted later in Batson, “Since the trial judge’s findings in the context under consideration here will turn on evaluation of credibility, a reviewing court ordinarily should give those findings great deference.” 476 U.S. at 98 n.21.
*379The Supreme Court in Hernandez reaffirmed the factual nature of the trial court’s decision in the third step of the Batson analysis:
“In Batson, we explained that the trial court’s decision on the ultimate question of discriminatory intent represents a finding of fact of the sort accorded great deference on appeal:
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] . . .
“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.’ [Citation omitted.] 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. . . . [the evaluation of which] lies ‘peculiarly within a trial judge’s province.’ [Citations omitted.]” (Emphasis added.) Hernandez, 500 U.S. at 364-65.
In addition to the credibility of the prosecutor, the trial court may consider “such circumstantial and direct evidence of intent as may be available.” Batson, 476 U.S. at 93. We have recognized that one “relevant factor” a trial court may consider in deciding whether the State acted with a discriminatory purpose is “the percentage of blacks on a venire panel and the percentage that ends up on the jury panel.” State v. Poole, 252 Kan. at 114. The percentage of African-Americans on the venire panel and eventual jury panel (including alternates) in the instant case was approximately the same, 29 percent.
The Batson analysis avoids placing a determinative emphasis on any one factor. The primary decision on the question whether the State has acted with an unlawful purpose is placed by Batson in the hands of the trial judge. The trial judge can objectively compare numbers or other facts and subjectively evaluate the credibility of the State’s counsel in explaining the reasons for each challenged strike.
In Belnavis, we reversed the conviction and remanded the case for a new trial based on a perceived Batson violation. 246 Kan. at 314. Belnavis concluded that the same factors the State offered in explaining why it struck two African-American panelists (one, *380because she might become too preoccupied with details in the case given the detail-intensive nature of her job, and the other, because she had a seven-month-old baby and might therefore be distracted) also existed to a sufficiently similar degree in several white panelists who were not struck. 246 Kan. at 312-14. We concluded that the State’s reasons were not credible, i.e., that its race-neutral explanations were merely pretextual and that the real basis for the strikes was race. In phrasing our holding, we said that the “State failed to present a racially neutral reason” for challenging the African-American panelists in question. 246 Kan. at 314.
Belnavis, in purporting to apply the Batson test, did not state the third and final part of the test: The defendant must carry the burden of proving purposeful discrimination. See 246 Kan. at 311. Belnavis did not characterize the third step as a factual finding, to be made by the trial court and to be afforded “great deference” on appeal. Belnavis came before Hernandez, which further clarified die Batson three-step analysis.
Walston relies upon the following statements in Belnavis:
“In order to determine whether the State has successfully provided a race-neutral explanation for the peremptory challenges, we must compare the characteristics of the individuals stricken with those not stricken. A prosecutor’s explanation fails to be racially neutral if characteristics of a person struck are present in white panel members not challenged by die State. U.S. v. Wilson, 853 F.2d 606, 610 (8th Cir. 1988).
“[T]he characteristics articulated to explain a neutral reason for challenging the black venireman were present in white jurors who went unchallenged. This is evidence of impermissible purposeful discrimination by the State. U. S. v. Wilson, 853 F.2d at 610.” (Emphasis added.) 246 Kan. at 312-13.
These Belnavis statements are too broad. There is no indication in Belnavis whether defense counsel informed the trial judge at the time of making the objection that some white jurors also did detailed work or had young children, and therefore the State’s reasons were not credible. Likewise, in the instant case, Walston’s counsel just made a general Batson objection. The State offered reasons, and the trial court ruled the reasons sufficient. Walston did not attack the State’s reasons by saying that certain white jurors had the same characteristics.
*381The concept behind the Belnavis statements quoted above holds a proper place within the Batson analysis. The ultimate decision is on whether the State has purposefully discriminated. The trial court can consider any “circumstantial and direct evidence of intent as may be available.” Batson, 476 U.S. at 93. The fact that a prosecutor strikes an African-American juror, supposedly for Reason A, but does not strike a white juror, who exhibited the same Reason A, is certainly circumstantial evidence that the State is purposefully discriminating against African-Americans. This kind of circumstantial evidence may be sufficiently compelling in some cases to convince a trial court that the State is purposefully discriminating and that its race-neutral reasons are pretextual. However, comparison-based circumstantial evidence should not be considered conclusive in every case, as a matter of law.
Whether the State’s reason is legitimate, or whether it is merely a pretext for a true discriminatory motive of seeking an all-white jury, should be evaluated based on all of the available direct and circumstantial evidence of intent. The similarity of white jurors who were not challenged should be included in the evaluation. Under the Batson framework, the trial court is given primary responsibility for making that evaluation. Appellate courts are to review the trial court’s decision with deference. We disapprove of any language in Belnavis that suggests the comparability of challenged and unchallenged jurors is evidence entitled to conclusive weight in determining a Batson violation.
Walston contends that regardless of whether his arguments comparing challenged African-American jurors to unchallenged white jurors have any merit, the trial court’s failure “to even mention the third step imposed by Belnavis must be viewed as discretionary abuse.” In other words, Walston claims that the trial court had an affirmative duty, upon finding a prima facie case in the first step under Batson, to ask the State for the reasons for its strikes against African-Americans and to compare those reasons with the information elicited from white panelists who were not challenged. Since Walston’s counsel did not come forward with any information about the unchallenged white jurors at the time *382the trial judge was considering the objection, Walston must be arguing that the trial court had a duty to make this inquiry sua sponte. He cites State v. Tittes, 245 Kan. 708, 715, 784 P.2d 359 (1989); State v. Huskey, 17 Kan. App. 2d 237, 240, 834 P.2d 1371 (1992); and State v. Ribadeneira, 15 Kan. App. 2d 734, 748, 817 P.2d 1105, rev. denied 249 Kan. 778 (1991), in support of his claim. Tittes and Huskey were cases in which the sentencing judges failed to comply with statutory requirements that certain findings be made at sentencing. Ribadeneira merely cites Tittes. We find no support in the cases Walston cites, or in Hernandez, Batson, or any other case, for imposing such a burdensome duty on the trial court when considering a Batson objection. To the contrary, the Tenth Circuit has stated in the Batson context that “[t]he burden of creating a record of relevant facts belongs to the defendants.” United States v. Esparsen, 930 F.2d 1461, 1466 (10th Cir. 1991).
Applying the Law to the Facts of This Case
In the case at bar, the trial court made a finding of fact that the State’s reasons for using four of its six peremptory strikes against African-Americans were racially neutral. We have repeatedly said that we review such findings with deference, reversing only if we find abuse of discretion or clear error, which are similar standards. Poole, 252 Kan. at 111, 114 (upholding trial court’s finding of no discriminatory purpose because it was not “clearly erroneous”); State v. Walker, 252 Kan. 117, 123, 843 P.2d 203 (1992) (expressly choosing to apply “abuse of discretion” standard over “clearly erroneous” standard in this context but noting “our conclusion would be the same under the clear error standard”). See Sward, Appellate Review of Judicial Fact-Finding, 40 Kan. L. Rev. 1, 7 and n.18 (1991) (suggesting that “the abuse of discretion standard is probably more deferential than the clearly erroneous standard” but noting that the “distinction between the two is difficult to draw” and any difference “may be more intuitive than scientific”).
Walston argues that an abuse of discretion is proven by the fact that the State did not strike several white jurors who had *383similar characteristics to those enunciated by the State as race-neutral reasons for striking African-Americans. In Walston s case, the comparative analysis advanced on appeal was not presented to the trial court at the time the objection was made. Walston carries the ultimate burden of persuasion on whether the State engaged in intentional discrimination. Batson, 476 U.S. at 94 n. 18.
Walston’s comparative analysis is not persuasive. No other panelist was comparable to V.T., who said the judicial system treated her son as a defendant at least partially “unfairly.” Walston identifies no other panelist who, like J.T., knew a potential character witness for the defense. While other unchallenged jurors had no children, like S.E., Walston fails to argue that those other jurors also exhibited hostile body language towards the prosecution or that S.E. did not. He also fails to challenge the State’s driving-experience justification for striking S.E., which, even if of questionable merit, still distinguishes S.E. from other unchallenged jurors with no children. Similarly, C.R., who was struck because she had no children and because of her body language, is distinguishable from the two white jurors who had no children.
A review of the record does not convey a sense that the State was purposefully trying to strike African-Americans. The State pointed out to the trial court that the victim was African-American, as were many witnesses, and therefore it had good reasons not to discriminate. Thus, regardless of whether some of the State’s reasons for striking certain jurors were slightly subjective, we cannot say that the trial court abused its discretion in finding such reasons racially neutral.
Affirmed.