DISTRICT COURT OF APPEAL OF THE STATE OF FLORIDA FOURTH DISTRICT
RANDOLPH THOMAS, Appellant,
v.
STATE OF FLORIDA, Appellee.
No. 4D2024-0718
[February 26, 2025]
Appeal from the Circuit Court for the Seventeenth Judicial Circuit, Broward County; Michael Lynch, V, Judge; L.T. Case No. 16- 008958CF10A.
Antony P. Ryan, Regional Counsel, Office of Criminal Conflict and Civil Regional Counsel, Fourth District, and Richard G. Bartmon, Assistant Regional Counsel, West Palm Beach, for appellant.
James Uthmeier, Attorney General, Tallahassee, and Heidi L. Bettendorf, Senior Assistant Attorney General, West Palm Beach, for appellee.
DAMOORGIAN, J.
Randolph Thomas (“Defendant”) appeals his conviction and sentence for two counts of sexual battery while standing in a position of familial or custodial authority of a person 12 years or older, but under 18 years old, of his girlfriend’s daughter (“Victim”). Defendant appeals on several grounds, including the trial court’s admission of DNA expert testimony using Bayes’ Theorem statistical analysis at trial regarding Defendant’s paternity of Victim’s child. We affirm the conviction in all respects and write only to address why we reject Defendant’s challenge of the use of Bayes’ Theorem.
By way of background, in 2015, Victim reported that Defendant, who was Victim’s mother’s live-in boyfriend, had been sexually assaulting her since she was a minor. Victim also alleged that Defendant was the biological father of Victim’s then seven-year-old son (“Child”). Police took DNA samples from Victim, Child, and Defendant, and DNA tests concluded that Defendant and Child’s DNA matched at all fifteen (15) genetic markers available to the lab at the time of testing. 1 Defendant was arrested and charged with two counts of sexual battery while standing in a position of familial or custodial authority of a person 12 years or older, but under 18 years old, Count 1 – penile penetration, and Count 2 – oral penetration.
The case proceeded to a jury trial. In relevant portion, Defendant sought to exclude at trial the testimony of the State’s DNA expert witness based upon his reliance on Bayes’ Theorem, a mathematical formula commonly used in paternity testing. The trial court denied Defendant’s motion in limine and allowed the State’s expert testimony. The State’s expert explained on cross-examination, in relevant portion, the use of prior probability under Bayes’ Theorem as follows:
[P]rior probability is a ratio of two things. One. We know that every child has to have a male and a female parent. So 50 percent of the DNA came from a man.
The other probability is, what’s the probability that any man could be the father? And we assign that to be 50 percent. So it’s 50 percent that any man could be the father or the fact that 50 percent of his DNA came from the father.
What that does is .5 over .5, is one. And it means therefore, the prior probability has no effect whatsoever on the posterior probability. It renders it effectively zero, and allows us then to just consider the genetic evidence.
The State’s expert explained at the conclusion of his analysis that “[t]he results were that it is 287 million times more likely that [Defendant] is the biological father of [Child] rather than a random untested man. That translates to a probability of paternity of greater than 99.9999996 percent.” At the end of the trial, the jury ultimately found Defendant guilty as charged on both counts.
On appeal, Defendant argues that the trial court abused its discretion by admitting the State’s expert testimony because it “assumed a 50% probability that [Defendant] had intercourse with Victim and was [the] father of her child” which “eviscerated [the] presumption of innocence” and abrogated Defendant’s rights, relying on State v. Skipper, 637 A.2d 1101 (Conn. 1994). The State responds that modern courts have rejected the premise that the use of Bayes’ Theorem violates the presumption of
1 The 16th marker was gender, which is not used in paternity calculations.
2 innocence, and that the use of “prior probability” under Bayes’ Theorem provides a neutral assessment of paternity in criminal cases. For the reasons outlined below, we decline to accept Defendant’s argument, and instead adopt the modern approach to Bayes’ Theorem outlined in Griffith v. State, 976 S.W.2d 241, 242 (Tex. App. Amarillo 1998), and its progeny.
Both sides, as well as the trial court, acknowledge that the only Florida case found through research that references Bayes’ Theorem is Brim v. State, 779 So. 2d 427 (Fla. 2d DCA 2000), but Brim did not directly address the issue at hand. Because this is a case of first impression in Florida, we look to other jurisdictions for guidance on the issue of prior probability using Bayes’ Theorem.
Defendant relies upon cases from the 1980s and 1990s in Connecticut and Wisconsin to support his position. In relevant portion, the Connecticut Supreme Court held, in Skipper, that the trial court reversibly erred by admitting paternity evidence that had utilized Bayes’ Theorem because it “permitted the introduction of evidence predicated on an assumption that there was a fifty-fifty chance that sexual intercourse had occurred in order to prove that sexual intercourse had in fact occurred. . . The fifty-fifty assumption that sexual intercourse had occurred was not predicated on the evidence in the case but was simply an assumption made by the expert.” Skipper, 637 A.2d at 1106.
Defendant also relies upon State v. Hartman, 426 N.W. 2d 320, 326 (Wisc. 1988). In Hartman, the Wisconsin Supreme Court majority held, in relevant portion:
[T]he probability of paternity is calculated based upon the assumption that the mother and putative father have engaged in sexual intercourse at least once during the period of possible conception. Because the probability of paternity assumes that sexual intercourse has occurred, it is improper to use this statistic to prove that sexual intercourse has occurred.
Id. (internal quotation and citation omitted).
However, Justice Steinmetz’s dissent in Hartman, in relevant portion, notes that any potential issues with jury confusion regarding prior probability with Bayes’ Theorem could be resolved by its admission through an expert witness:
3 The requirement of an expert to introduce the test results, coupled with the opportunity by defense counsel to examine the expert and thus explain the method of calculating paternity, sufficiently compensates for any alleged problems inherent in the Bayesian formula.
The 50 percent prior chance assumption does not require shifting the burden of proof to the defendant and is not an impermissible assumption; rather, it is part of a scientific theorem and the jury should be so told. Contrary to the majority’s assertion, the assumption is valid because the defendant has been named as the male having had sexual intercourse with the mother. In this case, the mother was the victim of a sexual assault and accused the defendant of being the father of her child as a result of his act. Therefore, the defendant was not randomly selected, but rather, had been named in a sworn complaint or information. That is, the “assumption” that is used in the probability of paternity computation was not presented into evidence in a vacuum, but was instead admitted only after a factual basis for the statistic was offered into evidence.
Id. at 327.
Free access — add to your briefcase to read the full text and ask questions with AI
DISTRICT COURT OF APPEAL OF THE STATE OF FLORIDA FOURTH DISTRICT
RANDOLPH THOMAS, Appellant,
v.
STATE OF FLORIDA, Appellee.
No. 4D2024-0718
[February 26, 2025]
Appeal from the Circuit Court for the Seventeenth Judicial Circuit, Broward County; Michael Lynch, V, Judge; L.T. Case No. 16- 008958CF10A.
Antony P. Ryan, Regional Counsel, Office of Criminal Conflict and Civil Regional Counsel, Fourth District, and Richard G. Bartmon, Assistant Regional Counsel, West Palm Beach, for appellant.
James Uthmeier, Attorney General, Tallahassee, and Heidi L. Bettendorf, Senior Assistant Attorney General, West Palm Beach, for appellee.
DAMOORGIAN, J.
Randolph Thomas (“Defendant”) appeals his conviction and sentence for two counts of sexual battery while standing in a position of familial or custodial authority of a person 12 years or older, but under 18 years old, of his girlfriend’s daughter (“Victim”). Defendant appeals on several grounds, including the trial court’s admission of DNA expert testimony using Bayes’ Theorem statistical analysis at trial regarding Defendant’s paternity of Victim’s child. We affirm the conviction in all respects and write only to address why we reject Defendant’s challenge of the use of Bayes’ Theorem.
By way of background, in 2015, Victim reported that Defendant, who was Victim’s mother’s live-in boyfriend, had been sexually assaulting her since she was a minor. Victim also alleged that Defendant was the biological father of Victim’s then seven-year-old son (“Child”). Police took DNA samples from Victim, Child, and Defendant, and DNA tests concluded that Defendant and Child’s DNA matched at all fifteen (15) genetic markers available to the lab at the time of testing. 1 Defendant was arrested and charged with two counts of sexual battery while standing in a position of familial or custodial authority of a person 12 years or older, but under 18 years old, Count 1 – penile penetration, and Count 2 – oral penetration.
The case proceeded to a jury trial. In relevant portion, Defendant sought to exclude at trial the testimony of the State’s DNA expert witness based upon his reliance on Bayes’ Theorem, a mathematical formula commonly used in paternity testing. The trial court denied Defendant’s motion in limine and allowed the State’s expert testimony. The State’s expert explained on cross-examination, in relevant portion, the use of prior probability under Bayes’ Theorem as follows:
[P]rior probability is a ratio of two things. One. We know that every child has to have a male and a female parent. So 50 percent of the DNA came from a man.
The other probability is, what’s the probability that any man could be the father? And we assign that to be 50 percent. So it’s 50 percent that any man could be the father or the fact that 50 percent of his DNA came from the father.
What that does is .5 over .5, is one. And it means therefore, the prior probability has no effect whatsoever on the posterior probability. It renders it effectively zero, and allows us then to just consider the genetic evidence.
The State’s expert explained at the conclusion of his analysis that “[t]he results were that it is 287 million times more likely that [Defendant] is the biological father of [Child] rather than a random untested man. That translates to a probability of paternity of greater than 99.9999996 percent.” At the end of the trial, the jury ultimately found Defendant guilty as charged on both counts.
On appeal, Defendant argues that the trial court abused its discretion by admitting the State’s expert testimony because it “assumed a 50% probability that [Defendant] had intercourse with Victim and was [the] father of her child” which “eviscerated [the] presumption of innocence” and abrogated Defendant’s rights, relying on State v. Skipper, 637 A.2d 1101 (Conn. 1994). The State responds that modern courts have rejected the premise that the use of Bayes’ Theorem violates the presumption of
1 The 16th marker was gender, which is not used in paternity calculations.
2 innocence, and that the use of “prior probability” under Bayes’ Theorem provides a neutral assessment of paternity in criminal cases. For the reasons outlined below, we decline to accept Defendant’s argument, and instead adopt the modern approach to Bayes’ Theorem outlined in Griffith v. State, 976 S.W.2d 241, 242 (Tex. App. Amarillo 1998), and its progeny.
Both sides, as well as the trial court, acknowledge that the only Florida case found through research that references Bayes’ Theorem is Brim v. State, 779 So. 2d 427 (Fla. 2d DCA 2000), but Brim did not directly address the issue at hand. Because this is a case of first impression in Florida, we look to other jurisdictions for guidance on the issue of prior probability using Bayes’ Theorem.
Defendant relies upon cases from the 1980s and 1990s in Connecticut and Wisconsin to support his position. In relevant portion, the Connecticut Supreme Court held, in Skipper, that the trial court reversibly erred by admitting paternity evidence that had utilized Bayes’ Theorem because it “permitted the introduction of evidence predicated on an assumption that there was a fifty-fifty chance that sexual intercourse had occurred in order to prove that sexual intercourse had in fact occurred. . . The fifty-fifty assumption that sexual intercourse had occurred was not predicated on the evidence in the case but was simply an assumption made by the expert.” Skipper, 637 A.2d at 1106.
Defendant also relies upon State v. Hartman, 426 N.W. 2d 320, 326 (Wisc. 1988). In Hartman, the Wisconsin Supreme Court majority held, in relevant portion:
[T]he probability of paternity is calculated based upon the assumption that the mother and putative father have engaged in sexual intercourse at least once during the period of possible conception. Because the probability of paternity assumes that sexual intercourse has occurred, it is improper to use this statistic to prove that sexual intercourse has occurred.
Id. (internal quotation and citation omitted).
However, Justice Steinmetz’s dissent in Hartman, in relevant portion, notes that any potential issues with jury confusion regarding prior probability with Bayes’ Theorem could be resolved by its admission through an expert witness:
3 The requirement of an expert to introduce the test results, coupled with the opportunity by defense counsel to examine the expert and thus explain the method of calculating paternity, sufficiently compensates for any alleged problems inherent in the Bayesian formula.
The 50 percent prior chance assumption does not require shifting the burden of proof to the defendant and is not an impermissible assumption; rather, it is part of a scientific theorem and the jury should be so told. Contrary to the majority’s assertion, the assumption is valid because the defendant has been named as the male having had sexual intercourse with the mother. In this case, the mother was the victim of a sexual assault and accused the defendant of being the father of her child as a result of his act. Therefore, the defendant was not randomly selected, but rather, had been named in a sworn complaint or information. That is, the “assumption” that is used in the probability of paternity computation was not presented into evidence in a vacuum, but was instead admitted only after a factual basis for the statistic was offered into evidence.
Id. at 327.
Additionally, as the State points out, and Defendant concedes, more recent cases from other states have allowed admission of paternity evidence involving the use of Bayes’ Theorem. Griffith, for example, relied on Justice Steinmetz’s dissent in Hartman and allowed the use of Bayes’ Theorem. 976 S.W.2d at 248–49. In Griffith, the state called a DNA testing expert witness who testified about DNA testing in general, in the case at hand, and as part of that, discussed Bayes’ Theorem. The Griffith court described the expert testimony regarding Bayes’ Theorem as follows:
The probability of paternity was calculated by using Bayes’ Theorem. Bayes’ Theorem, according to [the expert witness], states that prior to the testing, there is a prior probability of paternity. He stated that courts in the United States typically use a .5 or 50% prior probability because it is a neutral probability. The .5 prior probability indicates that the tested male either is or is not the father. [The expert] further testified that this calculation was a generally accepted principle, and was standard methodology in parentage testing, having been used for twenty or thirty years.
4 [The expert] further explained the theory and methodology involved in DNA testing generally. After explaining how DNA functions and how the tests are conducted, [the expert] discussed the specific results in this case. [The expert] stated that using the .5 prior probability, which was the standard prior probability reported in parentage tests, that appellant’s probability of paternity was 99.99%. At this point, the State passed [the expert] as a witness, and defense counsel cross- examined him.
Id. at 245. On cross-examination, the expert reiterated that the .5 prior probability was neutral and did not assume that the defendant was guilty or more likely guilty than not. Id. The expert also explained that, for example, if .1 had been used instead of .5, the probability of paternity, though represented by a lower number, would still have shown that the defendant was a match for the child at 6 genetic test sites. Id. at 245–46.
The Griffith court found the trial court properly admitted the paternity statistical evidence. The court also specifically rejected the holdings in Hartman and Skipper, explaining that it did not agree that the basic assumption that intercourse occurred was implicit in the statistic. Id. at 247–48. The older cases both relied, at least in part, on Peterson, A Few Things You Should Know About Paternity Tests (But Were Afraid To Ask), 22 Santa Clara L. Rev. 667 (1982). Id. at 247–48. The Griffith court noted that the law review article was written by an attorney and professor, as opposed to a statistician or geneticist, and the author had not relied upon any direct legal or scientific evidence to show that Bayes’ Theorem assumed intercourse did occur. Id. Instead, the court noted, “[l]ogically, the prior probability assumes intercourse could have occurred and thus the putative father could be the actual father, but the statistic does not necessarily assume intercourse did occur.” Id. at 248 (emphasis in original). In addition to citing Justice Steinmetz’s dissent in Hartman, the Griffith court found the expert testimony to be credible that the .5 prior probability was neutral and nothing about the statistic shifted the burden of persuasion to the defendant. Id. at 248–49.
Other courts have adopted Justice Steinmetz’s dissent in Hartman and the Griffith court’s reasoning. See Butcher v. Commonwealth, 96 S.W.3d 3, 9 (Ky. 2002) (holding the use of Bayes’ Theorem was allowed where the defendant had ample opportunity to question the expert witness on the use of prior probability and noting that prior probability under Bayes’ Theorem “merely acknowledges that intercourse preceded the birth of the child, and there is an equal chance that another individual engaged in that intercourse with the mother as there is a chance that the alleged father
5 did”); Jessop v. State, 368 S.W. 3d 653, 674–76 (Tx. App. Austin 2012) (holding, in relevant portion, that the prior probability invoked in Bayes’ Theorem is a statistically neutral probability and using Griffith and Butcher to reject the appellant’s contention that a prior probability must be zero instead of 0.5); People v. Gonis, 123 N.E. 3d 55, 60–61 (Ill. App. 3d Dist. 2018) (relying on Griffith, Butcher, and Jessup in holding that the trial court did not err in allowing in evidence of paternity based on Bayes’ Theorem as it did not violate the presumption of innocence). These cases have also specifically rejected Skipper and Hartman. See Butcher, 96 S.W.3d at 9–10; Jessop, 368. S.W. 3d at 676; Gonis, 123 N.E. 3d at 61– 62.
We adopt the reasoning of Griffith and its progeny. In Griffith, the child was a DNA match for the defendant at 6 genetic markers. More strongly, in the instant case, Child was a DNA match for Defendant at all 15 genetic markers tested. Notably, Defendant does not appear to challenge the DNA match at the 15 genetic markers, only the use of probability of paternity percentage arrived at using Bayes’ Theorem.
The State’s DNA expert testified that he found no errors with the DNA testing itself in this case, explained genetic testing in general and in the underlying case, and explained the mathematical formula used to arrive at his conclusion that the probability that Defendant is Child’s biological father is greater than 99.9999996 percent. The State’s expert explained how the use of the .5 prior probability is neutral and has no effect on posterior probability.
The presumption of innocence does not require a jury to assume it was impossible for a defendant to commit the crime charged. Rather, it requires the jury to assume as a starting proposition that the defendant did not commit the crime, until proven otherwise. The probability of paternity . . . is merely a way of expressing and interpreting the actual DNA test results. Thus, the statistic itself does nothing to shift the burden of going ahead to the defendant.
Griffith, 976 S.W.2d at 249. Accordingly, we find that the use of Bayes’ Theorem did not violate Defendant’s presumption of innocence and affirm the trial court did not err in denying Defendant’s motion in limine.
Affirmed.
MAY and GERBER, JJ., concur.
6 * * *
Not final until disposition of timely filed motion for rehearing.