The state appeals a judgment denying its motion to set aside the jury's verdict that D.P.K. is not the biological father of the child in question or, in the alternative, order a new trial, arguing that the jury should have been instructed that D.P.K. is rebuttably presumed to be the father pursuant to sec. 767.48(lm), Stats. Because two different sets of blood tests with conflicting statistical results as to the probability of parentage were presented at trial, we hold that the sec. 767.48(lm) presumption óf paternity is inapplicable and affirm the trial court.

In July 1982, the state, by the Langlade County district attorney, filed a petition for a determination of paternity, alleging that D.P.K. is the child's father. After court-ordered blood samples of the mother, the child and D.P.K. were drawn, Badger Labs of the American Red Cross reported the test results showed that the paternity of D.P.K. is unlikely. The state then agreed to dismiss the case.

In April 1986, Badger Labs notified the state that a clerical error had resulted in an incorrect interpretation of the blood test data and that, in fact, there was a 97.06% chance that D.P.K. is the father. The state then filed a new petition for determination of paternity alleging D.P.K. to be the father.1

*432Additional blood tests of the mother, child and D.P.K. were subsequently ordered. Genetic Design, Inc., reported the results of these samples as indicating that the probability of paternity is 99.45%.

On March 29, 1990, at the pretrial conference, the state moved that sec. 767.48(lm), Stats.,2 applies to this action and that, accordingly, the jury be instructed as to the rebuttable presumption of paternity created by sec. 767.48(lm) as contained in Wis J I — Civil 5001 (1989).3 The trial court denied the state's motions, reasoning that because this action was commenced before the effective date (October 1, 1987) of sec. 767.48(lm), the statutory presumption did not apply to this action. Following the rejection of the state's motion, the jury reached a verdict that D.P.K. is not the biological father. Because we affirm on different grounds, we need not reach the issue whether sec. 767.48(lm) retroactively applied before the effective date of the statute.

Section 767.48(lm), Stats., provides for a rebuttable presumption of paternity where an alleged father is shown to have a statistical probability of 99.0% or *433higher. The state contends that this presumption should apply to this action because the result of the second set of blood tests indicates that the probability of the accused's parentage is 99.45%. We disagree. Where more than one set of blood test results were presented to the jury, the sec. 767.48(lm) presumption did not apply where only one result reached the required statutory minimum of 99%. The record discloses no request to instruct the jury on the presumption if it chose to accept the higher test result, nor did the parties present evidence bearing on the superiority of one test over the other. We therefore do not address the propriety of a modified presumption instruction in such a case.

The application of a statute to a set of facts is a question of law we review de novo. Sturgis v. Town of Neenah Bd. of Canvs., 153 Wis. 2d 193, 198, 450 N.W.2d 481, 483 (Ct. App. 1989). We must give effect to the legislative intent and, to do so, we first look to the language of the statute itself. Id. Section 767.48(lm), Stats., states: "Under sub. (1), if the blood tests show that the alleged father is not excluded and that the statistical probability of the alleged father's parentage is 99.0% or higher, the alleged father shall be rebuttably presumed to be the child's parent." (Emphasis added.)

The statute is ambiguous because it is capable of being understood by a reasonably well-informed person in more than one way. See Sturgis, 153 Wis. 2d at 198, 450 N.W.2d at 483. When sec. 767.48(1m), Stats., is read in conjunction with subsec. (1), to which it refers, it is unclear whether "blood tests" refers to the statistical result of one set of blood tests done on the mother, child and alleged father, or whether it refers to the results of more than one set of blood tests.

*434We must interpret statutes so as to avoid absurd or unreasonable results. Sturgis, 153 Wis. 2d at 198, 450 N.W.2d at 483. It would be unreasonable to apply the sec. 767.48(1m), Stats., presumption in a paternity action where, as here, the results of one set of blood tests on the mother, child and alleged father show a statistical probability of the alleged father's parentage to be less than 99.0%, while the results of another set of tests show the probability to be 99.0% or higher. Where the results of multiple tests are admissible at trial, there is no reason for the trial court to conclude as a matter of law that the higher test result is the accurate one. Accordingly, we conclude that the phrase "blood tests” as used in sec. 767.48(1m) refers to the statistical result of a set of blood samples taken from the mother, child and alleged father and the presumption applies only where each set of admissible test results is 99.0% or higher.

Here, the statistical results of both sets of blood tests were presented to the jury at trial. Experts by the state testified in support of the accuracy of both tests, while D.P.K.'s expert refuted the accuracy of the results as well as the conclusions drawn from them. The credibility of the test results was properly left to the jury.

By the Court. — Judgment affirmed.