ELLINGTON, J.

Joseph Baker was convicted of first degree child molestation. The trial court admitted evidence of Baker's alleged molestation of his daughter 11 to 15 years earlier to show a common scheme or plan and to rebut his accident defense. We find the evidence admissible for both purposes, and affirm.

Facts

On the evening of the alleged molestation, Baker held a "slumber party" in his camper with his girlfriend's daughters, eight-year-old N.H., and her two sisters. N.H., wearing pajamas, shorts, and underpants, slept in bed with Baker. Baker rubbed her back as she went to sleep. She awoke later when she felt Baker rubbing her "privates" through her clothes. She immediately left to tell her mother what happened. Later N.H.'s mother told her that it was an accident because Baker was asleep, but N.H. thought he was awake.

Before trial, the State sought to admit testimony about Baker's other alleged assaults on sleeping children to show a common scheme or plan to sexually assault sleeping children and to show that the molestation was not an accident. At a preponderance hearing, the State presented testimony from Baker's daughter, Kathy Baker Mitchell, and Baker's half-sister, June Good.

Kathy Mitchell testified that when she visited Baker between the ages of seven or eight and eleven, he would sleep in bed with her, rub her back before she went to sleep, and she would awaken during the night with her underwear pulled down and his hand between her legs. The defense attacked Mitchell's credibility by way of testimony from other members of Baker's family. The court also heard testimony from a prosecutor who chose not to pursue charges against Baker for Mitchell's alleged molestation because Mitchell could be impeached on collateral issues.

The court found by a preponderance of the evidence that the touching of Kathy Mitchell occurred, that the incidents were substantially similar, and that her testimony was admissible to show that the touching of N.H. was part of a common scheme or plan and was not an accident. The court balanced the relevance of the evidence against its prejudice and found it admissible under State v. Lough, 125 Wn.2d 847, 889 P.2d 487 (1995), ruling that the abuse of sleeping children was similar to the drugging of victims because the vulnerable state of the victims created a diminished ability to describe what occurred. The court ruled that the necessity of this testimony was great, and that its relevance outweighed any unfair prejudice. The court limited Mitchell's testimony to the touching that occurred when she was around the same age as N.H., excluding testimony from Mitchell about later touchings or other bad acts.

In addition, the court considered June Good's testimony that Baker had raped her when she was a child. The court found that Good's allegations were also proven by a preponderance of the evidence, but excluded her testimony because it was not as reliable or necessary as Mitchell's testimony.

At trial, Baker testified that he did not believe N.H. would lie about his touching her, and that if he did touch her, it was done accidentally while he was asleep. Baker also testified that he never molested Kathy Mitchell. Before Mitchell testified, the court instructed the jury that it could consider her testimony only to determine whether the touching of N.H. was part of a common scheme or plan or was an accident.

As part of the jury instructions, the court again instructed the jury that Mitchell's testimony was only admissible to determine whether the alleged touching was part of a common scheme or plan and whether it was an accident. The court instructed the jury that it could not consider Mitchell's testimony as evidence of Baker's character or as evidence that Baker acted in conformity therewith in this case. The jury found Baker guilty.

Analysis

Baker contends that the trial court erred in admitting Kathy Mitchell's testimony about his prior bad acts. Before admitting evidence of other wrongs under ER 404(b), a trial court must (1) find that a preponderance of evidence shows that the misconduct occurred; (2) identify the purpose for which the evidence is being introduced; (3) determine that the evidence is relevant; and (4) find that its probative value outweighs its prejudicial effect. Lough, 125 Wn.2d at 852. In doubtful cases, the evidence should be excluded. State v. Smith, 106 Wn.2d 772, 776, 725 P.2d 951 (1986).

Baker first argues that a preponderance of the evidence did not show that he molested Mitchell. A court's preponderance determination will be upheld if it is supported by substantial evidence. State v. Benn, 120 Wn.2d 631, 653, 845 P.2d 289, cert. denied, 510 U.S. 944 (1993).

Here, the trial court considered Mitchell's testimony that during her visitation with Baker between the ages of seven and eleven, he slept in bed with her, rubbed her back until she went to sleep, and would then pull her underwear down and place his hands between her legs. Family members disputed whether Baker slept with Mitchell at their homes, but Baker had unrestricted visitation with Mitchell. In addition, contrary to Baker's contention on appeal, the prosecutor chose not to pursue charges based upon Mitchell's allegations because she could be impeached on collateral issues, not because her allegations were not credible.

Mitchell's testimony provided substantial evidence supporting the trial court's preponderance determination. To the extent that this determination involved resolution of credibility issues, they were for the trial court. When any reasonable view of disputed facts supports the trial court's finding, it will not be disturbed on appeal. See State v. Roth, 75 Wn. App. 808, 816, 881 P.2d 268 (1994), review denied, 126 Wn.2d 1016 (1995).

After making the preponderance determination, the trial court must identify a proper purpose for the admission of the evidence under ER 404(b). Lough, 125 Wn.2d at 853. The trial court here admitted Mitchell's testimony to show that the touching of N.H. was part of a scheme or plan, and to rebut Baker's accident defense. Baker argues that neither purpose provided a proper basis for admission.

Baker contends that the events Mitchell described were too remote in time and were not sufficiently similar to show a common scheme or plan. The scheme or plan exception applies "when an individual devises a plan and uses it repeatedly to perpetrate separate but very similar crimes." Lough, 125 Wn.2d at 855. Such evidence is admissible when it shows that a person committed "`markedly similar acts of misconduct against similar victims under similar circumstances.'" Lough, 125 Wn.2d at 856 (quoting People v. Ewoldt, 7 Cal.4th 380, 399, 867 P.2d 757, 767, 27 Cal.Rptr.2d 646, 656 (1994)); see also State v. Carleton, 82 Wn. App. 680, 683, 919 P.2d 128 (1996). Conduct is sufficiently similar when the similarity indicates design, not merely coincidence. Lough, 125 Wn.2d at 860.

The Lough court found sufficient similarity in allegations by four other women that they dated the defendant, and that he drugged and then raped them. Lough, 125 Wn.2d at 861. The Carleton court found sufficient similarity in allegations by another teenage boy that the defendant befriended him, revealed a homosexual alternate personality, and later molested him, but excluded testimony by another boy who was not told about the alternate personality. Carleton, 82 Wn. App. at 683-84.

Here, eight-year-old N.H. was the daughter of Baker's live-in girlfriend. On the night of the alleged molestation, N.H. slept in a bed with Baker, he rubbed her back until she went to sleep, and she awoke when she felt him massaging her vagina with his hand through her clothes. Mitchell testified that during visits with Baker from ages seven to eleven, she slept in bed with him, he rubbed her back until she went to sleep, and she would awaken with her underwear down and Baker's hand, head, or penis between her legs. The strong similarities in the relationships, the ages, the scenario, and the touchings described by N.H. and by Mitchell are indicative of design rather than coincidence. Unless the remoteness in time is determinative, the evidence is admissible under ER 404(b).

The lapse of time is not alone determinative, however. The trial court must consider common scheme or plan evidence in terms of the closeness in time, place, and modus operandi (meaning similarity, not signature). Lough, 125 Wn.2d at 858 (citing State v. Wermerskirchen, 497 N.W.2d 235, 240 (Minn. 1993)). A time lapse may decrease the likelihood of a common scheme or plan, but such a lapse may not be significant if the conduct is part of a pattern of similar misconduct over a number of years, or if intervening events make the time lapse insignificant. Lough, 125 Wn.2d at 858 (citing Wermerskirchen, 497 N.W.2d at 243 n.3).

Here, while the elapsed time weighed against admission, the other factors strongly favored admission. Further, prior bad act evidence is particularly relevant when the circumstances of the alleged crime create difficulty in assessing the credibility and memory of the complaining witness. Lough, 125 Wn.2d at 862 (prior act evidence particularly relevant because the use of drugs diminished alleged victim's recall). Both Mitchell and N.H. testified to being molested while they were sleeping with a trusted adult.

We review such relevancy determinations for an abuse of discretion. Lough, 125 Wn.2d at 861. Here, the trial court determined that, as in Lough, Mitchell's testimony was relevant to whether the molestation of N.H. occurred. The court found that Mitchell's testimony was necessary to support N.H.'s testimony because the circumstances of the alleged molestation of N.H. in the middle of the night while she was asleep were similar to the drug-induced state in Lough. The trial court did not abuse its discretion in determining that Mitchell's testimony was relevant to show a common scheme or plan.

Baker also contends that the evidence was not admissible to rebut his claim that any touching was an accident because it does not meet the threshold of noncoincidence. Evidence of prior acts is admissible to rebut the defense of accident. Roth, 75 Wn. App. at 818. In determining whether evidence of a prior act is admissible and relevant to rebut such a defense, however, the similarity of the acts must meet a threshold of noncoincidence. Roth, 75 Wn. App. at 819-20 (citing generally Eric D. Lansverk, Comment, Admission of Evidence of Other Misconduct in Washington To Prove Intent or Absence of Mistake or Accident: The Logical Inconsistencies of Evidence Rule 404(b), 61 WASH. L. REV. 1213, 1226-28 (1986) (recurrence of act creates improbability of coincidence, which becomes evidence of lack of accident)).

Since Baker asserted that any touching of N.H. was an accident, the question is whether the evidence of his alleged molestation of Mitchell was sufficiently similar to be relevant to rebut that claim and establish noncoincidence. As noted above, the alleged molestation of Mitchell was indeed similar to that alleged by N.H. And the existence of the similar allegation of molestation by Mitchell makes it less likely that Baker's behavior with N.H. was an accident. The court did not abuse its discretion in determining that Mitchell's testimony was admissible and relevant to show a common scheme or plan and to rebut Baker's defense of accident.

Finally, Baker contends that the trial court erred in balancing the probative value of Mitchell's testimony against its potential for prejudice. After the court determines that evidence of other acts is relevant, it must weigh the probative value of the evidence against its prejudicial effect. Lough, 125 Wn.2d at 862. When the other acts are uncharged offenses, they must have substantial probative value. Lough, 125 Wn.2d at 863. The weighing must appear on the record, and is reviewed for an abuse of discretion. Lough, 125 Wn.2d at 862-63.

When the allegation is child molestation, evidence of prior similar acts creates a likelihood that the jury will convict based solely upon character. State v. Krause, 82 Wn. App. 688, 696, 919 P.2d 123 (1996), review denied, 131 Wn.2d 1007 (1997). Because the inference is so prejudicial, some courts have held that the prejudice cannot be cured by an instruction. See Krause, 82 Wn. App. at 696. On the other hand, prior similar acts are highly probative of common scheme or plan, particularly in child sex abuse cases, because of (1) the secrecy in which the acts occur, (2) the vulnerability of the victims, (3) the lack of physical proof of the crime, (4) the degree of public opprobrium associated with the accusation, (5) the unwillingness of victims to testify, and (6) the jury's general inability to assess the credibility of child witnesses. Krause, 82 Wn. App. at 696 (citing Wermerskirchen, 497 N.W.2d at 240-41).

In affirming the trial court's balancing, the Lough court determined that (1) the evidence was highly probative of common scheme or plan, (2) the need for the testimony was great because the defendant's acts in drugging the victims rendered them unable to remember the events clearly, and (3) the court gave a limiting instruction. Lough 125 Wn.2d at 864; see also Krause, 82 Wn. App. at 696-97.

Here, the trial court acknowledged that the evidence was highly prejudicial, but determined that the prejudice was overcome by the necessity for the testimony under the facts of the case, including that the alleged molestation occurred while N.H. was asleep. Under these circumstances, the court reasonably concluded the need for the testimony outweighed its prejudice.

After conducting careful and correct pretrial procedures, the court gave similarly careful and thoughtful consideration to the posture of the parties' evidence and theories and reached well-reasoned conclusions. The trial court also minimized the prejudicial impact by giving the limiting instruction twice and by excluding other testimony that was less probative. The trial court did not abuse its discretion in determining that the evidence's probative value outweighed its prejudicial effect.

Affirmed.

COLEMAN and BECKER, JJ., concur.