This is an appeal from an order of the district court which affirmed the magistrate court’s dismissal of a petition filed by Health and Welfare under the Child Protective Act, Title 16, Chapter 16, Idaho Code. The petition sought temporary custody of a four-year-old child of the Symes, or in the alternative, protective supervision of the child in her own home. We reverse in part with instructions for further proceedings.
The Symes had been married approximately ten years and had three children, including four-year-old (X) who is the subject of the instant action. Donna Syme became suspicious that there was sexual activity between her husband and her four-year-old daughter. She left her husband and filed for divorce, which was granted on October 12, 1983. The divorce decree does not address the alleged sexual activity. In the divorce decree the Symes were awarded joint legal custody of their three children, including (X), with Mrs. Syme awarded physical custody subject to Mr. Syme’s right of visitation every other weekend, every other holiday, and four weeks during the summer.
During the pendency of the divorce, Mrs. Syme took (X) to a family physician whose examination was negative as to any sexual molestation of (X). Mrs. Syme also took (X) to a psychologist and to a counselor, both of whom reported they could not detect any evidence of sexual molestation. Mrs. Syme then arranged with a State mental health unit to have (X) seen by a psychologist, but (X’s) lack of verbal skills prevented him from reaching a conclusion as to the accusation of molestation. In September 1983, Mrs. Syme, through the Department of Health and Welfare, arranged for (X) to be placed in an incest group program, and (X) was thereafter involved in individual counseling and group counseling.
In October 1983, (X) was interviewed by Dr. McQueen, a psychologist, wherein (X) indicated that she had been involved in sexual activity with her father, her mother, and one of her brothers. That indication was made by (X) through the use of what are known as anatomically correct dolls, and by nodding her head “yes” or “no” in response to questions. Dr. McQueen conducted another videotaped interview of (X) on October 17, 1983. The tape was admitted into evidence at the hearing and shows (X) using the anatomically correct dolls. (X) indicated in the videotape that there were sexual “secrets” between she, her mother, her father, and her brother. On November 23, 1983, Dr. McQueen and a social worker conducted still another interview with (X) which was also videotaped. The results of that interview were similar to the previous ones except that (X) also implicated her second brother.
Mrs. Syme was out of town when this last interview took place, and (X) was about to spend the Thanksgiving weekend with her father. A Caldwell police detective declared (X) to be in imminent danger and she was temporarily placed in a foster home. On November 28, the present petition was filed under the Child Protective Act. After a shelter care hearing was held that same day the temporary custody of (X) was placed in the mother, and the father was prohibited from having any contact with the child until after an adjudicatory hearing. No criminal charges were, or ever have been, filed against the father.
At the request of Health and Welfare a Dr. Schaffer conducted a physical exam of (X) on November 29,1983, which was negative and inconclusive as to any sexual abuse or molestation. The father submitted to a polygraph examination on December 14, 1983, which resulted in an indi*46cation that the father was telling the truth when he denied sexual activity between him and her. A copy of that polygraph report was provided to other counsel, but no stipulation was reached as to the admission of that polygraph examination at trial.
An adjudicatory hearing was held on the petition before a magistrate. The State made a motion in limine to exclude any evidence of the polygraph exam, but that motion was denied. At the hearing Dr. McQueen testified that in his opinion (X) had been sexually abused by family members, and a social worker testified as to her observations of (X) in the incest group program.
Testifying on behalf of the respondent father was one John Dawson, a counselor who had interviewed (X). He contradicted the opinion of, and the techniques used by, McQueen. The father of the child also testified and denied any sexual involvement with (X). The polygraph examiner was permitted to testify that his exam indicated that Syme’s denial of involvement in sexual activity with (X) was truthful.
The magistrate concluded that the State failed to demonstrate the allegations in the petition by a preponderance of the evidence, that (X) did not come within the purview of the Child Protective Act, and he dismissed the case noting:
“There is no firm evidence before the court. Only the testimony of parties who have interviewed the child, and their own interpretation of the possibility of endangerment.
“The statute requires proof by a preponderance of evidence. There simply is no proof, opinions yes, speculations yes, but firm proof, no.”
That decision of the magistrate was appealed by Health and Welfare to the district court which affirmed the decision of the magistrate. Thereafter the prosecuting attorney of Canyon County purported to file an appeal to this Court from the order of the district court, affirming the magistrate’s decision. An amended notice of appeal was filed by the Attorney General’s office. Respondent moved to dismiss the appeal as being untimely filed since the action of the Office of the Attorney General was not taken within the required time limitation for filing an appeal. The Court has previously considered that motion to dismiss and it has been denied.
The appellant Health and Welfare contends that the district court erred in affirming the magistrate’s decision since the magistrate’s findings of fact were clearly erroneous. I.R.C.P. 52(a) requires a trial court to make findings of fact and separate conclusions of law in actions tried without a jury. Here the magistrate specifically found as follows:
“In a report to the court from the Department of Health and Welfare, filed December 2, 1983, the allegation is made that at that interview the child indicated a sexual contact with her father and brother. The child is not quite 5 years of age. From the testimony it appears there was no verbalization from the child. Apparently statements were made to her and she either nodded “yes” or “no”.
“It also appears she, [X], was placed in the SANE program. The report alleges here she verbalized the incidents, but such testimony was not presented to the court, only that the child finally interacted.
“Dr. McQueen’s interview, the interviews with the other staff members, the SANE program and peer associations at these meetings and the numerous interrogations this child has been subjected to, at the very least, gives this Court concern of an extreme contamination of any information she (the child) might supply because of the statements and suggestions she has experienced in all these examinations. Not only does there exist the very real possibility of contamination of any information the youngster might give, but there was another witness, Dr. John Dawson, with credentials equal to that of the State’s witnesses, who contradicted the methods used by Dr. McQueen and his findings were dramatically opposed to Dr. McQueen’s as to *47whether any activities had occurred after he interviewed [X].
“There is no firm evidence before the court. Only the testimony of parties who have interviewed the child and their own interpretation of the possibility of an endangerment.
“The statute requires proof by a preponderance of evidence. There simply is no proof, opinions, yes, speculations, yes, but firm proof, no.”
Appellant Health and Welfare asserts that the findings of the magistrate overemphasize a lack of verbalization by the child, and further assert that such verbalization is unnecessary in cases of a very young child alleged to have been sexually abused. As noted in our recent decision of State v. Snapp, 110 Idaho 269, 715 P.2d 939 (1986), wherein we discussed, albeit in a criminal context, admission of testimony relating to child sexual abuse syndrome, very young children may not possess verbalization skills and a child who has been the victim of sexual molestation may for a variety of reasons be unable to discuss the specific instances. For these reasons, sexual abuse has been called the “perfect crime.” Bowes, Libai’s Child Courtroom,: Is it Constitutional?, 7 JJuv.L. 31, 39 (1983).
It was indicated by a witness for Health and Welfare at the hearing that in her opinion there is an important distinction between “non-verbal” and “non-responsive” and that in the interviews with (X) she was never non-responsive but rather was alert, interested, and communicative.
We need not, however, base our decision in this case on the distinction, if any, between non-verbal and non-responsive since the record reflects both through the testimony of Dr. McQueen and through the videotape of the November 23, 1983 interview that (X), to some extent, did engage in verbalization. For example, Dr. McQueen testified that (X) was able to verbalize as to the family names used to describe sexual parts, and in the videotaped interview (X) was able to verbalize as to the “secrets” between her and her father and her brother, where they took place, and when they took place.
The videotape was admitted in evidence at the hearing, but the findings of the magistrate make no reference thereto. The magistrate may well have ignored the videotape because, as he found, the court was concerned, “of an extreme contamination of any information she (the child) might supply because of the statements and suggestions she has experienced in all these examinations.” The court also in its findings indicated:
“Not only does there exist the very real possibility of contamination of any information the youngster might give, but there was another witness, Dr. John Dawson, with credentials equal to that of the State’s witnesses, who contradicted the methods used by Dr. McQueen and his findings were dramatically opposed to Dr. McQueen’s as to whether any activities had occurred after he interviewed (X).”
We might therefore infer that the magistrate rejected the methods and the opinions of Dr. McQueen and rather adopted the opinions and findings of Dr. Dawson, and that in any event the magistrate believed that by the time the videotaped interview was conducted by Dr. McQueen that the child’s mind had been contaminated. However, in this particular case we believe it more desirable that the magistrate himself make clear why the verbalization described by Dr. McQueen and also shown in the videotaped interview by Dr. McQueen was not considered in the findings. Hence, we remand this case for a consideration of the verbalization evidence of Dr. McQueen and the verbalization evidence depicted in the videotaped interview. We do not require the magistrate to take any additional evidence, however he is not prohibited from doing so. If further evidence is to be taken before the magistrate this would call attention to the fact that the Idaho Rules of Evidence will apply and that I.R.E. 101 modifies I.J.R. 10 by making the rules of evidence applicable in all Child Protective Act proceedings ex*48cept temporary shelter care hearings. Hence, I.C. § 16-1608(b), “[t]he hearing shall be conducted in an informal manner,” is no longer governing in these matters. See comments to Rule 101.
Appellant Health and Welfare also asserts that the magistrate erred in admitting the testimony of the polygraph examiner who performed the polygraph test on Mr. Syme, the father of (X). At the in limine hearing the objection to that testimony was stated, “... we really don’t question [the examiner’s] ability to perform [a] polygraph, however we don’t feel that there is any authority to have that poly-, graph admitted into evidence in this case.” That objection was denied by the magistrate who stated:
“... In reviewing the Child Protective Act and the Y.R.A., our legislature gave the Court a great deal of leeway because they realize that the Court would have to hear areas involving, particularly in Child Protective Act cases, where it is based on the testimony of interviews of very young children, that sort of thing. My feeling is that both in the Y.R.A. and the C.P.A. that the legislature, and the legislative intent, was that these be informal hearings, they’ve even gone so far to say that, informal in nature where the Court may recess from time to time and I think that a single Judge hearing these matters, without a jury, must evaluate all of the evidence or all of the information, whether we call it evidence or information, that is brought before, because it is a broad area. Now, obviously, I think, the Court hears the evaluation of a psychologist, psychiatrist, possibly, at best, I think to [say] that these are absolutely scientific would be facetious because I don’t think that the gentlement who present these statements that they cab absolutely guarantee that their analysis is totally true. I think, what the Court has to look at is every bit of information that can be brought before it and then you’ll ... the parties will just have to have some faith that the Court will give such weight to each individual area in making a final decision. I’m not going to deny it, I think anything that this Court, under these circumstances, can be provided to make as intelligent a final decision as it possibly can. I’ll deny the State’s motion and I’ll consider that statement as well as any other witness’s statement and try to grant them such weight as I can in making an ultimate decision. At best it is a very difficult area for a Court at any time because as in so many of these cases we deal with minor children who cannot come forth before this Court and testify under oath and give exact information. I must take that information and the evaluation of the statements that they’ve made through a second party and their evaluation of what the child told them. So ultimately what this Court feels any source that is available to it is that much more helpful in making a final decision. So I’ll at this time overrule Mr. Breitsam-eter’s motion to quash or prohibit the testimony of this party in this matter.”
Respondents assert that Health and Welfare was not misled or suprised by the polygraph evidence, that the trial court did not materially rely on the evidence, and hence the admission of the polygraph testing does not constitute reversible error. Guillard v. Department of Employment, 100 Idaho 647, 603 P.2d 981 (1979); Erhardt v. Leonard, 104 Idaho 197, 657 P.2d 494 (Ct.App.1983). Counsel for Health and Welfare was provided a copy of the polygraph report approximately a month prior to the hearing, and hence we agree that Health and Welfare was not misled or su-prised by the polygraph evidence.
It is not clear from the record, however, whether the magistrate relied upon the polygraph evidence. There is no mention in the findings of the polygraph evidence or that Syme submitted to a polygraph examination. Likewise, in the findings there is no focus on Syme’s denial of sexual abuse or that the polygraph testimony buttressed his said denial. Rather, the magistrate in his findings, focused upon the possible contamination of the child’s mind from the numerous examinations and interviews.
*49Health and Welfare asserts that the findings of the magistrate substantially focus on the testimony of the defense witness, John Dawson, a psychological counselor, and in turn Dawson’s opinions were based on the results of the polygraph. While Dawson indeed testified that he considered the polygraph examination in forming his opinion, we find that Dawson relied on a number of other factors in his analysis, and the findings indicate that the magistrate focused his ruling more substantially on Dawson’s testimony as to these other factors. For example, Dawson testified that he could not substantiate that sexual abuse had occurred, and he questioned the validity of the diagnostic methods used by Dr. McQueen. He also indicated that the child appeared to have been questioned very frequently by a large number of people, and she was very capable of picking up cues and giving what she perceived to be the desired answers. Dawson further questioned the validity of the use of the anatomically correct dolls and noted that the reports of sexual abuse did not originate from the child, but upon the suspicions of the mother just prior to her seeking a divorce.
We hold that there is no showing of material reliance by the magistrate upon the polygraph evidence. Nevertheless, since this case must be remanded to resolve the videotape issue, we further hold that in this particular case the polygraph evidence need not be excluded. We agree with the concerns expressed by the magistrate in denying a motion to exclude the polygraph evidence and hold that those articulated concerns of the magistrate are legitimate.
In many cases of child sexual abuse there are no observers to the act other than the abuser and the victim, and the child is often very young and likewise often unable to adequately articulate the events. Equally troubling is an instance, as here, where it is suspected that the alleged victim is giving contaminated responses because of the number of interview and counseling sessions. No data has been submitted, and we find none, as to the use of anatomically correct dolls, as substantiating sexual abuse. In many cases, as here, physical examinations of the alleged victim are inconclusive. Too often cases rest solely on the testimony and reports of psychologists and counselors interpreting the child’s behavior and responses.
It is the respondent’s position that Health and Welfare is being hypocritical in arguing against the use of polygraph results in view of the esoteric tests conducted by Dr. McQueen which are no more well founded or scientifically based than is polygraph testing. While we do not necessarily agree with respondent’s position that Health and Welfare is being hypocritical, nevertheless we hold that fundamental fairness militates toward admitting the polygraph testimony in Child Protective Act cases in which there are accusations of sexual abuse, and the case against the defendant consists almost entirely of “evidence” received second hand from counselors and psychologists who have interviewed the alleged victim and who have then formed conclusions of sexual abuse based on actions or mental attitudes of the alleged victim. It is a type of evidence which would probably be admissible in no other type of case, and albeit the instant case is not cast in a criminal context, the ability to confront one’s accusers is highly curtailed, if not made impossible. See State v. Snapp, 110 Idaho 269, 715 P.2d 939 (1986).
We note further that hearings under the Child Protective Act are conducted by the court without a jury, and thus avoided is the common problem of an exaggerated popular opinion of the accuracy of such a technique with the potential to mislead a jury.
This Court has not previously addressed the question of admissibility of polygraph evidence in civil or criminal trials. Idaho, along with Rhode Island and Vermont,1 *50appear to be the only jurisdictions with no appellate decision on the issue of admissibility of polygraph evidence. In 22 states polygraph evidence is admissible upon stipulation, and in another 21 states polygraph evidence is not admissible in any circumstance. Four states, Louisiana, Massachu-settes, Michigan, and New Mexico, may admit polygraph evidence without stipulation. See Ansley, N., Quick Reference Guide to Polygraph Admissibility, Licensing Laws and Limiting Laws, (1984). The United States Supreme Court has not stated an opinion on polygraph evidence, and the Federal Circuit appears to be split, with four courts permitting such evidence on stipulation, five leaving the issue to the discretion of the trial court, and two excluding such evidence.
We note that the accuracy of polygraph examination is directly proportional to the skill of the examiner. Ferguson, R.J., Jr., and Allan L. Miller, The Polygraph in Court, 43 (1973). As was stated in People v. Anderson, 637 P.2d 354, 360 (Colo.1981), “polygraphy, albeit based upon a scientific theory, remains an art with an unusual responsibility placed on the examiner.” Also a consideration is the selection of the actual questions asked during the polygraph examination. People v. Anderson, supra, indicates that the crux of the polygraph technique is the development of neutral, control and relevant questions since all polygraph tests are preceded by a pretest interview in which background information is obtained from the subject and relevant questions are developed and reviewed with the subject. These questions were recently reviewed by the New Mexico Court of Appeals in State v. Anthony, 100 N.M. 735, 676 P.2d 262, 266, (Ct.App.1983), wherein it was stated:
“Experts are virtually unanimous that unless the ‘relevant’ questions are carefully formulated, the test results are suspect. A relevant question means a ‘clear and concise question which refers to specific objective facts directly related to the purpose of the examination and does not allow rationalization in the answer.’ ” (Citations omitted.)
These same concerns are dealth with in New Mexico’s rules of evidence wherein are established the minimum qualifications for a polygraph examiner and wherein are placed certain restrictions on the admissibility of the results. N.M.R.Evid. 707. See also State v. Dorsey, 88 N.M. 184, 539 P.2d 204 (1975); State v. Anthony, 100 N.M. 735, 676 P.2d 262 (Ct.App.1983). We have no such safeguard in our rules of evidence, nor is there a licensing requirement in Idaho for polygraphers.
In Idaho, Child Protective Act proceedings are civil in nature and adjudicated by a court without a jury. Because of the nature of the other type of evidence admitted into such proceedings, and the extreme difficulty as noted herein, of the proof, we hold that polygraph evidence' will be admissible in Child Protective Act proceedings where sexual abuse is alleged and the majority, if not sole, evidence offered to prove the sexual abuse is of the type in the case at bar. Such polygraph evidence will be admissible whether offered by or on behalf of the accused or the victim. The weight to be given polygraph examination testimony is in the discretion of the trial court in view of the qualifications and experience of the examiner and the particular questions asked of the subject.
This case is reversed and remanded to the magistrate court for consideration of the testimony of the verbalization of the alleged victim and the consideration of the videotape of the interview of the victim. Thereafter, the trial court shall enter specific findings of fact relating to his view of such evidence.
No costs allowed. No attorney fees on appeal.
DONALDSON, C.J., and HUNTLEY, J., concur.. In State v. Hamlin, 499 A.2d 45 (Vt.1985), the Vermont Supreme Court examined whether the *50lower court erred in refusing to decide whether the results of a polygraph test of the defendant’s companion were admissible at trial. The court did not focus on the proper standard for admissibility, and held that regardless of whether a trial court has discretion to admit polygraph examination results, the defendant failed to lay the proper foundation.