Caporale, J.
I. INTRODUCTION
This is a tort claims action brought under the provisions of Neb. Rev. Stat. § 81-8,209 et seq. (Reissue 1994) in which two minor girls, Bridgette A. Anderson and Candy S. Anderson, by and through their mother, the plaintiff-appellee, Cindy Anderson/Couvillon, allege that the defendant-appellant, Nebraska Department of Social Services, negligently placed the foster care of a minor boy in and with a friend of the mother and that as the proximate result, the boy sexually assaulted and damaged the girls. The district court entered partial summary judgment in favor of the girls on the issue of liability and upon trial awarded them damages as set forth in part IV(3) below. The department then appealed, and we, in order to regulate the caseloads of the two courts, on our own motion ordered the appeal transferred from the Nebraska Court of Appeals to this court.
As argued, the controlling errors assigned by the department may be summarized as claiming that the district court wrongly
(1) found the department liable, as its conduct neither violated a duty it owed the girls nor proximately caused their injuries;
(2) awarded hedonic damages as a separate and distinct category of damages; and (3) received and considered evidence concerning the computation of damages. The mother cross-appealed on behalf of the girls, assigning as error, in essence, the claimed inadequacy of the damages awarded.
[654]*654For the reasons which follow, we affirm the partial summary judgment on the issue of liability, but reverse the award of damages and direct a new trial on that issue. Our rationale in so mling makes consideration of the cross-appeal unnecessary.
II. SCOPES OF REVIEW
Summary judgment is proper only when the pleadings, depositions, admissions, stipulations, and affidavits in the record disclose that there is no genuine issue as to any material fact or as to the ultimate inferences that may be drawn from those facts and that the moving party is entitled to judgment as a matter of law. Horace Mann Cos. v. Pinaire, ante p. 640, 538 N.W.2d 168 (1995); Poppleton v. Village Realty Co., ante p. 353, 535 N.W.2d 400 (1995). On questions of law, an appellate court has an obligation to reach its own independent conclusions. Kropf v. Kropf, ante p. 614, 538 N.W.2d 496 (1995); McPherrin v. Conrad, ante p. 561, 537 N.W.2d 498 (1995).
IE. FACTS
The mother has three young daughters. Because her close friend, Terry Talle, and Talle’s husband enjoyed children but had none of their own, the mother permitted her daughters to spend time at the Talle home, often without her supervision.
In the fall of 1988, Talle and her husband applied for a foster care license from the department. Talle attended all required training sessions, where she learned the department’s rules and regulations. The training emphasized paperwork and financial matters but did not include discussion of proper discipline or control of older children, management of an older child’s anger, or the relationship between one suffering physical and sexual abuse and becoming an abuser. The Talles’ application ultimately was approved, and they accepted placement of two foster children for 2 months.
After those children left the Talle home, department caseworker Christy Johnson Strawder contacted Talle "about the possibility of placing then 13-year-old Ronald Heinen with the Talles for foster care. The department knew, but did not disclose to Talle, that Heinen was an exceptionally troubled boy. There existed a well-documented history of self-mutilation; [655]*655uncontrollable rage; emotional, physical, and sexual abuse; lying; stealing; and suicidal and homicidal ideations. Indeed, a memorandum prepared by the department’s Child Protective Services supervisor, Karen Griffith, noted that Heinen had been admitted to the Sandhills Psychiatric Unit of Great Plains Medical Center because he had said that he was going to commit suicide after killing his biological mother’s boyfriend.
Despite the foregoing, Strawder proceeded to seek a new foster home for Heinen. When Talle expressed interest, Strawder informed her only as to Heinen’s age, that he used hearing aids, that he was then hospitalized for a tonsillectomy, and that he had participated in a sexual fetish therapy group because of his fetish with women’s underwear. Talle was assured, however, that Heinen was not at risk of becoming a sexual offender.
A few days after talking with Strawder, Talle and the mother visited Heinen in the hospital. No one informed either Talle or the mother that in addition to the tonsillectomy, Heinen was also admitted because of his uncontrollable anger and suicidal and homicidal ideations. Following the hospital visit, Strawder took Heinen to spend the night at the Talle home. According to the mother, she informed Strawder at that time that her children spent a great deal of time at the Talle home, often without the mother, and asked Strawder if there were any “risks” with continuing the practice. The mother claims Strawder replied that there were no risks and, in response to another direct inquiry by the mother, stated that the children could be left unsupervised with Heinen. Strawder, however, testified that she recalled no conversation with the mother prior to Heinen’s being placed with the Talles.
On the basis of the sleepover and Strawder’s encouragement, the Talles agreed to act as foster parents for Heinen and executed a department “Child Placement Agreement,” in which the department agreed to “share with the foster care facility prior to placement and during placement, information known about the child’s life situation as appropriate and necessary . .
At no time did the department inform Talle about the opinion of Lynda Perry, a psychotherapist who was treating Heinen, that [656]*656foster care was not likely to be successful. Neither was Talle told that Heinen had a history of physical and verbal abuse that had caused his removal from prior foster care placements; that a former department caseworker believed the Talles did not possess the skill required to manage Heinen; that Heinen was thought to have suffered emotional, physical, and sexual abuse; that several caseworkers saw a correlation between being a victim of sexual abuse and becoming a perpetrator; that Heinen had homicidal and suicidal ideations; that caseworkers familiar with Heinen, his psychiatrist, and his psychologist believed Heinen’s case to be the most difficult they had ever seen; and that if left unsupervised, Heinen posed a distinct threat to children.
Perry advised Strawder in January 1990 that Heinen was at “high risk sexually” and should not be left alone with children. Strawder did not pass this information on to either Talle or the mother. One month later, Perry urged Strawder to make the Talles aware of Heinen’s sexual problems. Again, Strawder failed to communicate this to Talle or the mother.
Shortly thereafter, unbeknownst to Talle, Heinen began physically attacking Candy Anderson, who was then approximately 13 years old. He repeatedly attempted, with varying degrees of success, to kiss and fondle her; she, also with varying degrees of success, attempted to fight him off. In June 1990, while the children were in Denver with Talle and the mother, Heinen . began sexually assaulting Bridgette Anderson, who was then 7 years old.
On October 2, 1990, Perry informed Talle that Heinen had confessed to sexually assaulting the two girls. Talle confronted Heinen, and he repeated his confession to her. Talle thereupon immediately reported this to the department and informed the department that she was barring Heinen from her home.
Since Heinen’s removal from the Talle home, the two girls have exhibited symptoms of posttraumatic stress syndrome and have required, intense therapy. The younger girl was diagnosed with severe trauma associated with sexuai assault, has required intensive therapy, and has shown only minimal improvement. The older girl was also diagnosed with posttraumatic stress syndrome, and her psychologist observed that certain [657]*657preexisting conditions in her were significantly exacerbated by the assaults.
IV. ANALYSIS
With that factual background before us, we proceed to consider in turn each of the department’s arguments and controlling assignments of error.
1. Liability
In asserting that the district court erred in determining the department is liable to the girls, the department urges that although it owed a duty to Talle, it owed no duty to inform the mother and the girls of Heinen’s history, and further that its conduct was not the proximate cause of the girls’ injuries.
(a) Duty
We note at the outset of our consideration of the duty question the conflict in the evidence arising from the mother’s testimony concerning what Strawder told her when Heinen’s placement with Talle was being discussed and Strawder’s failure to recall any such conversations. However, any question of material fact such conflict may present becomes unimportant, for although the issue is within the reach of the department’s assignment that the district court erred in “holding that there was no genuine issue of material fact on the issue of liability,” this conflict in the evidence is not argued by the department.
Because a claimed prejudicial error must not only be assigned, but must be discussed in the brief of the asserting party, Brewer v. Brewer, 244 Neb. 731, 509 N.W.2d 10 (1993), an appellate court will not consider assignments of error which are not discussed in the brief, Florist Supply of Omaha v. Prochaska, 244 Neb. 776, 509 N.W.2d 209 (1993). Accordingly, we treat the mother’s testimony concerning her conversations with Strawder as being uncontroverted.
That having been established, we recall that in order to succeed in an action based on negligence, a plaintiff must establish the defendant’s duty not to injure the plaintiff, a breach of that duty, proximate causation, and damages. S.I. v. Cutler, 246 Neb. 739, 523 N.W.2d 242 (1994); Merrick v. Thomas, 246 Neb. 658, 522 N.W.2d 402 (1994); Moore v. [658]*658Slate, 245 Neb. 735, 515 N.W.2d 423 (1994). The question of whether a legal duty exists for actionable negligence is a question of law dependent on the facts in a particular situation. S.I. v. Cutler, supra; Merrick v. Thomas, supra; Lindsay Mfg. Co. v. Universal Surety Co., 246 Neb. 495, 519 N.W.2d 530 (1994).
“ ‘[D]uty’ is a question of whether the defendant is under any obligation for the benefit of the particular plaintiff; and in negligence cases, the duty is. always the same — to conform to the legal standard of reasonable conduct in the light of the apparent risk. . . .
“A duty, in negligence cases, may be defined as an obligation, to which the law will give recognition and effect, to conform to a particular standard of conduct toward another.”
Schmidt v. Omaha Pub. Power Dist., 245 Neb. 776, 786, 515 N.W.2d 756, 763 (1994) (quoting W. Page Keeton et al., Prosser and Keeton on the Law of Torts § 53 (5th ed. 1984)).
Foreseeability is also a factor in establishing a defendant’s duty. Merrick v. Thomas, supra; Schmidt v. Omaha Pub. Power Dist., supra. As Justice Cardozo stated in Palsgraf v. Long Island R. R. Co., 248 N.Y. 339, 344, 162 N.E. 99, 100 (1928), “The risk reasonably to be perceived defines the duty to be obeyed . . . .”
When the mother asked Strawder if her daughters could be left alone with Heinen without risk, the department had ample information that Heinen had a history of violent and abusive behavior. Thus, under the facts of this case, the risk of harm to the girls was foreseeable by the department, and it had a duty, upon direct questioning by the mother, to answer truthfully as to any knowledge it had or later acquired as to Heinen’s violent propensities and any danger that the girls might encounter by being left alone with him. Given the department’s knowledge and the precise nature of the mother’s question, Strawder and the department had a duty to respond truthfully to safety-related inquiries regarding Heinen.
(b) Proximate Cause
The department also asserts that the district court erred in finding that the department proximately caused the girls’ [659]*659injuries. The proximate cause of an injury is that which, in a natural and continuous sequence, without any efficient intervening cause, produces the injury, and without which the injury would not have occurred. Moore v. State, supra. A plaintiff must establish three basic requirements in establishing proximate cause, namely, that (1) without the negligent action, the injury would not have occurred, commonly known as the “but for” rule; (2) the injury was a natural and probable result of the negligence; and (3) there was no efficient intervening cause. Merrick v. Thomas, supra.
The first element of proximate causation is clearly established. The mother stated that had she known of Heinen’s background and the risks which were presented by allowing the girls to be around him without adult supervision, she would never have allowed them to be left alone with him.
The second element is also clearly established. By the department’s stating that there was no risk in leaving the girls alone with Heinen, it was a natural and probable result that the girls would be left alone with Heinen and would thereby be subjected to sexual abuse. A number of those involved in this case agreed that an adolescent victim of physical and sexual abuse has a risk of becoming a perpetrator, and Perry, Heinen’s treating psychotherapist, noted that Heinen should not be left alone with children because of the danger he posed. The foreseeability that Heinen would act in a sexually abusive manner was the hazard that made the department’s conduct negligent when it failed to respond truthfully to the mother’s questions.
Finally, the third element is established as well. An efficient intervening cause is a new, independent force intervening between the defendant’s negligent act and the plaintiffs injury by the negligence of a third person who had full control of the situation, whose negligence the defendant could not anticipate or contemplate, and whose negligence resulted directly in the plaintiff’s injury. Merrick v. Thomas, supra.
The department offers two different claims of intervening causes that resulted in the girls’ injuries. First, the department argues that Heinen’s intervening criminal acts were an intervening cause. However, the evidence is that the department [660]*660had knowledge of Heinen’s violent propensities and that he was at risk of becoming a sexual abuser. It therefore cannot be said that the department could not anticipate or contemplate Heinen’s attacks on the girls. The department had a duty to disclose to the mother, upon her direct questioning, the elements of Heinen’s history that would pose a hazard to her children. Once it is shown that a defendant had a duty to anticipate an intervening criminal act and guard against it, the criminal act cannot supersede the defendant’s liability. W. Page Keeton et al., Prosser and Keeton on the Law of Torts § 44 (5th ed. 1984). E.g., Erichsen v. No-Frills Supermarkets, 246 Neb. 238, 518 N.W.2d 116 (1994) (store owner liable to patron for foreseeable criminal act on premises); S.I. v. Cutler, 246 Neb. 739, 523 N.W.2d 242 (1994) (landlord liable to tenant for foreseeable assault by third party).
Second, the department argues that it was the mother’s negligence in placing the younger girl in the same room as Heinen on the overnight trip to Denver that caused the girls’ injuries. Not only does this argument not address the attacks on the older girl, it does not address the subsequent attacks on the younger one. Such an act might be negligent if the mother had been apprised of the risk that Heinen posed to her daughter; the department’s problem, however, is that the mother was not so apprised and therefore could not take precautions to guard against Heinen’s aggression. Again, based on Heinen’s prior history, the department could and should have anticipated Heinen’s criminal behavior, but the mother could not because the department did not inform her of Heinen’s violent propensities. It therefore was not an efficient intervening cause for the mother to place her daughter in the same room as Heinen on the overnight trip.
(c) Resolution
As a consequence, the district court correctly ruled that there was no genuine issue of material fact concerning the department’s liability.
2. Hedonic Damages
The department also contends that the district court erred in awarding hedonic damages, which we define in this context as [661]*661the damages arising from loss of the enjoyment of life, as measured separately from the economic productive value that an injured person would have had. See Black’s Law Dictionary 391 (6th ed. 1990).
The issue of hedonic damages is not new in Nebraska. In Swiler v. Baker’s Super Market, Inc., 203 Neb. 183, 187-88, 277 N.W.2d 697, 700 (1979), we wrote:
Loss of enjoyment of life may, in a particular case, flow from a disability and be simply a part thereof, and where the evidence supports it, may be argued to the jury. A separate instruction therein may be redundant. We do not recommend such an instruction be given, but find that under the facts of this particular case, where there is evidence from which the jury could find the injuries and resulting disability did cause loss of enjoyment of life, there was no error in giving the instruction, and we do not believe the jury was in any way misled.
Some courts and commentators have construed this passage as indicating that we have recognized hedonic damages as a separate and distinct category of damages, apart from pain and suffering. See, e.g., Mariner v. Marsden, 610 P.2d 6 (Wyo. 1980); Pamela J. Hermes, Loss of Enjoyment of Life— Duplication of Damages Versus Full Compensation, 63 N.D. L. Rev. 561 (1987). However, we did not go so far in Swiler as to hold that hedonic damages were to be recognized as a separate and distinct category of damages; we merely held that on the facts then before us, it was not error for the district court to have given the instruction on hedonic damages that it gave.
In this regard, Swiler is similar to the cases Huff v. Tracy, 57 Cal. App. 3d 939, 129 Cal. Rptr. 551 (1976), and Leiker v. Gafford, 245 Kan. 325, 778 P.2d 823 (1989), overruled on other grounds, Martindale v. Tenny, 250 Kan. 621, 829 P.2d 561 (1992). In both of those cases, the respective courts held that loss of the enjoyment of life was not a separate and distinct category of damages apart from pain and suffering, but that the instructions given which treated it as such did not amount to reversible error.
It is true that in Swiler, we declared that a majority of courts have approved hedonic damages as a separate category of [662]*662nonpecuniary damages that may be instructed upon. However, this statement appears to have been improvident.
The cases which have considered the issue generally fall into three categories: (1) those which totally reject loss of enjoyment of life as a consideration in awarding damages; (2) those which hold it is not a separate category of damages but may be considered as an element or component of pain and suffering and/or disability; and (3) those which hold loss of enjoyment of life is a separate category of damages.
Leiker v. Gafford, 245 Kan. at 339, 778 P.2d at 834. A slight majority of courts that have considered the issue have adopted the second category. Leiker, supra. See, Dugas v. Kansas City Southern Railway Lines, 473 F.2d 821 (5th Cir. 1973) (construing FELA); Sullivan v. U.S. Gypsum Co., 862 F. Supp. 317 (D. Kan. 1994) (wrongful death claim); Ortega v. Plexco, A Div. of Chevron Chemical Corp., 793 F. Supp. 298 (D.N.M. 1991) (wrongful death claim); Sterner v. Wesley College, Inc., 747 F. Supp. 263 (D. Del. 1990) (wrongful death claim); Nichols v. Estabrook, 741 F. Supp. 325 (D.N.H. 1989) (wrongful death claim); Funston v. United States, 513 F. Supp. 1000 (M.D. Pa. 1981) (construing Federal Tort Claims Act); Wilt v. Buracker, 191 W. Va. 39, 443 S.E.2d 196 (1993), cert. denied_U.S. _, 114 S. Ct. 2137, 128 L. Ed. 2d 867 (1994); Canfield v. Sandock, 563 N.E.2d 1279 (Ind. 1990); McDougald v. Garber, 73 N.Y.2d 246, 536 N.E.2d 372, 538 N.Y.S.2d 937 (1989); Poyzer v. McGrow, 360 N.W.2d 748 (Iowa 1985); Judd v. Rowley’s Cherry Hill Orchards, Inc., 611 P.2d 1216 (Utah 1980); Willinger v. Mercy Catholic Med. Ctr., Etc., 482 Pa. 441, 393 A.2d 1188 (1978) (wrongful death claim); Degen v. Bayman, 90 S.D. 400, 241 N.W.2d 703 (1976); Winter vs. Pa. R. R. Co., 45 Del. 108, 68 A.2d 513 (1949); Lombardo v. Hoag, 269 N.J. Super. 36, 634 A.2d 550 (1993), cert. denied 135 N.J. 469, 640 A.2d 850 (1994); Foster v. Trafalgar House Oil & Gas, 603 So. 2d 284 (La. App. 1992) ; Stroud v. Stroud, 299 S.C. 394, 385 S.E.2d 205 (S.C. App. 1989); Leonard v. Parrish, 420 N.W.2d 629 (Minn. App. 1988); Wingfield v. Peoples Drug Store, Inc., 379 A.2d 685 (D.C. App. 1977); Huff v. Tracy, supra; Underwood v. Atlanta & West Point R. Co., 105 Ga. App. 340, 124 S.E.2d 758 [663]*663(1962), reversed in part on other grounds 218 Ga. 193, 126 S.E.2d 785.
We conclude that the better reasoned rule is the rule we adopted in Swiler v. Baker’s Super Market, Inc., 203 Neb. 183, 277 N.W.2d 697 (1979), i.e., that loss of the enjoyment of life is not a separate category of damages but is an element or component of pain and suffering and of disability. We therefore adhere to that rule.
As noted in McDougald v. Garber, 73 N.Y.2d at 257, 536 N.E.2d at 376-77, 538 N.Y.S.2d at 941:
If we are to . . . approve a separate award for loss of enjoyment of life, it must be on the basis that such an approach will yield a more accurate evaluation of the compensation due to the plaintiff. We have no doubt that, in general, the total award for nonpecuniary damages would increase if we adopted the rule. That separate awards are advocated by plaintiffs and resisted. by defendants is sufficient evidence that larger awards are at stake here. But a larger award does not by itself indicate that the goal of compensation has been better served.
The advocates of separate awards contend that because pain and suffering and loss of enjoyment of life can be distinguished, they must be treated separately if the plaintiff is to be compensated fully for each distinct injury suffered. We disagree. Such an analytical approach may have its place when the subject is pecuniary damages, wliich can be calculated with some precision. But the estimation of nonpecuniary damages is not amenable to such analytical precision and may, in fact, suffer from its application. Translating human suffering into dollars and cents involves no mathematical formula; it rests, as we have said, on a legal fiction. The figure that emerges is unavoidably distorted by the translation. Application of this murky process to the component parts of nonpecuniary injuries (however analytically distinguishable they may be) cannot make it more accurate. If anything, the distortion will be amplified by repetition.
... We are confident . . . that the trial advocate’s art is a sufficient guarantee that none of the plaintiff’s losses will be ignored by the jury.
[664]*664Thus, while consideration of loss of the enjoyment of life may properly be considered as it relates to pain and suffering, and to disability, it is improper to treat it as a separate category of nonpecuniary damages. For that reason, the district court’s damages award must be reversed.
3. Computation of Damages
Lastly, the department argues that the district court erred in admitting the testimony of Stan Smith, whom the girls’ mother proffered as an alleged expert on establishing a formula for calculating the value of lost enjoyment of life.
Smith is an economist holding a bachelor of science degree who has worked toward a doctorate in economics and finance, served as a staff economist with various entities, worked for a financial and economics consulting firm, and now operates his own such firm.
Smith has calculated what he considers to be the average value per statistical life saved. Smith’s testimony was based on a “willingness to pay” approach, an approach that encompasses an individual’s willingness to pay for safety (e.g., airbags), the willingness to accept payment to endure risk of death in a job (e.g., coal miner), and the costs of government regulations. Utilizing the “willingness to pay” approach, Smith has calculated the average value per statistical life saved to be $2.3 million, in 1988 dollars. He then adjusts this figure based on the life expectancy of the individual in question and the fact that, except in a death case, not all of the enjoyment of life has been lost, but only some portion of it. Based on his calculations, Smith estimated that the older girl has suffered a loss of between $1,950,000 and $2,127,000, averaging $2,038,724, and that the younger girl has suffered a loss of between $2,442,000 and $2,817,000, averaging $2,629,996.
Although Smith claimed that the “willingness to pay” approach “is not only widely accepted but exclusively accepted as the proper basis for valuing human life” and that it is “exclusively employed in academia,” this approach has been subject to severe criticism from both courts and legal commentators. See, Mercado v. Ahmed, 91A F.2d 863 (7th Cir. 1992); Ayers v. Robinson, 887 F. Supp. 1049 (N.D. Ill. 1995); [665]*665Hein v. Merck & Co., Inc., 868 F. Supp. 230 (M.D. Term. 1994); Sullivan v. U.S. Gypsum Co., 862 F. Supp. 317 (D. Kan. 1994); Livingston v. U.S., 817 F. Supp. 601 (E.D.N.C. 1993); Sterner v. Wesley College, Inc., 747 F. Supp. 263 (D. Del. 1990); Montalvo v. Lapez, 77 Haw. 282, 884 P.2d 345 (1994); Wilt v. Buracker, 191 W. Va. 39, 443 S.E.2d 196 (1993), cert. denied __ U.S. _, 114 S. Ct. 2137, 128 L. Ed. 2d 867 (1994); Foster v. Trafalgar House Oil & Gas, 603 So. 2d 284 (La. App. 1992); Fetzer v. Wood, 211 Ill. App. 3d 70, 569 N.E.2d 1237 (1991); Southlake Limousine v. Brock, 578 N.E.2d 677 (Ind. App. 1991); Andrew J. McClurg, It’s a Wonderful Life: The Case for Hedonic Damages in Wrongful Death Cases, 66 Notre Dame L. Rev. 57 (1990). The majority of jurisdictions considering testimony based on “willingness to pay” studies have concluded that such testimony is inadmissible. Montalvo v. Lapez, supra; Wilt v. Buracker, supra. The criticism focuses on the fact that the basis of the “willingness to pay” approach is faulty and that the testimony does not satisfy the rules of evidence.
As a practical point, it is impossible to place a dollar figure on how each individual values his or her own life. Government regulators may attempt to do so for purposes of adopting regulations, but any amount they arrive at is a fiction, designed to escape the obvious answer anyone would give when asked what one’s own life was worth.
Consider the paradigmatic plight of Jessica McClure, the little girl who became trapped in the Texas water well in 1987. Extraordinary rescue efforts were undertaken to rescue her, at great cost. It is doubtful, however, that anyone was overheard suggesting that Jessica be left to die because the attempts to rescue her were too costly. In that situation, society deemed Jessica’s life to be priceless.
McClurg, 66 Notre Dame L. Rev. at 58.
The “willingness to pay” approach is based on the belief that the value of human life can be determined by “analyzing either production or consumption behavior with respect to risk reduction or avoidance.” Id. at 102. Three different models are used to arrive at the value of life: (1) consumer purchases of safety devices, (2) wage risk premiums to workers, and (3) [666]*666cost-benefit analyses of governmental regulations. There are critical errors in all three models used to arrive at hedonic valuation.
The first model is based on consumption activities of individuals. “From the price differential between a safer product and one less safe, and the reduction of risk of dying supposedly consequent therefrom, an extrapolation is made to arrive at the hedonic value of life.” Hein v. Merck & Co., Inc., 868 F. Supp. at 234. However, such reasoning is flawed because it fails to take into consideration the numerous factors that go into a consumer purchase and whether a consumer actually and accurately perceives the risk in making a purchase of one product rather than another.
For example, spending on items like air bags and smoke detectors is probably influenced as much by advertising and marketing decisions made by profit-seeking manufacturers and by government-mandated safety requirements as it is by any consideration by consumers of how much life is worth. Also, many people may be interested in a whole range of safety devices and believe they are worthwhile, but are unable to afford them. More fundamentally, spending on safety items reflects a consumer’s willingness to pay to reduce risk, perhaps more a measure of how cautious a person is than how much he or she values life. New of us, when confronted with the threat, “Your money or your life!” would, like Jack Benny, pause and respond, “I’m thinking, I’m thinking.” Most of us would empty our wallets. Why that decision reflects less the value we place on life than whether we buy an airbag is not immediately obvious.
Mercado v. Ahmed, 974 F.2d 863, 871 (7th Cir. 1992). When viewed in this manner, as one commentator pointed out, the willingness to reduce risk, and not the value of one’s life, is what drives consumer purchases, and this is the basic flaw in the first model:
For example, how much would a person demand to play Russian Roulette with a six-chamber revolver loaded with one bullet? The risk is easily computed as a one-in-six chance of death. To know the amount people would [667]*667demand to confront this risk would allow us to calculate with accuracy the value people attach to their lives. We can predict with confidence that the amounts would be tremendous, and that they probably would not vary widely between those who [buy cars with airbags] and those who do not.
Andrew J. McClurg, It’s a Wonderful Life: The Case for Hedonic Damages in Wrongful Death Cases, 66 Notre Dame L. Rev. 57, 105-06 (1990).
The second model is based on the belief that workers will demand higher wages in return for higher risks of death. “An extrapolation is then made from the differential between salaries for low-risk jobs and high-risk jobs to come up with the hedonic value of life.” Hein v. Merck & Co., Inc., 868 F. Supp. 230, 233-34 (M.D. Tenn. 1994). This model suffers from the fact that it assumes workers have free choice in the jobs they take, that they voluntarily assume the risk the job entails, and that they accurately perceive the risk. These assumptions cannot stand up to close scrutiny.
Diverse barriers — geographic, educational, language and racial, to name but a few — preclude free choice in employment decisions for many groups of people. For example, we cannot say that residents of Appalachian coal mining regions freely choose to be exposed to the risks of coal mining in return for a wage risk-premium. For many, coal mining is the only choice because of their lack of mobility.
McClurg, 66 Notre Dame L. Rev. at 104. Workers may also accept high-risk jobs for a variety of other reasons, including civic pride, the fact that one’s mother or father engaged in the work, or that it was the only job available in a tight job market. The fact is that workers accept high-risk jobs based on a number of factors other than any wage-risk premium, calling into question any formula which uses such employment choices as a basis to value a human life. Cf. Wilt v. Buracker, 191 W. Va. 39, 48, 443 S.E.2d 196, 205 (1993), cert. denied_ U.S. _, 114 S. Ct. 2137, 128 L. Ed. 2d 867 (1994) (“[a]nyone who is familiar with the wages of coal miners, policemen, and firefighters would scoff at the assertion that [668]*668these high risk jobs have any meaningful extra wage component for the risks undertaken by workers in those professions”).
The third model is based on the cost-benefit analysis conducted by government agencies in deciding whether to adopt a regulation. Smith stated that government studies on the cost-benefit analysis of government “show [a] willingness to implement legislation at a cost of approximately two million dollars per life saved; very little legislation beyond three million.” This model is also subject to the criticism that it does not take into account numerous other factors that go into such decisions. One court put it this way:
[G]ovemment calculations about how much to spend (or force others to spend) on health and safety regulations are motivated by a host of considerations other than the value of life: is it an election year? how large is the budget deficit? on which constituents will the burden of the regulations fall? what influence and pressure have lobbyists brought to bear? what is the view of interested constituents? And so on.
Mercado v. Ahmed, 974 F.2d 863, 871 (7th Cir. 1992). Another court pointed out that the amounts reached by government agencies in valuing a human life are often a result of political compromise:
“The Occupational Safety and Health Administration’s (OSHA’s) use of risk reduction values in regulatory analysis led the agency into a major conflict with the Office of Management and Budget (OMB) in 1985. OSHA initially attempted to use a value of $3,000,000 per anonymous life saved. The OMB decided that this value was too high and suggested that OSHA use a value of $1,000,000. Eventually, OMB and OSHA reached a compromise of $2,000,000 per life saved.” . . . “Although they do not always reveal their methodologies, federal agencies have set life values as low as $70,000 and as high as $132 million per life.”
(Citations omitted.) Ayers v. Robinson, 887 F. Supp. 1049, 1061 n.4 (N.D. Ill. 1995).
All three models that are used by Smith in valuing a life are thus flawed in one way or another.
[669]*669In addition to the problems with these three models, the testimony does not satisfy the standards for admissibility of expert testimony under the Nebraska Evidence Rules. In determining whether an expert’s testimony is admissible pursuant thereto, a court, in accordance with Frye v. United States, 293 F. 1013 (D.C. Cir. 1923), considers four preliminary and interrelated questions: (1) whether the witness qualifies as an expert pursuant to Neb. Evid. R. 702, Neb. Rev. Stat. § 27-702 (Reissue 1989); (2) whether the expert’s testimony is relevant; (3) whether the expert’s testimony assists the trier of fact to understand the evidence or determine a controverted factual issue; and (4) whether the expert’s testimony, even though relevant and admissible, should be excluded under Neb. Evid. R. 403, Neb. Rev. Stat. § 27-403 (Reissue 1989), because its probative value is substantially outweighed by the danger of unfair prejudice or other considerations. See State v. Reynolds, 235 Neb. 662, 457 N.W.2d 405 (1990). See, also, State v. Carter, 246 Neb. 953, 524 N.W.2d 763 (1994) (adheres to Frye test). Accord State v. Dean, 246 Neb. 869, 523 N.W.2d 681 (1994).
Rule 702 provides: “If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.”
One of the criticisms of expert testimony on the value of life is that there is “no basic agreement among economists as to what elements ought to go into the life valuation. There is no unanimity on which studies ought to be considered. There is a lack of reliability.” Mercado v. Ahmed, 756 F. Supp. 1097, 1103 (N.D. Ill. 1991), aff’d 974 F.2d 863 (7th Cir. 1992). The testimony is therefore not admissible as “scientific,” as the term is used in rule 702. See State v. Reynolds, 235 Neb. at 681-82, 457 N.W.2d at 418:
Under the test or standard enunciated in Frye, reliability for admissibility of an expert’s testimony, including an opinion, which is based on a scientific principle or is based on a technique or process which utilizes or applies [670]*670a scientific principle, depends on general acceptance of the principle, technique, or process in the relevant scientific community.
(Emphasis supplied.)
Even assuming that there was agreement among economists or that Smith’s process of valuation was scientific, there would still be a problem with such testimony because
the consensus [would be] that of persons who are no more expert than are the jurors on the value of the lost pleasure of life. Even if reliable and valid, the evidence may fail to “assist the trier of fact to understand the evidence or determine a fact in issue” in a way more meaningful than would occur if the [trier of fact] asked a group of wise courtroom bystanders for their opinions.
Mercado v. Ahmed, 756 F. Supp. at 1103. “Given the flaws of the willingness-to-pay theory and the wide-ranging values it generates, the extent to which willingness-to-pay evidence meaningfully assists a jury in calculating damages for the intrinsic value of life is suspect at best.” Andrew J. McClurg, It’s a Wonderful Life: The Case for Hedonic Damages in Wrongful Death Cases, 66 Notre Dame L. Rev. 57, 106 (1990).
The measurement of the joy of life is intangible. A jury may draw upon its own life experiences in attempting to put a monetary figure on the pleasure of living. It is “a uniquely human endeavor . . . requiring the trier of fact to draw upon the virtually unlimited factors unique to us as human beings. ”... Testimony of an economist would not aid the jury in making such measurements because an economist is no more expert at valuing the pleasure of life than the average juror. “[T]he loss of enjoyment of life resulting from a permanent injury is . . . not subject to an economic calculation. ”
(Citation omitted.) Montalvo v. Lapez, 77 Haw. 282, 303, 884 P.2d 345, 366 (1994).
Smith’s testimony therefore fails the first and third prongs of the factors listed in Reynolds. That being so, we need not decide whether in any event such testimony should be excluded under rule 403 because its probative value is substantially outweighed by the danger of unfair prejudice or other considerations, as [671]*671some courts have concluded. See, e.g., Ayers v. Robinson, supra; Foster v. Trafalgar House Oil & Gas, 603 So. 2d 284 (La. App. 1992).
Thus, the district court erred in admitting Smith’s testimony. Ordinarily, the erroneous admission of evidence in a bench trial of a law action is not reversible error if other relevant evidence, admitted without objection or properly admitted over objection, sustains the trial court’s necessary factual findings. Miles v. Box Butte County, 241 Neb. 588, 489 N.W.2d 829 (1992); Maresh v. State, 241 Neb. 496, 489 N.W.2d 298 (1992); State v. Lomack, 239 Neb. 368, 476 N.W.2d 237 (1991); Suess v. Lee Sapp Leasing, 229 Neb. 755, 428 N.W.2d 899 (1988); Alliance Nat. Bank v. State Surety Co., 223 Neb. 403, 390 N.W.2d 487 (1986). However, in such a case, a reversal is warranted if the record shows that the trial court actually made a factual determination, or otherwise resolved a factual issue or question, through the use of erroneously admitted evidence. Miles v. Box Butte County, supra; State v. Lomack, supra.
In awarding its damages, the district court here specifically recited that it found Smith to be an expert whose specialized knowledge was helpful. Moreover, in finding that the older girl sustained $26,860.28 in damages, it included $25,000 for loss of the enjoyment of life, and in finding that the younger girl sustained $348,341.83 in damages, included $300,000 for loss of the enjoyment of life. The record thus shows that although the district court did not fully accept Smith’s testimony, it did rely on the erroneously admitted evidence. That circumstance provides an additional reason for rejecting the district court’s damages award.
V. JUDGMENT
Accordingly, as noted in part I, the summary judgment finding the department liable to the girls is affirmed; however, the judgment awarding them damages must be, and hereby is, reversed and the cause remanded for a new trial on the damages issue alone.
Affirmed in part, and in part reversed
AND REMANDED FOR A NEW TRIAL
ON THE. ISSUE OF DAMAGES.