SMITH, Senior Judge.

Appellants, Larry E. Gray and Frederick White, appeal from a final order of the Commission on Human Relations adopting the hearing officer's recommended order and ruling that appellee Russell Corporation (Russell) did not discriminate against appellants on account of their race when it discharged them from employment. Finding no error, we affirm.

Appellants have framed several issues on appeal so as to present questions of law. However, we find that disposition of the issues on appeal turns primarily upon our review of factual issues determined by the hearing officer below. After careful review, we conclude that the hearing officer's finding that there was no discriminatory intent in the discharge of appellants is supported by competent, substantial evidence, and was not based on erroneous applications of the law.

Appellants, Gray and White, both black males, were employed by Russell from March 1988 and August 1987, respectively. At the time of their discharge, both held positions as lift truck operators, the highest paying hourly rate job in the plant except for the lead person position. Neither had prior disciplinary problems with Russell. They were both discharged from their employment in November 1991 after an incident in which the two disrupted a company-approved luncheon on company premises being hosted by personnel of the company's Embroidery Department. Appellants did not work in the Embroidery Department and were not invited to participate in the luncheon. The evidence established, as found by the hearing officer, that appellants entered the room where the luncheon was being set up, and that they took and consumed food prepared for others after being denied permission to do so by one of the employees who was helping to set up the food for the luncheon. Complaints were brought to the attention of managerial personnel, and a report was submitted to Russell's review committee which is composed of five supervisors and department heads, one of whom is black. The review committee concluded that the incident amounted to appellants' having "willfully destroyed the property of others," in violation of company rule, and therefore discharged appellants on that basis. Appellants filed charges of discrimination with the Florida Commission on Human Relations, alleging that they had been denied promotions and had been discharged on account of their race in violation of Florida Statutes, and specifically section 760.10, et seq., Florida's Human Rights Act of 1977.

This may be characterized as a "disparate treatment" case and thus subject to the burden of proof as explained by the Supreme Court in McDonnell Douglas v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973); Texas Department of Community Affairs v. Burdine, 450 U.S. 248, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981), and subsequent cases. We have previously articulated at length the burden of proof in discrimination cases under the Florida Act, and no useful purpose would be served by reiteration of those requirements. See Florida Department of Community Affairs v. Bryant, 586 So.2d 1205 (Fla. 1st DCA 1991). Because the Act is patterned after Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e, et seq., federal case law interpreting Title VII is applicable. Id. at 1209.

We initially observe that there was no direct evidence of a discriminatory motive on the part of Russell in discharging appellants. The hearing officer found, however, that appellants had established a prima facie case of discrimination by evidence regarding similarly-situated white employees who purportedly violated the same or similar rules having to do with damage to or destruction of property of the company or others, but who were not discharged. Although the hearing officer questioned the similarity of the "comparison" evidence, he found, nevertheless, that some of this evidence was sufficiently similar so that an ordinary person might "reasonably infer" discrimination from the facts shown, if those facts were unrebutted. See Goldstein v. Manhattan Industries, Inc., 758 F.2d 1435, 1443 (11th Cir.), cert. den'd, 474 U.S. 1005, 106 S.Ct. 525, 88 L.Ed.2d 457 (1985). The hearing officer then concluded that evidence presented by Russell concerning the facts considered by the review committee, and the reasons for its action in the instant case, provided a legitimate, nondiscriminatory reason for its decision to discharge appellants. Accordingly, the hearing officer found the burden was upon appellants to prove, by a preponderance of the evidence, that Russell's articulated reasons for discharging appellants were pretextual, and that they had failed to meet that burden.

The evidence concerning comparable incidents of discipline or discharge of both white and black employees was of critical importance in the case, and we find that the hearing officer in his order has carefully summarized and evaluated the evidence pertaining to each incident. Numerous federal decisions have dealt with the requirements of "comparability" of evidence in the context of discrimination cases, as outlined by the federal court in Williams v. Public Warehouse, 9 Fla. L. Weekly F. D14, 1995 WL 224423 (M.D. Fla. April 6, 1995):

Thus, to be deemed "similarly-situated", the individuals with whom the plaintiff seeks to compare his/her treatment must have dealt with the same supervisor, have been subject to the same standards and have engaged in the same conduct without such differentiating or mitigating circumstances that would distinguish their conduct or the employer's treatment for them for it. Mazzella v. RCA Global Communications, Inc., 642 F. Supp. 1531 (S.D.N.Y. 1986), aff'd., 814 F.2d 653 (2d Cir. 1987); Lanear v. Safeway Grocery, 843 F.2d 298 (8th Cir. 1988) (plaintiff must prove that he and white employee were similarly situated in all respects and that the other employee's acts were of comparable seriousness to his own); Cox v. Electronic Data Systems Corp., 751 F. Supp. 680 (E.D.Mich. 1990)

[Quoting Mitchell v. Toledo Hosp., 964 F.2d 577 (6th Cir. 1992).] Thus, a Plaintiff must show "that a person of another race would not also have been discharged under similar circumstances." Rush v. McDonald's Corp., 966 F.2d 1104, 1112 (7th Cir. 1992) (quoting McDonald v. Santa Fe Trail Transportation Co., 427 U.S. 273, 281-84 [96 S.Ct. 2574, 2679-81, 49 L.Ed.2d 493] (1976)).

Id. at D16.

In the case before us, the hearing officer found that facts available to the review committee showed that appellants had taken food belonging to employees of the Embroidery Department, and further, that they had "taunted those employees, exhibiting a disregard for their feelings and a lack of remorse or apology for their conduct." This, according to the hearing officer, "was an aggravating factor considered by the review committee in terms of its concern for the morale of their fellow employees, and the maintenance of peace and good relationships in the work place. . . ."

Appellant White argues that a distinction should be made between his conduct and that of appellant Gray, in that White exhibited none of the "taunting, mocking attitude" toward the other employees displayed by Gray. White's actions were apparently less noticeably egregious than the actions of Gray. Nevertheless, he knowingly participated in the incident with Gray, consuming food himself, and offering a piece of chicken to a fellow worker who was also not a member of the Embroidery Department. Both White and Gray failed to show any remorse or attempt to apologize for their actions, and White nodded in assent when Gray, upon being questioned, denied having eaten the food. White later, on two occasions, admitted having eaten the food without permission.

In his order the hearing officer, referring to the firings as a "seemingly harsh remedy," noted that it was specifically the absence of remorse or apology on the part of appellants and their misrepresentation of their actions that caused the review committee, including the black member, to unanimously vote to discharge appellants. The hearing officer found that the committee "was sensitive to a perceived need to take strong disciplinary action in order to preserve employee morale of those victimized by the incident and other employees who knew of it, and to show the committee's sensitivity to the necessity that co-workers respect each other's property and feelings."

It is apparent from our review that none of the comparative evidence involved incidents of willful destruction of the property of other employees, nor, more significantly, did any of these instances deal with disrespectful or intimidating conduct such as that engaged in by appellants. We reject appellants' contention that by requiring them to provide evidence of instances in which white employees were not discharged under similar circumstances, the hearing officer required them to "prove too much." The determination of comparable seriousness is a factual issue within the province of the hearing officer, and it is not our function to second-guess his conclusions as to the comparability and seriousness of the conduct involved in each instance. Heifetz v. Department of Business Regulation, Division of Alcoholic Beverages Tobacco, 475 So.2d 1277 (Fla. 1st DCA 1985) (factual issues susceptible to ordinary methods of proof and not infused with policy considerations are the prerogative of hearing officer, and ultimate findings of fact supported by competent, substantial evidence may not be changed by reviewing agency or appeals court, even if reviewing tribunal would have ruled otherwise had it been trier of fact). Further, the question of intentional discrimination is a "pure question of fact" and, as such, the factfinder's determination will not be set aside unless clearly erroneous. Pullman-Standard v. Swint, 456 U.S. 273, 286, n. 16, 102 S.Ct. 1781, 1789, n. 16, 72 L.Ed.2d 66 (1982).

Appellants also urge that the hearing officer erred in excluding testimony by appellants' expert witness. They contend that by rejecting the expert testimony, appellants were deprived of the opportunity to establish a "pattern and practice of discrimination," which imposes a higher burden upon the employer. As stated by the court in Cox v. American Cast Iron Pipe Co., 784 F.2d 1546 (11th Cir.), cert. den'd, 479 U.S. 883, 107 S.Ct. 274, 93 L.Ed.2d 250 (1986):

In other words, once a pattern and practice of discrimination is established, a rebuttable presumption that plaintiff was discriminated against because of her sex and is entitled to recovery obtains. The employer may overcome this presumption only with clear and convincing evidence that job decisions made when the discriminatory policy was in force were not made in pursuit of that policy.

Id. at 1559.

Appellants' expert was accepted by the hearing officer as an expert witness in the field of Labor Economics. However, during his testimony serious questions arose concerning the reliability and accuracy of the data upon which his proffered opinions were based. In his recommended order the hearing officer found that appellants had failed to establish that the statistical evidence and analysis by their expert witness consisted of, or was based upon, "properly-gathered, formulated and analyzed statistical proof." Consequently, at the conclusion of the expert's testimony the hearing officer ruled that his testimony and written report were inadmissible.

Although the hearing officer devoted little space in his recommended order to the expert witness issue, the record itself clearly reveals the basis for his decision to exclude the expert evidence. The expert was produced, according to appellants' counsel, for the purpose of testifying only regarding the likelihood of Russell's black employee discharge decisions occurring absent racial bias, based upon statistical computations involving the number of black and white employee discharges in relation to the racial composition of Russell's work force. The witness was accepted by the hearing officer as an expert in the area of Labor Economics as that relates to making calculations and analyses from statistical data and rendering opinions as to what such calculations and analyses show regarding Russell's hiring, promotion and termination practices.

During the discovery process below, appellants' counsel received copies of a wide range of documents from Russell, including some mandated by federal law, disclosing information as to the size and racial composition of Russell's work force, termination lists and logs, and reports of disciplinary actions for the three-year period preceding appellants' discharges, and for an additional year ending September 1992. From these documents, which were placed in evidence before the hearing officer, appellants' law firm prepared a computer data base which was then reduced to summary form by a paralegal employed by the law firm, who separated, classified and categorized the reprimand and discharge information from the computer data base without instruction or direction from the expert witness. Testimony at the final hearing revealed that the expert's written report and testimony were based entirely upon his statistical analysis of the information contained in the summary. The witness never examined or even saw the underlying data from which the summary was prepared, was unaware of the manner in which the information had been compiled and categorized, and was unaware that certain essential information regarding disciplinary reprimands had been excluded from the summary. The person or persons who compiled the data and prepared the summary were not called as witnesses.

In particular, the summary excluded all disciplinary actions and discharges of probationary or orientation employees. Also excluded were all reprimands for tardiness or absenteeism. The expert believed that he made the decision to exclude probationary employees based upon the manner in which those employees were "characterized" to him by appellants' counsel. From this information he made certain assumptions concerning the manner in which Russell subjected them to disciplinary actions, concluding that they were treated differently than non-probationary employees. This assumed difference, in his opinion, justified eliminating them from his study. He made no investigation of his own to determine whether the assumptions upon which he based the exclusion decision had any basis in fact.

The expert testified that his analysis of the termination statistics depended in part upon certain further assumptions concerning the productivity of black and white workers as the basis for Russell's discharge decisions. He stated that reprimands for tardiness and absenteeism would relate to productivity, and that consideration of these reprimands would be critical to his conclusions concerning disciplinary actions for work-related reasons. His testimony indicated that reprimands related to performance or productivity were included in his study and analysis. However, it was then pointed out to him that his written report indicated these reprimands had been excluded. He stated that he was unaware that this exclusion had been made in the summary furnished to him, but agreed that inclusion of these reprimands would be essential to the making of a valid study. We find that Russell's counsel objected throughout the expert's testimony because of these and other omissions and deficiencies. Appellants' assertion that Russell's only concern giving rise to an objection based on section 90.704, Florida Statutes, was the count of how many white and black employees were discharged during the time frames at issue appears incorrect.

After careful examination of the testimony given by the expert witness, both on direct examination and during his protracted cross-examination, we cannot say that the hearing officer abused his discretion in rejecting it in its entirety. The hearing officer recognized the rule, as codified in the Florida Evidence Code, that an expert may rely upon data or information from other sources, and that such data and information itself need not be admissible in evidence, provided "the facts and data are of a type reasonably relied upon by experts in the subject to support the opinion expressed." § 90.704, Florida Statutes. The hearing officer ruled that the standard for admissibility had not been met by appellants' evidence.

The expert witness admitted that he had relied upon data compiled and categorized by a paralegal of appellants' law firm; that he had given no instructions as to how to compile the data, what separations or categorizations should be made and the reasons for them; and that he had not reviewed the underlying records or documents from which the information supplied to him had been compiled. Upon direct questioning by the hearing officer, the witness stated that he really "couldn't confirm, one way or another," whether experts in his field customarily relied upon data assembled and provided to them under such circumstances as presented in this case.

The hearing officer concluded that the expert's source of data or information upon which he based his opinions was not that commonly relied upon by an expert practicing in the field of Labor Economics. He also pointed out that the obvious omission from the reprimands study of critical data would furnish an additional basis for exclusion of the expert's written report, Petitioners' Exhibit 35. With respect to the testimony regarding terminations, the hearing officer observed that even if he found it admissible, despite its failure to meet the standard for admissibility under section 90.704, it would have "scant probative value."

The appellate courts in this state have long followed the rule that the acceptance or rejection of expert testimony is a matter within the sound discretion of the lower tribunal, and such decision will not be overturned on appeal absent a showing of abuse of discretion. See Behm v. Division of Administration, State Department of Transportation, 292 So.2d 437, 441 (Fla. 4th DCA 1974), approved, 336 So.2d 579 (Fla. 1976) (trier of fact may accept or reject expert opinion evidence in its discretion); Husky Industries, Inc. v. Black, 434 So.2d 988 (Fla. 4th DCA 1983) (opinions and inferences of expert are inadmissible unless underlying facts or data form sufficient basis for expert's opinion and are themselves relevant); Gulley v. Pierce, 625 So.2d 45, 50 (Fla. 1st DCA 1993), rev. denied, 637 So.2d 236 (Fla. 1994) (determination that a witness is qualified as an expert in a particular field does not necessarily mean that the specific expert opinion sought to be elicited should be admitted). We find no abuse of discretion in the hearing officer's exclusion of the expert's testimony and written report.

Appellants place a great deal of emphasis on the issue of statistical evidence in this case. It is undoubtedly true that statistics "play an important and often controversial role" in employment discrimination cases. Capaci v. Katz Besthoff, Inc., 711 F.2d 647, 661 (5th Cir. 1983). "Where gross statistical disparities can be shown, they alone may in a proper case constitute prima facie proof of a pattern or practice of discrimination." Hazelwood School District v. United States, 433 U.S. 299, 307-08, 97 S.Ct. 2736, 2741, 53 L.Ed.2d 768 (1977). citing International Brotherhood of Teamsters v. United States, 431 U.S. 324, 339, 97 S.Ct. 1843, 1856, 52 L.Ed.2d 396 (1977). It is also true, as Judge Reavley wrote for the court in Capaci, supra, that "numerical data is not irrefutable and must be used properly." 711 F.2d at 661. To the same effect, in somewhat different language, the court in Teamsters, supra, had earlier stated:

We caution only that statistics are not irrefutable; they come in infinite variety and, like any other kind of evidence, they may be rebutted. In short, their usefulness depends on all of the surrounding facts and circumstances.

431 U.S. at 340, 97 S.Ct. at 1856-57.

With the foregoing principles in mind, we have considered appellants' contention that the statistical showing made by appellants could not properly be ignored by the hearing officer. Appellants cite Falcon v. General Telephone Co. of Southwest, 626 F.2d 369, 381 (5th Cir. 1980), rev'd on other grounds, 457 U.S. 147, 102 S.Ct. 2364, 72 L.Ed.2d 740 (1982), in which the court stated that

once the plaintiff has offered some evidence that is probative of disparity that may be statutorily significant, it is then the defendant's burden to come up with more specific statistical evidence to rebut the plaintiff's proof. (Emphasis added.)

Appellants also direct our attention to that part of the opinion in Capaci, supra, in which the court admonished:

The defendant must do more than raise theoretical objections to the data or statistical approach taken; instead, the defendant should demonstrate how the errors affect the results . . . particularly in cases where the plaintiff has demonstrated gross disparities in employer practices.

711 F.2d at 653-54. [Citations omitted.]

Appellants urge that because all the records necessary to establish the disparity in discharges of black employees was in evidence before the hearing officer, and because these discharge numbers were either conceded by opposing counsel, or could have been counted by the hearing officer, it was error for the hearing officer to fail to consider this evidence in determining whether Russell was guilty of unlawful discrimination. We have given some consideration to this argument.

During cross-examination of the expert, Russell's counsel indicated to the hearing officer that there had been a total of 94 discharges during the three-year period in question, and that of these, 50 were probationary employees, and 44 were non-probationary employees. At oral argument before this court, Russell's counsel indicated that as to the 44 non-probationary employees, the numbers utilized by appellant appeared to be correct. We may assume, therefore, that of the 44 discharged, 26 were white and 18 were black. According to one benchmark figure computed from data prepared by appellants' counsel and submitted to the expert witness, the percentage of black employees in Russell's work force, excluding officials, managers, professionals and technicians, averaged 27 percent over the three-year period. Based on these figures, it appears that the percentage of black non-probationary employees discharged exceeds the percentage of blacks in Russell's work force. However, although as previously noted probationary employees were not included in the expert's analysis of discharges, both counsel agreed at oral argument that the benchmark, or work force number, utilized in appellants' calculations included both probationary and nonprobationary employees. Appellants offered no analysis of discharges for probationary employees or other indication of what effect inclusion of the discharges might have on the numerical disparity shown to exist, using only the figures for non-probationary employees. Our own examination of the probationary discharge figures found in Respondent's Exhibit 8 (a tabulation of discharge data), reveals that of the 50 probationary employees discharged during the three-year period, 15 were black, and 35 were white. Thus, the percentage of black probationary discharges corresponds fairly closely to the percentage of blacks in Russell's work force.

We make these observations not for the purpose of making our own fact-finding determinations, but simply to point out the obvious: that a change in the manner in which the numerical and other data in the voluminous record before us is assembled, classified and analyzed could change the statistical result. Here, it appears that inclusion of the probationary discharge figures would have tended to favor Russell. Whether the reasons given for exclusion of probationary employees from the numerical study were or were not persuasive was a matter within the province of the hearing officer. In any event, we conclude that without a more complete and accurate explanation and analysis of Russell's discharge and reprimand records, which is absent here, we have no alternative but to agree with the hearing officer's assessment that the proffered statistical evidence has "scant probative value."

We do not as a matter of law foreclose the possibility that a hearing officer, in a given case, might find upon his own examination of documents in the record, unaided by expert testimony, sufficient statistical basis for a conclusion as to the likelihood of discriminatory practices. We do not believe this to be such a case. We observe that not even appellants' expert was asked to provide an opinion as to racial bias based solely on the discharge numbers, without consideration of supporting data concerning disciplinary actions from which further conclusions might be drawn so as to eliminate other possible reasons for a higher discharge rate for blacks. For a court to infer that racial considerations are responsible for a statistically significant disparity, it should be established, among other things, that "race is the only evident variable separating the two groups." See Taylor v. Teletype Corporation, 648 F.2d 1129, 1133 (8th Cir.), cert. den'd, 454 U.S. 969, 102 S. Ct. 515, 70 L.Ed.2d 386 (1981). The evidence before the hearing officer clearly failed to satisfy this requirement.

To reverse the ruling of the hearing officer for his failure to make a finding of disciplinary practices based upon his own statistical analysis and conclusions, as urged by appellants, would require us to fault the hearing officer for his failure, in effect, to step into the shoes of an expert in the field of Labor Economics — a role for which so far as the record shows he possesses no qualifications — and to perform the calculations and analyses appellants' expert failed, or was unable, to do because he did not receive complete information. Significantly, we note that although the hearing officer reserved a final ruling on the admissibility of the expert evidence pending the production of further evidence or legal authority by appellants, the memorandum of law and fact filed below by appellants' counsel in support of appellants' proposed findings of fact and conclusions of law contained no legal argument or citation of authority on the expert evidence issue, or with regard to statistical evidence. Furthermore, appellants' proposed findings of fact and conclusions of law submitted to the hearing officer contained no proposed findings of fact or conclusions of law relating to these issues. Further, although appellants filed exceptions to the recommended order, no argument or objection was made regarding the hearing officer's rejection of the expert testimony or his failure to make findings based on the statistical evidence. See Couch v. Commission on Ethics, 617 So.2d 1119, 1124 (Fla. 5th DCA 1993) (party waives right to challenge recommended order's findings of fact by failing to file exceptions to hearing officer's recommended order; and party cannot argue on appeal matters not properly excepted or challenged before the agency and thus not preserved for appellate review). It would be improper, in our view, to find error under the circumstances presented here.

Appellants raise other issues concerning the hearing officer's findings as to the credibility of certain witnesses presented by appellee. We affirm on these issues without discussion.

Finding no reversible error, the order appealed is AFFIRMED.

BARFIELD, C.J., and KAHN, J., concur.