Raad v. Alaska State Commission for Human Rights

86 P.3d 899, 2004 Alas. LEXIS 5, 85 Empl. Prac. Dec. (CCH) 41,629, 93 Fair Empl. Prac. Cas. (BNA) 178, 2004 WL 57418

• Court: Alaska Supreme Court
• Decided: January 9, 2004
• Docket: S-10718
• Status: Published

Opinion

OPINION

EASTAUGH, Justice.

I. INTRODUCTION

Nada I. Raad claimed that the Fairbanks North Star Borough School District discriminated and unlawfully retaliated against her by not hiring her for any of thirty-one teaching positions. The Alaska State Commission for Human Rights dismissed her discrimination complaint. Raad appeals a superior court decision affirming the commission’s dismissal. Because there was some evidence that the explanations offered by the school district for not hiring Raad were pretextual, and because it is not clear from the record whether the commission adequately considered that evidence, we reverse and remand.

II. FACTS AND PROCEEDINGS

Nada I. Raad is a Lebanese woman who is Muslim. 1 Raad was certified and endorsed by the State of Alaska to teach secondary science and math. She completed her teacher certification at the University of Alaska Fairbanks, and was a student teacher and substitute teacher in the Fairbanks North Star Borough School District in the early 1990s. In January 1991 Raad applied for full-time employment with the district, and in 1992 the district placed her in its hiring pool. The district did not hire Raad for a full-time position for the 1992-93 school year.

Raad was not eligible for a district teaching position for the 1993-94 school year, as a result of disciplinary action imposed by the district. Raad had been a finalist for a district teaching position in 1993 but was not hired. Upset that she was not hired, Raad entered the school district administrative offices in August 1993, accused the district of discriminating against her, and made an alleged threat. The district suspended Raad from the applicant pool for the 1993-94 school year following this incident.

Raad formalized her discrimination accusations against the district in a complaint she filed in 1993 with the Alaska State Commission for Human Rights. The commission investigated the allegations, but dismissed her complaint for lack of substantial evidence. This dismissed complaint forms the basis for the retaliation claims at issue in this appeal.

Raad continued to seek a full-time position with the district after her one-year suspension expired, but the district did not hire her. District hiring decisions are made by individual principals at district schools.

On March 20, 1995 Raad filed another complaint with the commission. She alleged that the district discriminated against her on the basis of her national origin and her religion, and unlawfully retaliated against her for filing the earlier discrimination complaint by not hiring her for positions that became available after her suspension ended. She also alleged that the district discriminated against her on the basis of her sex, excluded her from consideration for substitute teaching positions, and mishandled her son’s assessment for the district’s gifted program for discriminatory reasons. Her claims as amended were directed at district hiring decisions for the 1994-95, 1995-96, and 1996-97 school years.

The commission investigated Raad’s claims and determined that substantial evidence supported her allegations that the district discriminated against her on the basis of her national origin and religion and retaliated against her for filing a discrimination complaint, by not hiring her for teaching positions. The commission found that substantial evidence did not support her other allegations.

The executive director of the commission certified that attempts to eliminate the alleged discrimination by conference, conciliation, or persuasion failed, and that a hearing was required to hear the merits of the case under AS 18.80.120. A hearing on Raad’s complaint was held in November and December 1999 before Hearing Examiner Nathaniel B. Atwood.

Raad’s amended complaint alleged that the district discriminated or unlawfully retaliated against her by not hiring her for any of thirty-one teaching positions for the 1994-95, 1995-96, and 1996-97 school years. It is undisputed that Raad was at least minimally qualified for each of the thirty-one positions. District principals interviewed Raad for six of the positions, but did not hire her for any of them. The hearing examiner found that the district admitted that none of the successful applicants for any of the thirty-one positions was Lebanese or Muslim.

With respect to Raad’s 1993-94 suspension from the hiring pool and prior discrimination complaint, the hearing examiner determined that the behavior resulting in Raad’s suspension was “relevant only to the limited extent that any of the principals making the 31 hiring decisions at issue here, knew of that behavior and were influenced by it in considering whether Raad was a good candidate to fill one of the disputed positions.” The hearing examiner did not allow Raad to testify about the 1993 events or the propriety of the district’s suspension, but did allow testimony about the hiring principals’ knowledge of the 1993 events if that knowledge played a role in their decisions not to hire Raad.

The hearing examiner found that even if it were assumed that Raad had established a prima facie case of discrimination as to one or more of the thirty-one positions, the district had articulated legitimate, non-discriminatory, and non-retaliatory reasons for its hiring decisions for each of the thirty-one positions. The hearing examiner also found that Raad did not show that the reasons offered by the district were pretextual. The hearing examiner concluded that there was “no evidence or even any suggestion from the facts adduced at the hearing that Raad’s national origin, her religion, or her having filed a prior complaint of discrimination had any impact whatsoever on any of the 31 hiring decisions at issue.” (Emphasis added.) The hearing examiner recommended that Raad’s complaint be dismissed in its entirety.

On February 5, 2001 the commission issued a final order that adopted the hearing examiner’s findings of fact, conclusions of law, and proposed order, and dismissed Raad’s complaint. 2 Raad appealed the commission’s order to the superior court. The superior court ruled that the commission’s decision was supported by substantial evidence, and affirmed the order. Raad now appeals the superior court’s decision to this court.

Raad also filed suit against the school district in federal court. 3 The United States District Court of the District of Alaska granted summary judgment against Raad, but that decision was reversed by the United States Court of Appeals for the Ninth Circuit. 4 The federal case does not involve claims that are before this court. Raad’s federal claims concern hiring decisions the district made between 1991 and 1993, and the 1993-94 disciplinary action. 5 In contrast, this appeal involves hiring decisions the district made in 1994, 1995, and 1996. In reversing the grant of summary judgment, the Ninth Circuit held that Raad established her prima facie case of discrimination, 6 and that, viewing the evidence in the light most favorable to Raad, there were debatable issues of material fact regarding whether the school district’s reasons for not hiring Raad were pretextual. 7

III. DISCUSSION

A. Standard of Review

In administrative appeals we “independently review the merits of an administrative decision.” 8 We review an agency’s factual findings to determine whether they are supported by substantial evidence. 9 In applying the substantial evidence test we view the evidence in favor of the findings. 10 Where evidence is conflicting, this court “will not reweigh the evidence and substitute its judgment for that of the trier of fact.” 11 We review questions of law not involving agency expertise under tlie substitution of judgment test. 12 The adequacy of findings of fact and conclusions of law presents an issue of law which we review de novo. 13

B. The Legal Framework

Under Alaska law it is “unlawful for an employer to refuse employment to a person, or to bar a person from employment ... because of the person’s race, religion, color, or national origin.” 14 Employers are also prohibited from retaliating against persons who have filed discrimination complaints with the commission. 15

In cases such as this one, where there is no direct evidence of discriminatory intent, both Aaska and federal courts use a three-part analysis known as the McDonnell Douglas test to determine whether an individual is the subject of unlawful discrimination. 16 The McDonnell Douglas test involves a series of shifting burdens.

The first step of the analysis places the burden on the complaining party to establish a prima facie ease of discrimination. 17 If a prima facie case of discrimination is established, the burden shifts to the employer to articulate a legitimate, nondiscriminatory reason for the employment action. 18 If legitimate, nondiscriminatory reasons are presented, the burden shifts back to the complaining party to show that the reasons offered by the employer are pretextual. 19

The elements of the prima facie showing of discrimination are typically the McDonnell Douglas elements we adopted in Aaska State Comm’n for Human Rights v. Yellow Cab. 20 The complainant must show that (1) the complainant belongs to a protected class; (2) the complainant applied for and was qualified for a job for which the employer was seeking applications; (3) the complainant was rejected despite the complainant’s qualifications; and (4) after the complainant’s rejection, the position remained open and the employer continued seeking applications from persons with the complainant’s qualifications. 21 This test is not applied mechanistically, though, and varies with the factual circumstances of the particular case. 22 For example, if an employer has not left the disputed position open, and has instead hired someone else, the fourth element of the prima facie case is the hiring of an individual not within the same protected class as the complainant. 23

This general framework also applies in retaliation cases. 24 To establish a prima facie case of discriminatory retaliation, a complainant must establish that (1) the complainant engaged in a protected activity (e.g., opposed a discriminatory practice); (2) the complainant suffered an adverse employment action; and (3) there was a causal link between the protected activity and the employer’s action. 25 Causation may be inferred from the proximity in time between the protected action and the allegedly retaliatory action. 26

If a complainant establishes a prima facie case of discrimination or retaliation, the burden shifts to the employer to articulate legitimate, nondiscriminatory reasons for the employment action. 27 We have noted that “[t]o satisfy this burden, the employer ‘need only produce admissible evidence which would allow the trier of fact rationally to conclude that the employment decision had not been motivated by discriminatory animus.’ ” 28 An employer cannot “compose fictitious, post-hoc justifications for an action taken against an employee. Rather, the employer must articulate legitimate business reasons existing at the time the employment decision was made and supported by admissible evidence.” 29

If the employer provides legitimate, nondiscriminatory reasons for the allegedly discriminatory action, the burden shifts back to the complainant to show that discriminatory reasons more likely motivated the employer. 30 Usually a complainant satisfies this burden by showing that the employer’s proffered explanation is a pretext for discrimination. 31 The legal standard for evaluating pretext is not as clear as the standards for evaluating the first two prongs. As Raad noted in her prehearing brief, the United States Supreme Court has held that a complainant may demonstrate pretext “either directly by persuading the court that a discriminatory reason more likely motivated the employer or indirectly by showing that the employer’s proffered explanation is unworthy of credence.” 32 Because a complainant can show pretext either directly or indirectly, a variety of factors can evidence a pretextual justification.

The commission states on appeal that there are generally three types of evidence used to show pretext: (1) direct evidence of discrimination; (2) comparative evidence; and (3) statistics. 33 As the commission observes, courts have explained that statistical evidence may help determine whether an employer’s actions “conform[ ] to a general pattern of discrimination,” 34 but the “[w]eight to be accorded such statistics is determined by the existence of independent corroborative evidence of discrimination.” 35 Thus, the commission argues, “the fact that the successful applicants were not members of Raad’s protected class is insufficient to establish pretext.”

Citing the district court’s ruling in Raad’s federal case, the superior court noted that a disparity in qualifications among candidates can serve as a basis for a finding of pretext. But the superior court also agreed with the district court's reliance on Bullington v. United Air Lines, Inc., and with the proposition, drawn from Bullington, that the disparity in candidates’ qualifications “must be so apparent as ‘to jump off the page and slap us in the face’ to support a finding of pretext.” 36 The superior court expressly agreed with the district court’s analysis and held that nothing in the record in this case would support finding such a disparity.

The Ninth Circuit rejected the Bullington analysis in reversing the district court’s decision. The court of appeals held that “[u]n-like the Tenth Circuit, we have never followed the Fifth Circuit in holding that the disparity in candidates’ qualifications ‘must be so apparent as to jump off the page and slap us in the face to support a finding of pretext.’ ” 37 In reversing the summary judgment, the court of appeals held that “Raad has demonstrated a genuine factual dispute as to whether the District’s proffered reasons were pretextual.” 38

The Ninth Circuit, citing its earlier decision in Odima v. Westin Tucson Hotel, 39 noted that “[i]n this Circuit, we have held that a finding ‘that a Title VII plaintiffs qualifications were clearly superior to the qualifications of the applicant selected is a proper basis for a finding of discrimination.’ ” 40 The Ninth Circuit observed that “[i]n Odima, we held that the plaintiffs superior qualifications standing alone were enough to prove pretext and, on that basis, we affirmed the district court’s entry of judgment for the plaintiff following a bench trial.” 41

Although the Ninth Circuit held that disparity in qualifications can demonstrate pretext in disparate treatment cases, it noted that “there was no logical reason” that disparity in qualifications would be probative of pretext in retaliation claims. 42 It also noted that in discrimination cases, a disparity in qualifications “naturally gives rise to an inference that the non-discriminatory explanation offered by the employer is pretextual.” 43 By contrast, the fact that an employer hired a less qualified applicant does not give rise to an inference that an employer knew of a complainant’s prior discrimination complaints. 44

C. The Hearing Examiner’s Approach

The hearing examiner first set out the McDonnell Douglas framework, and then discussed how the test would apply to Raad’s allegations. The allegations involved hiring decisions made by twelve school principals for thirty-one teaching positions over the course of three years. The hearing examiner then conducted a McDonnell Douglas analysis for the hiring decisions for each of the thirty-one positions.

The hearing examiner observed that it was undisputed that Raad’s religion is Islam, that her national origin is Lebanese, and that these are protected classes under Alaska law. 45 The hearing examiner then found that “[i]t cannot be disputed that Raad’s membership in these protected classes is not readily apparent.” The hearing examiner observed that “Raad is a Caucasian woman with light skin and dark hair. Overall, her features do not bespeak of any particular national origin. She does speak with a foreign accent and ... her accent will be relevant to her prima facie case of national origin discrimination. Nothing about her appearance indicates any particular religious preference.” (Emphasis deleted.)

Alaska Statute 18.80.220(a)(1) makes it illegal for an employer to refuse to hire a person “because of’ the person’s religion or national origin. Because the hearing examiner found that Raad’s religion, national origin, and the fact that she filed a prior complaint were not readily apparent, the hearing examiner read AS 18.80.220(a)(1) and relevant case law as requiring Raad, in establishing her prima facie case, to show that the hiring principals knew that she was a member of the relevant protected class. The hearing examiner noted that under subsection .220(a)(1), an employer could not discriminate “because of’ a person’s national origin or religion if the employer was not aware of it. This approach is consistent with case law cited by the hearing examiner. 46

The hearing examiner then noted that it was undisputed that the district hired applicants for the thirty-one teaching positions that Raad applied for, that Raad was at least minimally qualified for each position, and that Raad was not hired for any of the positions. The hearing examiner therefore concluded that Raad would need to prove the following elements to establish her prima facie case of discrimination as to any given position: (1) that the principal who made the hiring decision knew of Raad’s national origin or religion; and (2) that the individual hired for the position was neither Lebanese nor Muslim.

With respect to the second requirement, the hearing examiner found that the district admitted that none of the successful applicants for the thirty-one positions was a member of Raad’s protected classes. The examiner based this finding on Paragraph No. 6 of the district’s answer to Raad’s complaint. Paragraph No. 6 stated in part: “Respondent affirmatively asserts its belief that complainant is the only Lebanese or Moslem applicant who has applied for a teaching position with respondent.” Because of the district’s admission, the hearing examiner concluded that Raad established the second prong of her prima facie showing. Under the framework applied by the hearing examiner, Raad only needed to show that a hiring principal was aware of her national origin or religion to make her prima facie showing on a given discrimination claim.

With respect to Raad’s claims of national origin discrimination, the hearing examiner’s finding summary stated that “[t]here was some evidence presented that a few of the hiring principals may have known Raad’s national origin. The more persuasive evidence is that most did not.” The hearing examiner then stated that “[e]ven if Raad had established a prima facie case as to one or more of the 31 positions, the principals articulated legitimate, non-discriminatory and non-retaliatory reasons for their hiring decisions.”

The hearing examiner’s summary findings suggested that only “a few of the hiring principals may -have known Raad’s national origin.” But the hearing examiner’s specific findings-which addressed each principal’s hiring decisions individually — indicated that the hearing examiner concluded outright, or assumed for the purposes of the decision, that, for twenty-eight of the thirty-one hiring decisions, the principals either knew of Raad’s national origin or were aware that she was of some foreign national origin. The hearing examiner concluded that there was no evidence presented that the hiring principals for the other three positions had, or could have had, any knowledge of Raad’s national origin. The hearing examiner concluded that there was insufficient evidence presented to show that Principals Ofelt, Thibodeau, or Conwell knew of Raad’s national origin at the time of the relevant hiring decisions.

Raad argues on appeal that her name, “Nada Hashem Itani Raad,” is identifiably “Arab,” and that she “speaks with a distinctive identifiable Middle East accent.”

The hearing examiner concluded that knowledge of Raad’s accent would be relevant, but not necessarily dispositive, evidence of knowledge of Raad’s national origin. 47

Despite the hearing examiner’s initial suppositions to the contrary in the findings summary, we accept for the purposes of this appeal that Raad established a prima facie case of discrimination on the basis of national origin for the twenty-eight positions for which the-hearing examiner concluded or assumed that she had established a prima facie case. Raad’s resume on file with the district indicated that she had both training and work experience in Beirut and that she was fluent in Arabic. In light of Raad’s name, her accent, and indications throughout her resume that she was Lebanese, we accept for the purposes of appellate review Raad’s contention that Raad established that the principals either knew of Raad’s national origin, or knew that she was of some foreign origin. Raad claims that she established a prima facie case of discrimination for all thirty-one positions, but we accept the hearing examiner’s conclusion that Raad failed to establish a prima facie case of discrimination on the basis of national origin with respect to the three positions filled by Principals Ofelt, Thibodeau, and Conwell. The record demonstrates that substantial evidence supports the hearing examiner’s conclusion that those principals were unaware of Raad’s national origin.

With respect to Raad’s claims of discrimination on the basis of religion, the hearing examiner found that “[t]he evidence was insufficient to establish that any of the principals knew Raad’s religion. Thus, her religious discrimination claims fail from the outset.” Because Raad’s religion was not readily apparent, and because substantial evidence supports the hearing examiner’s conclusions on this issue, we accept that Raad failed to establish her prima facie case of discrimination on the basis of religion.

With respect to Raad’s claims of retaliation, the hearing examiner found that it was undisputed that Raad had filed a prior complaint of discrimination and that she was therefore a member of the protected class identified in AS 18.80.220(a)(4). The hearing examiner also found that she was subject to adverse employment action — i.e., not hired for the thirty-one positions.

The hearing examiner concluded that to establish a prima facie case of retaliation Raad would have to show: (1) that the principal making the hiring decision knew that Raad made a prior complaint; and (2) that there was a causal connection in time between Raad’s prior complaint and any of the thirty-one hiring decisions, or that there was some other evidence establishing retaliatory motive.

The hearing examiner found that there was evidence that two of the principals, responsible for four of the thirty-one hiring decisions, “may have known that Raad had filed a prior complaint of discrimination. Thus, her retaliation claims as to the remaining 27 positions fail.” We accept the hearing examiner’s conclusion that Raad established her prima facie case of retaliation with respect to only four of the thirty-one positions because the conclusion is supported by substantial evidence. The hearing officer concluded that Raad established her prima facie case of retaliation with respect to the one position filled by Principal Murphy, and for three positions filled in 1996 by Principal Layral. The record supports the hearing examiner’s findings regarding Murphy and Layral’s knowledge of Raad’s prior corn- plaint, and the timing of the employment action.

After outlining the requirements for the prima facie showings of discrimination and retaliation, the hearing examiner considered the reasons offered by the principals for not hiring Raad. The hearing examiner concluded broadly in his summary findings that “[e]ven if Raad had established a prima facie case as to one or more of the 31 positions, the principals articulated legitimate, non-discriminatory and non-retaliatory reasons for their hiring decisions.” (Emphasis deleted.) Despite this broad declaration, the hearing examiner seems to have only considered the district’s reasons for not hiring Raad for twenty-eight positions. The hearing examiner did not make specific findings regarding reasons offered for not hiring Raad for the three positions filled by Principals Ofelt, Thi-bodeau, and Conwell. For these three positions, the hearing examiner simply concluded that Raad did not establish a prima facie case of discrimination or retaliation. In his position-by-position analysis, the hearing examiner found that the district offered legitimate, non-discriminatory, and non-retaliatory reasons for not hiring Raad for each of the twenty-eight positions for which he concluded or assumed that Raad had established a pri-ma facie case of discrimination on the basis of national origin. These included the four positions for which the hearing examiner concluded that Raad had established a prima facie case of retaliation.

Our review of the record confirms that substantial evidence justified the hearing examiner’s conclusion that the district offered legitimate, non-discriminatory, and non-retaliatory reasons for not hiring Raad for these twenty-eight positions. Under the McDonnell Douglas framework, the burden then shifted back to Raad to show that the reasons proffered by the district were pretextual.

D. Pretext

In recommending the dismissal of Raad’s claims, the hearing examiner’s findings summary stated that Raad “did not meet her burden of showing that [the district’s] reasons were pretextual.” But the hearing examiner’s approach to the issue of pretext is not clear from the record.

In setting out the general McDonnell Douglas framework, the hearing examiner stated that if an employer offers legitimate non-discriminatory reasons for an employment action “the burden shifts back to the complainant to persuade the court (or the Commission) that discriminatory reasons more likely motivated the employer. Usually the complainant satisfies this burden by proving that the employer’s articulated reason for its action is in fact pretextual.”

In his position-by-position analysis, the hearing examiner concluded that the reasons offered by the hiring principals were legitimate, non-discriminatory, and non-retaliatory; the hearing examiner then stated that the burden shifted back to Raad to show by a preponderance of the evidence that discriminatory reasons more likely motivated the principals. As to the twenty-eight positions for which Raad established a prima facie case of discrimination, the hearing examiner’s decision then concluded that “[n]o evidence was presented to support such a finding.” (Emphasis added.)

It is not clear how the hearing examiner analyzed the issue of pretext. Similarly, it is not clear what evidence the hearing examiner considered relevant to the pretext issue.

Raad, who represents herself on appeal, seems to argue that she should have an opportunity to show that the reasons the principals gave for not hiring her were, in fact, pretextual. Raad claims she “is asking [for] an opportunity to be fully heard regarding the pretext to [prove] her case of discrimination against the District if the Court can not find it in the record.”

Raad points to some evidence in the record that at least raises questions about the reasons offered by the district. The following two examples are illustrative. The hearing examiner concluded that Principal McGill offered legitimate, non-discriminatory reasons for not hiring Raad. The hearing examiner mentioned McGill’s testimony that she had observed Raad as extraordinarily distressed and nervous prior to an interview in 1993. The hearing examiner accepted McGill’s ob servations of Raad’s behavior as a legitimate, non-discriminatory reason for not hiring her. On appeal Raad points to passages in McGill’s testimony that suggest that McGill might not have attended the interview she claimed to have observed. McGill had testified that her “direct observations [of Raad] were prior to the interview I was part of and during the interview.” But referring to that interview, McGill also later testified: “I’ve told you that I don’t remember being in the room. So would it surprise me that other people don’t remember me being in the room, I can’t say that either way.” This latter testimony permits an inference that McGill was not actually at the interview she earlier claimed she “was part of.” Although not dispositive of the issue, the inconsistency casts some doubt on the reason McGill gave for not hiring Raad, and is at least some evidence of pretext.

Raad raises a similar question about Principal McDaniel’s reasons for not hiring Raad. The hearing examiner accepted McDaniel’s testimony that he did not hire Raad because he was looking for a teacher who could work well with underachieving or “at risk” students. Raad argues that this reason was pretextual, and points to two letters of recommendation on file with the district specifically praising her for her work with “at risk” children. One of the letters reads, “On her own initiative and time, Nada set up a tutoring session for students at risk of failing. These sessions had a positive impact on the students involved.” The recommendation letter does not necessarily undermine the reasons the hearing examiner accepted from Principal McDaniel, but the recommendation casts some doubt on those reasons, and permits an inference that the reasons he gave were pretextual.

As mentioned above, a complainant has various ways of showing that the reasons an employer gives for a discriminatory hiring decision are pretextual. 48 A complainant may show pretext either directly or indirectly. 49 Even as to claims on which the evidence is barely sufficient to establish a prima facie case of discrimination, that evidence remains relevant to the issue of pretext.

For this reason, and in light of the permissible inferences identified by Raad regarding the reasons Principals McGill and McDaniel gave for not hiring her, the hearing examiner’s conclusion that there was “no evidence” of pretext is problematic. The hearing examiner’s conclusion that Raad “did not carry her burden” with respect to pretext suggests that he may have considered evidence of pretext and found it unpersuasive. But because we cannot determine from the record whether the hearing examiner adequately considered Raad’s claim that the district’s proffered reasons for not hiring her were pretextual, we must remand.

We do not substitute our view of the evidence for that of the hearing examiner. However, we do see some evidence in the record that at least suggests pretext, yet we cannot tell from the hearing examiner’s findings how, or if, he considered that evidence at the pretext stage.

In reviewing the commission’s decision, the superior court cited the Bulling-ton 50 test and stated that evidence of disparity in qualifications did not sufficiently “slap” the court in its face to demonstrate pretext. The Ninth Circuit rejected the Bullington test, and so do we. Evidence of disparity in qualifications need not be sufficient to “slap” a hearing examiner in the face to justify a finding of pretext. It is not clear whether the hearing examiner incorrectly applied such a strict standard to Raad’s complaint, but, again, we cannot say with assurance what standard the hearing examiner did apply-

To the extent the hearing examiner found that there was “no evidence” of pretext, that finding is erroneous. As Raad implies on appeal, the record includes at least some evidence permitting an inference that some of the district’s proffered reasons for not hiring her were pretextual. In resolving the pretext dispute, the hearing examiner should have again considered any evidence Raad used in establishing her prima facie case of discrimination that may have undermined the justifications offered by the district. The hearing examiner should also have considered any direct evidence Raad presented undermining the district’s reasons for not hiring her.

E. Disputed Issues on Remand

Although we accept the hearing examiner’s assumptions or conclusions for the purposes of appellate review, we do not mean to prevent the parties from relitigating unresolved contested fact disputes on remand. For example, in Part III.C we accept for purposes of discussion that Raad established a prima facie ease of national origin discrimination with respect to twenty-eight positions, and a prima facie case of retaliation with respect to four positions. Our acceptance of those propositions should not be interpreted as a holding that Raad established a prima facie case with respect to those positions. And it should not be read as precluding the parties on remand from litigating any genuine factual disputes about whether she established prima facie cases of discrimination or retaliation. This would be so as to those hiring decisions for which the hearing examiner made no conclusive factual findings and merely assumed for the sake of argument that Raad had established a prima facie case of national origin discrimination or retaliation, requiring him to examine the reasons why Raad was not hired.

On the other hand, the hearing examiner made several determinations that will bind the parties on remand. The hearing examiner concluded that Raad satisfied her burden of proving a prima facie ease of national origin discrimination as to those principals who the hearing examiner found either knew Raad was from Lebanon or “more likely than not knew that Raad was of some foreign nationality.” Likewise, the hearing examiner concluded that Raad established her prima facie case of retaliation as to at least one principal found to have known of Raad’s complaint against the district. The commission did not appeal these factual findings and is therefore not free to relitigate these determinations on remand.

Whether the hearing examiner concluded, after making appropriate factual findings, or simply assumed for discussion’s sake that Raad established prima facie cases, he treated the validity of the district’s reasons for not hiring Raad as the determinative issue. Because the hearing examiner reached the district’s reasons for not hiring Raad, he had to consider whether those reasons were pre-textual. And as stated above, we remand because it is not clear from the record whether the hearing examiner adequately considered evidence of pretext.

IY. CONCLUSION

For these reasons, we REVERSE and REMAND. 51

1 . Raad's Human Rights Commission complaint described herself as a "woman whose National Origin is Lebanese and whose religion is Mos-lem.” The hearing examiner used similar terminology. Although Raad’s appellate briefs also refer to herself as "Moslem,” we use the phrase "Muslim.”

2 . The commission first inadvertently issued an order adopting the hearing examiner's preliminary findings. During the subsequent appeal the superior court issued a limited remand back to the commission to correct this mistake. The commission then issued a corrected final order on January 28, 2002 incorporating the hearing examiner's final recommended findings of fact, conclusions of law, and order.

3 . Raad v. Fairbanks North Star Borough Sch. Dist., No. F97-0068-CV slip op. (D.Alaska, July 17, 2000).

4 . Raad v. Fairbanks North Star Borough Sch. Dist., 323 F.3d 1185 (9th Cir.2003). The Ninth Circuit reversed and remanded with respect to Raad’s claims of disparate treatment based on national origin and religion, and Raad's claim of retaliation relating to her disciplinary suspension. As amended on May 8, 2003, the Ninth Circuit's opinion noted that Raad did not appeal summary judgment on several claims, and affirmed summary judgment on Raad’s other claims. Id. at 1198.

5 . Raad, 323 F.3d at 1188, 1192.

6 . Id. at 1193, 1197.

7 . Id. at 1194-95, 1197.

8 . Newmont Alaska Ltd. v. McDowell, 22 P.3d 881, 883 (Alaska 2001) (quoting Blanas v. Brower Co., 938 P.2d 1056, 1059 (Alaska 1997)).

9 . Id. at 883; Strand v. Petersburg Pub. Sch., 659 P.2d 1218, 1220 (Alaska 1983).

10 . Alaska State Comm'n for Human Rights v. Yellow Cab, 611 P.2d 487, 490 (Alaska 1980).

11 . Id.; see also Oceanview Homeowners Ass’n v. Quadrant Constr. & Eng'g, 680 P.2d 793, 798 (Alaska 1984) ("[I]t is not our function to evaluate the strength or weakness of evidence presented to administrative agencies.... '[W]hat matters is whether the determination of the [agency] is supported by substantial evidence on the whole record.' " (quoting Anderson v. Employers Liab. Assurance Corp., 498 P.2d 288, 290 (Alaska 1972))).

12 . Newmont Alaska Ltd., 22 P.3d at 883.

13 . W.R. Grasle Co. v. Alaska Workmen's Comp. Bd., 517 P.2d 999, 1003 (Alaska 1974).

14 . AS 18.80.220(a)(1).

15 . AS 18.80.220(a)(4) makes it unlawful for an employer to "discriminate against a person because ... the person has filed a complaint, testified, or assisted in a proceeding under [AS 18.80].”

16 . The McDonnell Douglas test is named after the United States Supreme Court case in which it was first enunciated, McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973). Alaska adopted the McDonnell Douglas test in Brown v. Wood, 575 P.2d 760, 770 (Alaska 1978). See also Veco, Inc. v. Rosebrock, 970 P.2d 906, 918-19 (Alaska 1999); State, Dep't of Fish & Game v. Meyer, 906 P.2d 1365, 1374-75 (Alaska 1995); Haroldsen v. Omni Enters., 901 P.2d 426, 430 (Alaska 1995).

In cases in which there is direct evidence of discrimination we apply a mixed-motive analysis instead of the McDonnell Douglas test. Under the mixed-motive approach, if a complainant presents direct evidence of discriminatory intent her ultimate burden of proof is somewhat lower than that required to satisfy the McDonnell Douglas test. Era Aviation, Inc., v. Lindfors, 17 P.3d 40, 44 (Alaska 2000). On appeal Raad seems to imply that a mixed-motive analysis might be appropriate in the present case, but she points to no direct evidence of discriminatory intent. The mixed-motive framework therefore does not apply.

17 . Haroldsen, 901 P.2d at 430.

18 . Id.

19

20 . Alaska State Comm’n for Human Rights v. Yellow Cab, 611 P.2d 487, 490 (Alaska 1980).

21

22 . Haroldsen, 901 P.2d at 431; see also McDonnell Douglas, 411 U.S. at 802 n. 13, 93 S.Ct. 1817.

23 . See, e.g., Yellow Cab, 611 P.2d at 492 (holding prima facie case was established in hiring discrimination case by showing that position remained open before employer hired candidates not belonging to complainant’s protected class); see also Haroldsen, 901 P.2d. at 430-31 (holding that fourth factor of prima facie case requires showing "that others, who are not within the protected class, were treated more favorably”).

24 . Yeco, 970 P.2d at 918-19.

25 . Id. at 919; see also Raad v. Fairbanks North Star Borough Sch. Dist., 323 F.3d 1185, 1196-97 (9th Cir.2003).

26 . Veco, 970 P.2d at 919.

27

28 . Id. (quoting Miller v. Fairchild Indus., Inc., 797 F.2d 727, 731 (9th Cir.1986)); see also Thomas v. Anchorage Tel. Util., 741 P.2d 618, 624 (Alaska 1987) (holding that evidence of legitimate, nondiscriminatory justification is sufficient if it "allow[s] the trier of fact rationally to conclude that discriminatory animus was not the motivating factor in the employment decision”).

29 . Thomas, 741 P.2d at 624 (original emphasis).

30 . Meyer, 906 P.2d at 1375.

31 . Id.; see also Era Aviation, 17 P.3d at 44.

32 . Texas Dep’t of Cmty. Affairs v. Burdine, 450 U.S. 248, 256, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981).

33 . Penk v. Oregon State Bd. of Higher Educ., 816 F.2d 458, 462 (9th Cir.1987) (citing B. Schlei & P. Grossman, Employment Discrimination Law 1314 (2d ed.1983)).

34 . McDonnell Douglas, 411 U.S. at 805, 93 S.Ct. 1817.

35 . Am. Fed. of State, County, & Mun. Employees, AFL-CIO v. Washington, 770 F.2d 1401, 1407 (9th Cir.1985).

36 . The superior court cited Raad v. Fairbanks North Star Borough School District, No. F97-0068-CV (HRH) at 40 (D.Alaska 2000) (quoting Bullington v. United Air Lines, Inc., 186 F.3d 1301, 1319 (10th Cir.1999)).

37 . Raad v. Fairbanks North Star Borough Sch. Dist., 323 F.3d 1185, 1194 (9th Cir.2003) (original emphasis). We similarly reject the Bulling-ton test, and discuss it below in Part III.D.

38

39 . Odima v. Westin Tucson Hotel, 53 F.3d 1484 (9th Cir.1995).

40 . Raad, 323 F.3d at 1194 (quoting Odima, 53 F.3d at 1492).

41 . Id. (original emphasis).

42 . Id. at 1197.

43

44 . Id. at 1198.

45 . See AS 18.80.220(a)(1).

46 . See, e.g., Geraci v. Moody-Tottrup, Int'l, Inc., 82 F.3d 578, 581 (3d Cir.1996) (holding that where disability is not readily apparent, employee must allege knowledge and present evidence from which rational jury could infer knowledge to establish prima facie case of discrimination).

47 . This approach is consistent with the case law cited by the hearing examiner. See, e.g., Ang v. Procter & Gamble Co., 932 F.2d 540, 549 (6th Cir.1991) (holding that discrimination based on manner of speaking can be national origin discrimination, and noting that "accent and national origin are inextricably intertwined”).

48 . See supra notes 30-41 and accompanying text.

49 . Texas Dep’t of Cmty. Affairs v. Burdine, 450 U.S. 248, 256, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981).

50 . Bullington v. United Air Lines, Inc., 186 F.3d 1301, 1319 (10th Cir.1999).

51 . Raad, a pro se appellant, also seems to argue on appeal that the hearing examiner improperly excluded evidence regarding her previous discrimination complaint and the events surrounding her 1993 disciplinary suspension. Her arguments addressing the admission of evidence of the 1993 suspension can be clarified on remand.