Debra Stoe v. William Barr

United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT

Argued March 2, 2020 Decided May 29, 2020

No. 18-5315

DEBRA STOE,
APPELLANT

v.

WILLIAM P. BARR, ATTORNEY GENERAL, U.S. DEPARTMENT
OF JUSTICE,
APPELLEE

Appeal from the United States District Court
for the District of Columbia
(No. 1:16-cv-01618)

Susan E. Huhta argued the cause for appellant. With her
on the briefs was Julia T. Quinn.

Daniel P. Schaefer, Assistant U.S. Attorney, argued the
cause for appellee. With him on the brief were Jessie K. Liu,
U.S. Attorney at the time the brief was filed, and R. Craig
Lawrence, Assistant U.S. Attorney.

Before: TATEL and PILLARD, Circuit Judges, and
EDWARDS, Senior Circuit Judge.

Opinion for the Court filed by Senior Circuit Judge
EDWARDS.
 2

EDWARDS, Senior Circuit Judge: On August 10, 2016,
Debra Stoe, Appellant, filed suit in District Court against the
Attorney General of the United States in his official capacity as
head of the Department of Justice (“DOJ”). Stoe’s complaint
alleged that DOJ had denied her a promotion to a Division
Director position because of her gender, in violation of 42
U.S.C. § 2000e-16, and her age, in violation of 29 U.S.C.
§ 633a. On August 28, 2018, the District Court issued an order
and memorandum opinion granting summary judgment in
favor of DOJ. Stoe v. Sessions, 324 F. Supp. 3d 176 (D.D.C.
2018). Stoe filed a timely notice of appeal on October 24, 2018.

Summary judgment is appropriate only if “there is no
genuine dispute as to any material fact and the movant is
entitled to judgment as a matter of law.” FED. R. CIV. P. 56(a).
We review the District Court’s grant of summary judgment de
novo, considering the record taken as a whole, Reeves v.
Sanderson Plumbing Prods., Inc., 530 U.S. 133, 150 (2000),
viewing the evidence in the light most favorable to Stoe, and
drawing all reasonable inferences in her favor, Iyoha v.
Architect of the Capitol, 927 F.3d 561, 565 (D.C. Cir. 2019).
We may not make credibility determinations or weigh the
evidence. Id.

After reviewing the record in this case pursuant to these
standards, we conclude that a reasonable jury could find that
DOJ’s proffered nondiscriminatory reason for denying Stoe the
promotion that she sought “was pretextual and that
discrimination was the real reason.” Hamilton v. Geithner, 666
F.3d 1344, 1347 (D.C. Cir. 2012). Accordingly, we reverse the
grant of summary judgment against Stoe and remand the case
so that it may proceed to trial.
 3
I. BACKGROUND

A. Factual Background

1. Stoe’s Background and Experience

In 1998, Debra Stoe began working for DOJ as a GS-11
Social Science Analyst in the Office of Research and
Evaluation at the National Institute of Justice (“NIJ”). Before
starting at DOJ, Stoe had earned a master’s degree and had
worked in the private sector for twelve years as an industrial
engineer. In 2004, Stoe began working as a GS-14 scientist in
the Policy and Standards Division (“Division”) of DOJ’s
Office of Science and Technology (“OST”). Her position was
reclassified as “Physical Scientist” in 2010, and she held that
position until 2018.

The Division is responsible for developing performance
standards for law enforcement equipment and technology (e.g.,
bulletproof body armor), overseeing conformity assessment
and compliance testing programs, and developing policy for
the adoption and use of law enforcement-related technology.

Stoe’s Work in the Policy and Standards Division. For
many years, Stoe was the only program manager in the
Division. She developed “new, non-traditional approaches . . .
needed for the NIJ Standards and Testing Program to reach its
full potential,” helping to update outmoded standards. Joint
Appendix (“JA”) 1184. Among her many accomplishments,
Stoe redesigned the system for NIJ standards development; she
also oversaw the training and work product of other program
managers in standards development protocols. One of the
program managers who worked under Stoe’s direction was
Mark Greene, who was later selected for the Division Director
position that Stoe was denied.
 4
Stoe “implemented a new method for developing
standards at NIJ called ‘The Special Technical Committee
(STC) Process.’” JA 1331. Her “outstanding leadership led to
the publication of the CBRN Certification Document,” which
was characterized as “the gold standard” in NIJ’s “new STC
process.” It was predicted to “impact the safety of public safety
in immeasurable ways.” JA 1409.

Between 2004 to 2014, Stoe’s work performance record
was indisputably exemplary. She published at least ten
performance standards, including the first standard for
chemical, biological, radiological, and nuclear protective
ensembles, and a unique ballistic body armor standard, which
became the most downloaded document on NIJ’s website. JA
1179 ¶ 10, 1235, 1409. She also oversaw the development of at
least 20 additional standards and test methods. JA 1179 ¶ 10.

According to her superiors, Stoe’s sterling efforts
“morphed a one[-]person standards program into a multi-
person standards program and in the process morphed her roles
and responsibilities from managing grants and interagency
agreements to managing a program and administering the
activities of others.” JA 1433. Stoe’s “leadership and
revolutionary transformation of a moribund program . . .
demonstrated capabilities that the agency never before
experienced and . . . obtained previously unattainable goals and
objectives.” JA 1188. And her “leadership and management of
NIJ’s standards and testing program . . . contributed
significantly to taxpayer value,” saving the government
“several millions [of] dollars.” JA 1254. Her “radical revision”
of interagency responsibilities eventually received approval
from the White House’s Office of Management and Budget.
JA 1185, 1335-36.
 5
Stoe also significantly revised the Division’s compliance
testing and conformity assessment programs to conform them
to international protocols. JA 1186, 1402. She brought
compliance testing in-house to DOJ, assumed responsibility for
working directly with manufacturers and law enforcement
stakeholders, and increased the “confidence in the
manufacturing quality control process and ultimately in the
safety and effectiveness of the equipment sold to [U.S.]
criminal justice agencies.” JA 1186. Stoe published the
Division’s first compliance testing administrative manual,
which was adopted by other federal agencies. JA 1409.

Stoe’s Work with the Interagency Committee. In 2011,
DOJ appointed Stoe to serve as its alternate representative on
the Interagency Committee on Standards Policy (“ICSP”). The
ICSP is a cabinet-level working group on standards, and it is
comprised of representatives from over thirty federal agencies.
Although Stoe was designated as DOJ’s “alternate” ICSP
representative, she was “doing the real work,” JA 1370,
because the primary DOJ representative did not attend
meetings, JA 1380.

Stoe’s Grants Management Responsibilities. In
connection with her standards and compliance testing
programs, Stoe also had substantial grants management
responsibilities. By 2014, she had distributed over $30 million
in federal funds to support government-funded third-party
research. JA 1180 ¶ 12. The record indicates that Stoe did “an
outstanding job managing grants and portfolios, consistently
tracking and following” grants with “no late grant closeouts . . .
or outstanding issues.” JA 1235.

Stoe’s Performance Evaluations from 2010 to 2014.
Between 2010 and 2014, Davis Hart was Stoe’s first-line
supervisor and George “Chris” Tillery was Stoe’s second-line
 6
supervisor. Both acknowledged that, although Stoe was in a
GS-14 position, she was working at a GS-15 level. See
JA 1191, 1179 ¶¶ 7-9, 1110. In 2010, Stoe spoke to Tillery
about the discrepancy between her classification and the actual
work she was doing, and he agreed that she was performing
GS-15 level work. However, Stoe had to raise the issue several
times with Tillery before he formally requested a desk audit
for her in May 2012. Some of the GS-15 level work performed
by Stoe included supervising and managing the standards
testing, conformity assessment, and compliance testing
programs, and representing DOJ on the ICSP.

Stoe’s performance reviews, co-authored by Tillery and
Hart, consistently rated her at “exceeds expectations.” See,
e.g., JA 1191, 1382, 1392, 1401, 1408. Her personnel file is
replete with glowing comments on the quality of her work and
the significant impact of her efforts on public safety. See, e.g.,
JA 1409. In 2010, Stoe received the Assistant Attorney
General Employee of the Year Award, and in 2012 and 2013,
she was nominated for the Samuel Heyman Service to
America for Justice and Law Enforcement Award. And she
was a coveted speaker at national and international
conferences on standards and testing. JA 1185, 1383, 1402.

In 2013, Stoe’s review stated that her “performance
exceeded expectations to an exceptional degree in all
elements. She consistently demonstrated unusually high
initiative in performing job responsibilities and consistently
performed in a manner, which is significantly beyond what is
expected.” JA 1383. It also noted that her performance was
“made particularly exceptional by the fact that her immediate
supervisor’s position was vacant during much of the reporting
period.” JA 1383. And her superiors acknowledged that they
could not “do full credit to the span of [her] work in the space
available.” JA 1383. At her 2014 mid-year evaluation, Stoe
 7
asked Tillery if there was anything she could improve on, and
he responded, “[N]o, you cannot improve . . . on excellence.
You can’t improve on somebody that exceeds at everything.”
JA 106.

2. Alleged Gender Bias Issues with Tillery

Stoe was one of two women in OST, which had
approximately fourteen employees. Tillery ran OST and was
Stoe’s second-level supervisor beginning in 2010. Christine
Crossland, who was in the Office of Research and Evaluation,
a parallel division within NIJ which works closely with OST,
shared office space with Stoe and routinely interacted with her.
JA 1173 ¶¶ 2, 4. In her Declaration offered in the proceedings
before the District Court, Crossland stated that Tillery “created
and promote[d] a male-dominated workplace culture that [was]
hostile to women.” JA 1174 ¶ 6; see also JA 1181 ¶ 15.
According to Crossland, Tillery treated Stoe in a “markedly
worse” way than he treated male subordinates, and “this biased
treatment ha[d] existed for many years.” JA 1174 ¶ 7.

In her own Declaration regarding her relationship with
Tillery, Stoe stated:

Over the years, especially at meetings and
presentations, Mr. Tillery has interrupted me while
speaking, refused to let me finish speaking,
challenged my authority and belittled me in front of
male colleagues, become angry when I have corrected
a mistake or incorrect statement that he made, or
sometimes rephrase what I had said a moment prior,
as a way of taking credit for my ideas, or suggesting
that I had been inarticulate and that he needed to
translate. Very often, I was and still am the only
woman in the room. Furthermore, Mr. Tillery does not
 8
treat my male colleagues in this manner. With them
he is respectful, deferential, and complimentary.

JA 1181 ¶ 16. Both Stoe and Crossland declared that Tillery’s
treatment of them and other women at DOJ convinced them of
Tillery’s “bias against women.” JA 1181 ¶ 15, 1174 ¶ 6.

Crossland left no doubt that, based on what she had seen,
Tillery treated Stoe differently than he treated her male
colleagues:

Frequently, I have observed Chris talk to Debra in
a way that I would describe as patronizing,
condescending, belittling and sexist. Even though she
is one of the most outstanding performers at NIJ,
Chris frequently speaks to Debra as if he thinks she
does not know what she is talking about—even
though she clearly does. On multiple occasions, I have
heard Chris interrupt, undermine, and insult Debra in
meetings in a way that I cannot imagine he would ever
speak to one of her male peers. In fact, I have never
observed Chris speak to a male colleague in the
dismissive way he frequently speaks to Debra.

JA 1174 ¶ 8.

Although Tillery denied any gender bias, he did admit in
his deposition that he had never helped promote a woman to a
position above the GS-13 level. JA 1162. He claimed that there
were not “a lot of women applying for positions in our office”
and that “we have been under a hiring freeze anyway; so, we
have had very little opportunity to hire anyone.” JA 1161-62.
However, the record indicates that, between 2010 and 2014,
Tillery made at least six promotion decisions elevating men to
GS-14 or GS-15 level positions. JA 1178 ¶ 5.
 9

In 2010, Stoe applied to fill the vacancy in the Division
Director position. She was one of two finalists. The other
finalist was Hart, a male employee who was working in a GS-
14 position. Tillery recommended Hart to the Acting Director
of NIJ, and Hart was chosen. In his 2010 Hiring Memorandum,
Tillery acknowledged that “Ms. Stoe [had] an advantage over
Mr. Hart in regard to having a more detailed and in-depth
understanding of a significant component of OTD’s mission;
specifically, with regard to managing NIJ’s standards
development and compliance-testing programs.” JA 1422. He
also acknowledged that Hart “[did] not have Ms. Stoe’s depth
of experience in compliance-testing and standards
development,” but pointed out that “Mr. Hart’s experience in
those areas [was] not negligible.” JA 1422. Tillery gave Hart
the edge because he had more supervisory and operational
experience in compliance testing and standards than Stoe.

When Stoe was informed of this hiring decision, Tillery
told her that the reason she was not selected was because of her
lack of formal supervisory experience. JA 1178-79 ¶ 6. Tillery
suggested to Stoe that it would be a good idea for her to get
formal supervisory training, which she did. JA 1179 ¶ 6. By
2014, Stoe had completed 80 hours of supervisory management
training.

3. The Events Surrounding Stoe’s Non-Selection for
the Division Director Position

It was well understood that, for a number of years, Stoe
was successfully performing GS-15 level work even though
she was employed in a GS-14 position. Between 2012 and
2014, Hart and Tillery requested “desk audits” to reclassify
Stoe’s position. They made these requests because they
recognized that Stoe was routinely working above her grade
 10
and performing GS-15 level work. When these requests were
denied, Tillery then opted to remove GS-15 level work from
Stoe’s position and reassign this work to the Division Director
position.

In March 2014, Hart announced his retirement. This
resulted in a vacancy in the Division Director position, which
was a GS-15 Supervisory Program Manager position. Tillery
posted the position vacancy in April 2014.

Tillery effectively controlled the process for the selection
of a new Division Director. Tillery selected two other panelists
to help him review applications and interview candidates:
Gordon Gillerman, an expert in conformity assessment from
the National Institute of Science and Technology, and Maria
Swineford, a grants management specialist who did not have a
science background. There were serious discussions about
removing grants management from the Division Director
position, but Tillery still chose Swineford to serve on the panel.
In her deposition testimony, Swineford conceded that she was
“in no position to really make any sort of assessment about [the
candidates’] true backgrounds” in standards and testing and
conformity assessment. JA 1012-13.

In May 2014, Tillery communicated with officials in
Human Resources regarding the candidates on the first
certificate list that they had sent him. Tillery objected to the list
because, in his view,

[t]he applicants in the active certifications [were] a
mixed bag. Many [had] no experience in conformity
assessment (standards and testing). Those that [did]
tend[ed] to have it in only in the area of IT. That
bode[d] ill for their ability to replace [Stoe] as the
Department’s alternate Standard[s] Executive . . . .
 11

JA 1458.

In June 2014, in response to Tillery, Human Resources
sent another certificate list with 77 names. Stoe was ranked in
21st place on this list and Greene was ranked in 72nd place.
Tillery forwarded this certificate list to Gillerman and
Swineford, along with a note stating that “the individual that is
selected will replace the member of our staff currently serving
as one of the two alternate Standards Executives for DOJ” on
the ICSP, which was Stoe. JA 1416. The panelists then divided
up the 77 applications for review to determine whom they
would interview.

Stoe’s application was assigned to and reviewed by
Swineford. After her initial review, Swineford did not
recommend Stoe for an interview because she thought that Stoe
lacked the requisite supervisory experience. However, Tillery
“made the argument to Ms. Swineford that . . . the work that
Ms. Stoe had done with coordinating the standards activities,
with leading the standards activities within NIJ, and with
representation on other standards bodies, in fact, justified or
supported her leadership skills.” JA 264. Swineford then
changed her initial score for Stoe, and this resulted in Stoe
being added to the interview list.

Greene was not initially selected for an interview because
Gillerman concluded that Greene was unqualified “based on
the assumption that the alternate standards exec role and
conformity assessment . . . are critical to the job.” JA 1203.
Gillerman never changed his view that Greene’s application
did not justify an interview. JA 830. And Tillery apparently
agreed with Gillerman’s assessment that Greene did not meet
the criteria to be interviewed. JA 1084. As it turned out, the
only reason Greene was interviewed was because of “a policy
 12
that if one Office of Justice Programs employee on a certificate
was interviewed (in this case, Stoe), then all employees on that
certificate who work in the same component must also be
interviewed.” Br. for Appellee at 10.

“In 2014, Greene had worked for the Division for only four
years, less than two of them as a GS-14. Greene’s first exposure
to standards and conformity assessment at DOJ came from
assignments Stoe was ‘managing.’ His relatively limited
portfolio of standards and grants management work was on
projects that had been shifted to him from Stoe. By 2014,
Greene had not published a single standard, had never
performed GS-15 level work, and he lacked supervisory
experience and supervisory training.” Br. for Appellant at 16-
17 (footnote and citations omitted). Greene had a Ph.D. in
Materials Science and Engineering, completed a postdoctoral
fellowship at the National Institute of Science and Technology,
and had worked in the private sector. However, Greene had
never performed GS-15 level work as Stoe had done. Greene
was 38 years old and Stoe was 60 years old at the time of the
2014 selection.

On or about June 20, 2014, the panelists deliberated and
Tillery then decided whom to interview. Seven candidates were
interviewed. In preparation for the interviews, Tillery wrote
five questions that the panelists should raise with each
candidate.

As Appellant explains, “[t]he first question related to the
‘ability to analyze organizational and operational problems and
develop solutions’ and specifically referenced service on the
ICSP and asked the candidates to ‘describe a situation in which
you identified a problem related to conformity assessment
activities. . . .’ The second question related to ‘knowledge of
program management principles,’ referenced technology
 13
policy, and asked for a situation in which the candidate
‘provided program management advice or assistance. . . .’ The
third related to the ‘ability to provide advice and guidance on
business and program management issues’ and asked the
candidates to ‘describe a situation in which you provided
advice or guidance’ related to ‘grants and/or contract
management.’ The last two questions related to supervisory/
leadership abilities and asked the candidates to ‘describe
situation[s]’ in which they ‘performed a leadership role’ and
‘dealt with a variety of stakeholders.’” Br. for Appellant at 17
n.13; see also JA 1445-52. (The interview questions appear in
the APPENDIX.)

Tillery determined that each question was to be weighted
equally: each panelist was to grade each interviewee on each
question, using a scale of 1 to 5, with 5 being the highest score
possible. Only the first question focused on conformity
assessment and the ability to serve as DOJ’s representative on
the ICSP. None of the questions asked about standards
development or about overall experience.

The interviews took place on July 11 to 16, 2014, with
thirty minutes allotted for each interview. Gillerman and
Swineford followed a pattern of taking notes on each of the
candidates’ responses to each question. Tillery, however, did
not take notes on Greene’s answers to four of the five
questions. For each candidate, Tillery wrote initial scoring
ranges for each question and then picked one score at some
point after the interviews. For example, Tillery gave Greene an
initial score of “3-5” on every question. After the interviews
were over, however, he changed Greene’s scores to either “4”
or “5” for each question (even though he had no notes for
Greene on four of the five questions). Tillery’s initial scoring
ranges for Stoe were “2-3” and “4-5”; however, in his final
assessments, he changed Stoe’s scores to the low end of the
 14
range for three of her five responses. Tillery had notes for all
Stoe’s answers.

A graphic example of Tillery’s scoring is seen in
connection with his assessments of the answers given by Stoe
and Greene to the third question on grants management. Tillery
initially gave Stoe a score of “2-3” on the grants management
question, while giving Greene a score of “3-5” for the same
question. JA 550, 559. Although Tillery did not have any notes
on Greene’s answer, he later went back and assigned Greene a
5, while dropping Stoe’s score to 2.

Stoe’s Interview. Stoe’s interview lasted less than thirty
minutes, apparently because Swineford showed up late. And
Stoe was not asked any follow-up questions. On Stoe’s
response to her first question, Gillerman and Swineford both
graded Stoe a 5. In his initial scores, Tillery gave Stoe a “4-5”
on this question, but then downgraded it to a 4 in his final
scoring. When asked at his deposition about Stoe’s score,
Tillery could not recall why it had been lowered and his notes
do not reveal anything negative. JA 1127-28, 548.

Tillery was critical of Stoe’s response to the second
question because it was focused on her area of expertise.
Although he testified that there had been “[a]bsolutely nothing
wrong with [her response],” he gave her a score of 3. JA 1130.
Swineford and Gillerman each graded Stoe a 4. At one point
during Stoe’s interview, Tillery shook his head to express a
negative reaction to one of her responses. Stoe felt that Tillery
was “attempt[ing] to influence the other two panelists to maybe
grade [her] down or downgrade [her] a little bit on that
particular question.” JA 966.

Swineford felt that, in her response to the third question,
Stoe showed that she had “good experience” in grants
 15
management. Swineford was “impressed” with Stoe’s answer
and gave her a score of 4. JA 1000, 1597. Gillerman gave Stoe
a 3, while Tillery gave her a final score of 2. For the final two
questions, Stoe received scores of 4, 3, and 3 on one, and 5, 5,
and 4 on the other. See JA 634.

Greene’s Interview. Greene’s interview lasted
approximately 45 minutes, with follow-up questions. With
respect to the first question, Gillerman believed Greene’s
“discussion was not focused on the issues of standards
executives or the policy associated with standards executive
work.” JA 849. Tillery also found that Greene “did not have a
specific understanding of the role of the standards executive
within the federal government” and wrote in his notes that
Greene did a “little tap dance around [the] standards
executive’s role.” JA 1123-24. Gillerman and Tillery still gave
Greene a 4 on this answer, and Swineford gave him a 5.

Greene’s answer to the second question focused on his
experience moderating a talk on smart gun technology, during
which he walked “around the room with a microphone like Phil
Donahue” and made sure to keep them from having a
discussion on gun control policy. JA 910-11. Tillery testified
that Greene’s response “involved a more complex issue or set
of issues than Ms. Stoe’s response.” JA 276. Yet, Tillery had
found Stoe’s response to be too focused on her area of expertise
when she discussed her standards program during this question.
Tillery and Gillerman gave Greene’s response a score of 4, and
Swineford gave him a 5 on this question.

Regarding Greene’s response to the third question, dealing
with grants, Swineford noted that Greene “described extensive
experience with not only grants but with interagency
agreements,” but she could not remember during her deposition
whether Greene had discussed all aspects of the grants process.
 16
JA 1031-33. Swineford gave Greene’s response a 4, and Tillery
and Gillerman both gave Greene a 5.

Gillerman’s notes appear to question Greene’s response to
the fourth question: “What was accomplished[?] What did he
do? Leadership?” JA 587. Gillerman gave Greene’s answer a
3, while the other panelists graded it a 4. For the fifth question,
Greene again relied on his moderator example, and was given
a 4 by Tillery and Gillerman and a 5 by Swineford.

Higgins’ Interview. Gillerman and Swineford both
thought Kathleen Higgins, the third finalist, was unprepared
and ranked her the lowest on their combined scores. Tillery was
“very disappointed” with Higgins’ interview, saying that “she
struggled most with the two leadership questions” and that “she
could have done much, much better.” JA 1093-94. According
to Tillery, Higgins “had a wealth of experience, and . . . it did
not come across as well as it should.” JA 1093. Nevertheless,
Tillery gave Higgins a 5 on the first question and a 5 on both
questions dealing with supervisory responsibility. When asked
who Tillery believed did better between Stoe and Higgins
during the interview, Tillery answered that Stoe did, but later
tried to qualify his answer. JA 1105-06.

On the final day of interviews, even though he had not
discussed the deliberation process with the other two panelists
or seen their actual scores, Tillery contacted his supervisor to
tell him that the panel was choosing between Greene and
Higgins. Then on July 18, Tillery sent an email to Gillerman
and Swineford giving them his preliminary scores. No
discussion between the panelists preceded Tillery’s email.
Gillerman and Swineford did not share their preliminary scores
until July 21. In his email to Gillerman and Swineford, Tillery
proposed that the panel “sum [their] final totals for each
candidate and divide by 3 to come to a final score.” JA 626.
 17

Tillery’s final scores had Greene and Higgins tied for first
place with 21 points, and Stoe in third place with 18 points.
Gillerman had Greene and Stoe tied for first place with 20
points, and Higgins in third place with 17 points. Swineford
had Greene in first place with 23 points, Stoe in second place
with 19 points, and Higgins in third place with 18 points.
Swineford scored Greene higher than Stoe on all questions
except for the question on grants, on which she gave them both
a 4. However, at her deposition, Swineford mentioned grants
management as the only area in which Greene was the stronger
candidate. JA 1046-47.

On July 21, the panelists deliberated by telephone.
Swineford did not remember anyone advocating for or even
mentioning Higgins in deliberations. JA 1018-19. Rather, she
stated that “it was primarily a conversation between
recommending Debra or Mark.” JA 993.

In his 2014 Hiring Memorandum, Tillery wrote that “it
was the consensus of the panel that Dr. Greene was best
qualified to fill this position.” JA 387. However, in his
deposition, Gillerman said that he did not recall any such
deliberations, and he said he could not recall coming to a
consensus on the final selection. JA 868-70, 873-74. Gillerman
made it clear that he “was not the selecting official,” which he
clarified to mean that he “was not the final decision maker” in
selecting Greene. JA 873. Tillery assumed that role.

On July 21, 2014, after the panel had conferred by
telephone, Tillery emailed his supervisor to inform him of
Greene’s selection. He said that: “The major differentiators in
Mark’s favor were his detailed understanding of the . . . grants
processes and their issues and his ability to provide guidance
on technology policy.” JA 1436. Yet, Tillery had praised Stoe
 18
for doing “an outstanding job managing grants,” testified that
Stoe had more experience in grants management and business
processes than Greene, and never indicated in Stoe’s
performance reviews that her grants management work was
substandard in any way. JA 1235, 1140.

At one point in his deposition, Tillery stated that he scored
the candidates based solely on how they responded to the
interview questions. JA 1120-21. This is perplexing because
Tillery used a range of scores (“3-5”) in his initial assessments
of Greene on every question; he had no notes on Greene for
four of the five questions; and then, after the interviews, he
gave Greene final scores of 4 or 5 on every question. After
selecting Greene for the position, Tillery wrote to an official in
Human Resources to say that: “I am well aware of [Greene’s]
skills and capabilities having worked closely with him on
numerous projects over the years. Contacting references is not
necessary in this case.” JA 1435.

Tillery later informed Stoe that, although she had not been
selected for the position, she had “scored better than the other
candidates on four out of five interview questions, and second
highest on the question dealing with grants management.”
JA 1180 ¶ 13. This was not true. See JA 634. When Stoe asked
him to clarify why she had not received the position, Tillery
explained that “a candidate would not be selected if they had
scored lower than a ‘3’ on any of the interview questions,
implying that this would be justification for denying a
candidate the position.” JA 1180 ¶ 13. However, there is
nothing in the record to confirm this. And there is nothing to
indicate that Gillerman or Swineford operated on this
assumption.

Tillery testified that he ultimately selected Greene due to
“Dr. Greene having a greater appreciation for the issues of
 19
technology policy and [his] understanding that Dr. Greene had
a more nuanced application of conformity assessment than Ms.
Stoe.” JA 1101. “This was the only time conformity assessment
was cited as a reason for Greene’s selection. It squarely
conflicts with the panelists’ scoring, as no one scored Stoe
lower than Greene on the first question about conformity
assessment.” Br. for Appellant at 25. And Tillery could not
recall any specifics regarding Greene’s discussion of
conformity assessment. Yet, Tillery explained that while Stoe
had more experience with conformity assessment and
understood the process better, Greene had a better
“understanding of potential.” JA 1102-03.

Following her non-selection, most of the work that Stoe
had been doing was reassigned to Greene. As the Division
Director, Greene was assigned to run the standards program,
serve on the ICSP, work on body armor projects, and oversee
conformity assessment duties. All of these duties were
formerly performed by Stoe, who consistently received high
praise for her work in the Division. In addition, Greene’s new
position was stripped of all grants management
responsibilities, just as had been suggested at the start of the
process to fill the vacancy in the Division Director position.

B. Procedural History

On August 10, 2016, Stoe filed a complaint in District
Court, alleging that DOJ denied her a promotion because of her
gender, in violation of 42 U.S.C. § 2000e-16, and her age, in
violation of 29 U.S.C. § 633a. On August 28, 2018, the District
Court entered a final order granting DOJ’s motion for summary
judgment. On October 24, 2018, Stoe filed a timely notice of
appeal.
 20
II. ANALYSIS

A. Standard of Review

“This court reviews the District Court’s ruling on
summary judgment de novo.” Feld v. Fireman’s Fund Ins. Co.,
909 F.3d 1186, 1193 (D.C. Cir. 2018). This means that we must
review the record “taken as a whole.” Matsushita Elec. Indus.
Co. v. Zenith Radio Corp., 475 U.S. 574, 587 (1986). And in
conducting this review, we are required to “examine the facts
in the record and all reasonable inferences derived therefrom in
a light most favorable to the nonmoving party.” Robinson v.
Pezzat, 818 F.3d 1, 8 (D.C. Cir. 2016) (internal quotation marks
and citation omitted). We must then determine whether “there
are any genuine factual issues that properly can be resolved
only by a finder of fact because they may reasonably be
resolved in favor of either party.” Anderson v. Liberty Lobby,
Inc., 477 U.S. 242, 250 (1986). Under this standard, “[t]he
evidence of the non-movant is to be believed.” Id. at 255.

“Summary judgment is appropriate only if the pleadings,
depositions, answers to interrogatories, admissions, and
affidavits filed pursuant to discovery show that, first, ‘there is
no genuine issue as to any material fact’ and, second, ‘the
moving party is entitled to a judgment as a matter of law.’”
Holcomb v. Powell, 433 F.3d 889, 895 (D.C. Cir. 2006)
(quoting FED. R. CIV. P. 56(c)). In other words, if there are no
genuine issues of material fact, the moving party is entitled to
judgment as a matter of law if the nonmoving party “fails to
make a showing sufficient to establish the existence of an
element essential to that party’s case, and on which that party
will bear the burden of proof at trial.” Celotex Corp. v. Catrett,
477 U.S. 317, 322 (1986).
 21
In considering a motion for summary judgment, judges
must ask themselves not whether they think “the evidence
unmistakably favors one side or the other but whether a fair-
minded jury could return a verdict for the plaintiff on the
evidence presented.” Anderson, 477 U.S. at 252. That is, “there
must be evidence on which the jury could reasonably find for
the plaintiff.” Id. And in viewing the evidence in the light most
favorable to the nonmoving party and drawing all reasonable
inferences in her favor, the court’s role is not “to weigh the
evidence and determine the truth of the matter but to determine
whether there is a genuine issue for trial.” Id. at 249. In short,
“[c]redibility determinations, the weighing of the evidence, and
the drawing of legitimate inferences from the facts” are not the
functions of the court. Id. at 255. These are matters for a jury if
the court determines that there is a genuine issue for trial.

B. The Applicable Legal Standards

Stoe’s causes of action in this case arise under 42 U.S.C
§ 2000e-16 and 29 U.S.C. § 633a. Both provisions cover
employment in the federal government. Section 2000e-16(a)
states that “[a]ll personnel actions affecting employees or
applicants for employment . . . in executive agencies . . . shall
be made free from any discrimination based on . . . sex.” 42
U.S.C. § 2000e-16(a). Section 633a(a) states that “[a]ll
personnel actions affecting employees or applicants for
employment who are at least 40 years of age . . . in executive
agencies . . . shall be made free from any discrimination based
on age.” 29 U.S.C. § 633a(a).

Regarding claims arising under § 2000e-16(a), we have
explained the burden-shifting requirements, as follows:

[When] the record contains no direct evidence that the
adverse employment action of which the plaintiff
 22
complains was caused by prohibited discrimination,
we turn to the burden-shifting framework of
McDonnell Douglas Corp. v. Green, 411 U.S. 792,
802–05 (1973), to analyze the claim. Under this
framework, the plaintiff must first establish a prima
facie case of discrimination by showing that: (1) she
is a member of a protected class; (2) she applied for
and was qualified for an available position; (3) despite
her qualifications, she was rejected; and (4) either
someone filled the position or it remained vacant and
the employer continued to seek applicants.
....

Once the plaintiff has established a prima facie
case, the burden shifts to the defendant to produce
evidence that the plaintiff was rejected for a
legitimate, nondiscriminatory reason. . . .
....

. . . If the defendant produces such evidence [of a
nondiscriminatory reason], the McDonnell Douglas
framework—with its presumptions and burdens—
disappears, and the sole remaining issue is
discrimination vel non. At this point, to survive
summary judgment the plaintiff must show that a
reasonable jury could conclude from all of the
evidence that the adverse employment decision was
made for a discriminatory reason. By “all of the
evidence,” we mean any combination of (1) evidence
establishing the plaintiff’s prima facie case; (2)
evidence the plaintiff presents to attack the
employer’s proffered explanation for its actions; and
(3) any further evidence of discrimination that may be
available to the plaintiff, such as independent
evidence of discriminatory statements or attitudes on
 23
the part of the employer. Aka v. Wash. Hosp. Ctr., 156
F.3d 1284, 1289 (D.C. Cir. 1998) (en banc).

Holcomb, 433 F.3d at 895-97 (internal quotation marks and
citations omitted).

With respect to claims of age discrimination arising under
§ 633a(a), we have explained that “plaintiffs can establish
liability . . . in one of two ways. First, they can make use of
the McDonnell Douglas evidentiary framework to establish
that age was the but-for cause of the challenged personnel
action. . . . Second, plaintiffs may establish liability by
showing that age was a factor in the challenged personnel
action.” Ford v. Mabus, 629 F.3d 198, 207 (D.C. Cir. 2010);
see also Reeves, 530 U.S. at 142 (applying the McDonnell
Douglas framework to a claim arising under the Age
Discrimination in Employment Act).

There is no dispute here that Stoe meets all of the criteria
necessary to establish a prima facie case under § 2000e-16(a)
and § 633a(a): (1) she is a woman over the age of forty; (2) she
was qualified and applied for the Division Director position;
(3) she was rejected; and (4) the position went to Greene. It is
also undisputed that DOJ produced evidence that Stoe was
denied promotion for a facially legitimate, nondiscriminatory
reason. See Reeves, 530 U.S. at 142. DOJ claims that the
selection decision was based on the candidates’ interview
performance, and Greene was selected because, according to
DOJ, his “answers better demonstrated his relevant experience
and narrowly superior qualifications across the board than did
Stoe’s.” Br. for Appellee at 21.

Thus, a “central inquiry” here is whether Stoe produced
sufficient evidence for a reasonable jury to find that DOJ’s
“asserted non-discriminatory reason was not the actual reason
 24
and that the employer intentionally discriminated against [her]
on a prohibited basis.” Hamilton, 666 F.3d at 1351 (internal
quotation marks and citation omitted). “Because in appropriate
cases a ‘factfinder’s disbelief of the reasons put forward by the
defendant’ may support an inference of intentional
discrimination, St. Mary’s Honor Ctr. v. Hicks, 509 U.S. 502,
511 (1993), we do not routinely require plaintiffs ‘to submit
evidence over and above rebutting the employer’s stated
explanation in order to avoid summary judgment.’ Aka, 156
F.3d at 1290.” Id.

With these standards in mind, and viewing the evidence in
the light most favorable to Stoe and drawing all reasonable
inferences in her favor, “taking care neither to make credibility
determinations nor to weigh the evidence before us,” id., we
conclude that “a fair-minded jury could return a verdict for
[Stoe] on the evidence presented.” Anderson, 477 U.S. at 252.
The parties do not analyze the evidence separately in relation
to Stoe’s age and sex discrimination claims, instead treating the
claims as rising or falling together on the presence or not of
material factual disputes over whether the employer’s reasons
were pretextual. See Br. for Appellant at 1; Br. for Appellee at
2. We do the same.

C. The Candidates’ Relative Qualifications

Stoe points out that, “[t]ellingly, DOJ never contends that
Greene was the best qualified for the job, and all but concedes
that a jury could find Stoe better qualified than Greene. See
DOJ Br. at 3[5]-[38] (Stoe ‘had more experience in standards
and conformity assessment,’ ‘more relevant experience’ at
OST ‘developing and publishing standards,’ and ‘was in some
ways more qualified on paper than was Greene’).” Reply Br.
for Appellant at 8 n.4. Appellant’s Reply Brief amplifies these
points. We quote at length, both because the material is fully
 25
supported by the record and the differences between Stoe’s and
Greene’s qualifications are aptly highlighted:

[Stoe] had been successfully performing a
majority of the Division Director responsibilities for
many years, while Greene had no comparable
background. This evidence alone would allow a
reasonable juror to conclude that Stoe was
substantially better qualified. DOJ’s brief completely
ignores this record evidence – a damning omission.

DOJ also ignores the uniquely relevant and
superlative quality of Stoe’s “revolutionary work,”
which allowed her to achieve “previously
unattainable goals and objectives.” Indeed, Stoe spent
ten years creating the Office of Science and
Technology’s (“OST”) standards development and
conformity assessment programs, whereas Greene
had only worked on “discrete parts” of the standards
development and conformity assessment programs,
for eighteen months under Stoe’s direction. Simply
put, Stoe was an undisputed expert in the core areas
of responsibility for the position, and Greene was not.
While DOJ stresses that Greene worked on
“technology performance and equipment
performance standards,” it cannot dispute that he had
never published a single standard, whereas Stoe had
personally published ten standards and oversaw the
development of at least twenty more. Likewise, Stoe
was the only candidate with experience serving on the
Interagency Committee on Standards Policy
(“ICSP”), one of the key Division Director
responsibilities. Greene had no ICSP experience
(which led the panel to initially deem him
unqualified), and he was only granted an interview
 26
due to a DOJ policy whereby Greene had to be
interviewed because Stoe received one.

As for government experience, Stoe had twelve
more years of experience working for DOJ than
Greene. She had eight more years of GS-14
experience than Greene and had worked at the GS-15
level for at least four years, whereas Greene had no
GS-15 experience. . . .

. . . DOJ suggests Greene’s experience in the
private sector, his fellowship experience at National
Institute of Science and Technology (“NIST”), and
his PhD, rendered the gap between his and Stoe’s
credentials insubstantial. But DOJ ignores evidence
that before joining DOJ, Stoe had twelve years of
relevant private sector experience, whereas Greene
had at most five. DOJ also fails to explain how
Greene’s NIST fellowship meaningfully enhanced his
qualification for a supervisory position in OST, and
overlooks Stoe’s experience working with NIST for
many years. DOJ’s suggestion that Stoe did not have
substantially better supervisory and leadership
qualifications [DOJ Br. at [37-38]] ignores the fact
that Stoe for years had (informally) supervised the
work of multiple GS-13 and GS-14 program
managers, including Greene himself. DOJ tries to
downplay Stoe’s 80 hours of supervisory training, but
that was 80 more hours than Greene had when he was
selected to lead the division. . . .

Next, DOJ claims Greene had a “good deal” of
experience with grants and other “business-related
aspects” of the position, but ignores evidence that
Stoe had done an “outstanding job” managing grants
 27
for many years. Even Tillery admitted that Stoe had
greater experience managing grants than Greene.
Moreover, DOJ completely ignores evidence that
Tillery knew at the time of the selection decision that
the Division Director role was about to be divested of
grants management responsibilities and was, in fact,
divested of those responsibilities within a year of
Greene’s selection.

Reply Br. for Appellant at 2-5 (footnote and citations omitted).
Even the District Court recognized these marked differences
in the qualifications of Stoe and Greene. Stoe, 324 F. Supp. 3d
at 187-88.

Unsurprisingly, Stoe argues that she should prevail in this
matter because, based on the evidence in the record, a
reasonable jury could find that she was substantially better
qualified for the Division Director position than was Greene.
The Supreme Court has held that “qualifications evidence may
suffice, at least in some circumstances,” to demonstrate that an
employer’s proffered explanation is pretext for discrimination.
Ash v. Tyson Foods, Inc., 546 U.S. 454, 457 (2006). Following
Ash, this court developed a framework for evaluating claims
“involving a comparison of the plaintiff’s qualifications and
those of the successful candidate.” Hamilton, 666 F.3d at 1352
(internal quotation marks omitted) (quoting Aka v. Wash.
Hosp. Ctr., 156 F.3d 1284, 1294 (D.C. Cir. 1998) (en banc)).
In Aka, we noted that:

If a factfinder can conclude that a reasonable
employer would have found the plaintiff to be
significantly better qualified for the job, but this
employer did not, the factfinder can legitimately infer
that the employer consciously selected a less-
qualified candidate—something that employers do
 28
not usually do, unless some other strong
consideration, such as discrimination, enters into the
picture.

156 F.3d at 1294. However, we have been cautious in
explaining the inferences to be drawn from comparative
qualifications evidence:

“[W]e must assume that a reasonable juror who might
disagree with the employer’s decision, but would find
the question close, would not usually infer
discrimination on the basis of a comparison of
qualifications alone.” For this reason, a disparity in
qualifications, standing alone, can support an
inference of discrimination only when the
qualifications gap is “great enough to be inherently
indicative of discrimination”—that is, when the
plaintiff is “markedly more qualified,” “substantially
more qualified,” or “significantly better qualified”
than the successful candidate.

Hamilton, 666 F.3d at 1352 (citations omitted) (first quoting
Aka, 156 F.3d at 1294; then quoting Holcomb, 433 F.3d at
897).

Although the record in this case surely supports Stoe’s
claim that she was better qualified than Greene, we need not
decide whether a jury would find that she was ‘significantly’
or ‘markedly’ more qualified. Holcomb, 433 F.3d at 897; see
also Hamilton, 666 F.3d at 1352 (“Given the record in this
case . . . [this] is a question we need not conclusively
resolve.”). As we noted in Hamilton,

[P]laintiffs are “expressly not limited to comparing
[their] qualifications against those of the successful
 29
applicant; [they] may seek to expose other flaws in the
employer’s explanation.” Holcomb, 433 F.3d at 897;
see also Ash, 546 U.S. at 458 (noting approvingly the
Eleventh Circuit’s suggestion that “superior
qualifications may be probative of pretext when
combined with other evidence”). Here, [Stoe] relies
not only on comparative qualifications evidence, but
also “seek[s] to expose,” Holcomb, 433 F.3d at 897,
procedural irregularities in a highly subjective
selection process. Reviewing the record as a whole,
we agree that the evidence of [Stoe’s] superior
qualifications taken together with “other flaws in the
employer’s explanation,” id., creates a genuine issue
of material fact that only a jury can resolve.

666 F.3d at 1352.

In an attempt to counter the foregoing case authorities,
DOJ relies heavily on Fischbach v. District of Columbia
Department of Corrections, 86 F.3d 1180 (D.C. Cir. 1996).
The court in that case held that “Title VII liability cannot rest
solely upon a judge’s determination that an employer
misjudged the relative qualifications of admittedly qualified
candidates.” Id. at 1183. Suffice it to say that this case involves
very different issues, regarding whether superior
qualifications may be probative of pretext, procedural
irregularities in the selection process, false testimony, and
accumulated evidence of gender bias against the claimant. As
we have explained, the decisions in Reeves, Iyoha, Hamilton,
Holcomb, Aka, and Ash, not Fischbach, control the disposition
of this case.
 30
D. A Reasonable Jury Could Find in Stoe’s Favor Based
on Her Superior Qualifications, the Accumulated
Evidence of Gender Discrimination, and Pretext

As shown above, Stoe offered compelling evidence to
prove that Tillery had: (1) discriminated against Stoe on the
basis of her gender in the past; (2) ruled Stoe out of
consideration for the promotion even before she was
interviewed; (3) designed the interview questions and process
to mask Stoe’s superior qualifications; (4) allowed gender bias
to taint his scoring of the candidates in their interviews; (5)
influenced the other panelists’ scoring to the detriment of Stoe;
(6) infected the selection process with his bias by sending out
his votes to the panel in advance of their final deliberations; (7)
in effect reported to his superior that Greene had been selected
before deliberating with other members of the panel; and (8)
provided shifting and false rationales for his inconsistent
actions. According to Stoe, viewing this evidence in the light
most favorable to her, and drawing all reasonable inferences in
her favor, a reasonable jury could find that DOJ’s proffered
nondiscriminatory reason for denying her the promotion that
she sought was pretextual and that discrimination was the real
reason. We agree.

First, Stoe provided testimony that Tillery discriminated
against her on the basis of gender prior to the contested
promotion action at issue in this case. It is undisputed that
evidence of discriminatory statements or attitudes by a
decision-maker may support a finding of discrimination, even
if the disparaging comments were not made in the context of
the contested employment action. See, e.g., Reeves, 530 U.S. at
152. A reasonable jury could conclude that Stoe’s evidence of
Tillery’s treatment of her and other female colleagues revealed
his “bias against women.” See JA 1181 ¶ 15. For example,
Tillery took credit for Stoe’s ideas, challenged her authority,
 31
belittled her in front of male colleagues, interrupted her while
she was speaking, and became angry if she corrected him. JA
1181 ¶ 16. Further, in her Declaration, Stoe’s colleague,
Crossland corroborated this view of Tillery’s sexist treatment
against women. JA 1173-76. Crossland described Tillery’s
treatment of Stoe as “patronizing, condescending, belittling
and sexist.” JA 1174 ¶ 8. As even the District Court
acknowledged, Stoe’s evidence of Tillery’s treatment of her
and other female colleagues revealed a “pernicious” form of
“sexism.” Stoe, 324 F. Supp. 3d at 197. A jury might find this
to be compelling evidence of discriminatory motive, which
caused Tillery to act against Stoe because of her gender.

Second, Stoe provided testimony that Tillery ruled her out
of consideration even before she was interviewed. For
example, Tillery expressed a desire to replace Stoe on the
ICSP, even though serving on the ICSP would be one the main
responsibilities of the Division Director position. In an email
to Human Resources, Tillery communicated that he was not
satisfied with the list of candidates “to replace Debra.”
JA 1458-59. And later when he forwarded a revised list of
candidates to his fellow panel members, he stated that the
selectee would “replace” Stoe on the ICSP. JA 1416. He knew
at the time that Stoe was applying for the Division Director
position. A reasonable jury certainly might view Tillery’s
emails as further evidence of discriminatory motive showing
that Tillery meant to prevent Stoe from being promoted
because of her gender.

Third, Stoe provided evidence that Tillery designed the
interview process to mask her superior qualifications and to
manipulate the scoring of candidates. A reasonable jury could
easily see this as evidence of pretext. Tillery made the decision
to base the selection entirely on the interviews. Tillery also
wrote and weighted the questions himself. And Tillery adopted
 32
a scoring system that was easily manipulated. The measure of
interview performance is hardly an exact science. Different
interviewers may hear the same thing from the same candidate
and grade it very differently. The evidence in this case proves
the point. This is why, “although employers may of course take
subjective considerations into account in their employment
decisions, courts traditionally treat explanations that rely
heavily on subjective considerations with caution.” Aka, 156
F.3d at 1298. It is well understood that assessments of
interview performances and dubious scoring systems can be
used to cover up discriminatory hiring practices. See, e.g.,
Hamilton, 666 F.3d at 1355-56.

In Hamilton, the interview panelists did not create a
written record of their deliberations or reasons for choosing the
final selectee, leaving the court with no evidence of the
decision-making process. Id. The situation in this case is worse
because of “[t]he irregularities of Tillery’s scoring
methodology, including initially assigning a score range
during the interviews (thereby reserving his ability to settle on
final scores to fit his desired outcome), and then later assigning
Greene scores at the high end of the range and Stoe the low
end, then further lowering one of Stoe’s scores again without
explanation.” Reply Br. for Appellant at 17-18. For example,
Tillery lowered his scoring of Stoe’s response to the first
question but could not recall why during his deposition, nor
did his notes reveal anything negative. JA 1127-28, 548.
Another example is Stoe’s response to the second question:
Tillery had given Stoe a 3 for her response (not a very strong
score), but during his deposition he said that there was nothing
wrong in the way she had responded. JA 1128-30. The record
reveals that during the interviews Tillery assigned Stoe and
Greene a range of scores for each question; he then
subsequently went back and assigned concrete scores at some
point after the interviews. JA 545-53, 555-62. A reasonable
 33
jury could infer that Tillery delayed assigning concrete scores
to Stoe’s and Greene’s interview answers so that he could
manipulate the scores. And a reasonable jury could conclude
that the resulting scores were merely pretextual and gender
bias was the real reason behind Tillery’s scores.

A most telling example is seen in Tillery’s initially giving
Stoe a raw score of “2-3” on the grants management question,
while giving Greene a raw score of “3-5” for the same
question. JA 550, 559. Although Tillery had no notes on
Greene’s answer, he later went back and assigned Greene a 5,
while assigning Stoe a score of 2. In the scoring format
adopted by Tillery, this three-point differential between Stoe’s
and Greene’s scores had a huge impact on the final result.

It is also noteworthy that, on July 16, 2019, immediately
after the interviews concluded and before the panelists’ scores
had been tabulated, Tillery sent a message to his superior
proclaiming that “[i]ts between Mark Greene and Kathy
Higgins.” JA 1442. A reasonable jury might conclude that
Tillery meant to lock in a result favoring Greene before the
panel had deliberated and reached consensus.

Tillery’s interview questions also suggest an attempt to
distract from Stoe’s superior qualifications. Only one of the
five interview questions Tillery wrote focused on a major
responsibility of the Division Director position – the standards
and conformity assessment programs and ICSP representation
– which Stoe had already been performing for years. Although
this was the most relevant question for the position (Stoe was
ranked the highest on this question), Tillery weighed it equally
to the other four questions. Moreover, Tillery did not ask any
questions regarding experience relevant to the position. This is
critical, because Stoe had been performing many of the
responsibilities of the Division Director position already. In
 34
Salazar v. Washington Metropolitan Area Transit Authority,
401 F.3d 504 (D.C. Cir. 2005), the court found that a
reasonable jury could infer that the interview panelists
“selected an interview agenda which, though facially
acceptable, was designed to downplay [the plaintiff’s]
strengths.” Id. at 510. Use of the interview process to minimize
a candidate’s strengths, as a reasonable jury might conclude
occurred in this case, can be taken as pretextual to cover
proscribed discrimination against the candidate. See id. at 509-
10.

Fourth, Stoe provided further evidence that the selection
process created and run by Tillery was anything but fair and,
in fact, likely a pretext to cover his gender bias. For example,
Tillery sent his votes to the panel in advance of their
deliberations. A reasonable jury might view this as an attempt
to influence the panel. Tillery was the lead official throughout
the selection process, which obviously carried significant
weight. He wrote the questions, determined how the
candidates would be scored, sent his scores to the other
panelists before their deliberations, and notified his superior of
a decision that had yet to be reached. Tillery claimed that the
panel reached “consensus” on the selection of Greene.
However, Gillerman said he recalled no consensus ever being
reached. Tillery was also the only person on the panel who
knew the three finalists, and he had a history of gender bias
against Stoe (which the other panelists may not have known
about). Tillery was the only one on the panel who could claim
close familiarity with the Division Director position, which of
course would carry weight in panel interactions. Indeed,
Swineford candidly acknowledged that she had doubts about
her ability to judge the candidates fairly. In this context, a
reasonable jury could find Tillery was in a good position to
influence the scoring of the other two panelists and that he did,
in fact, employ the process that he had created as pretext for
 35
unlawful discrimination. The law is clear that an unfair
selection process is no defense to a claim of discrimination.

The decision in Iyoha, citing Salazar, 401 F.3d at 509,
pointedly states that “when an employer seeks to rely on a
‘fairly administered’ process to justify an employment action,
the process must in fact be fair.” 927 F.3d at 570. The court in
Iyoha further explains that “[a] selection process that relies on
numerical scores given by a panel of interviewers is only as
fair as the panelists who give the scores.” Id. In Iyoha, the court
found that a reasonable jury could find that because the senior
member of the panel had past acts of discrimination toward the
plaintiff and was in a position to potentially influence the
scores given by the other panelists, the plaintiff was not
provided “a fairly administered selection process, and that [the
defendant’s] claim to the contrary is pretextual.” Id. (internal
quotation marks and citation omitted). We understand that, in
this case, the District Court did not have the benefit of the
Iyoha decision, issued in 2019, when it decided the Attorney
General’s summary judgment motion in August 2018.

Fifth, Stoe presented evidence that Tillery provided
shifting and false rationales for the selection of Greene. For
example, Tillery told Stoe that she was not selected because
candidates had to score at least a 3 on every question to be the
final selectee. However, nothing in the record supports this,
and there is nothing to indicate that the other two panelists
operated on this assumption. Furthermore, the record indicates
that Tillery told his supervisor that Greene was selected
because he performed better than Stoe on grants management
and technology policy. JA 1436. And the record additionally
indicates that Tillery testified that Greene was selected due, in
part, to his more “nuanced” appreciation of conformity
assessment. JA 1101. This last claim not only differs from the
prior two explanations, but it cannot be squared with the
 36
panelists’ scoring of the candidates on the conformity
assessment question. No panelist scored Greene higher than
Stoe on Question 1.

A factfinder “can reasonably infer from the falsity of the
explanation that the employer is dissembling to cover up a
discriminatory purpose. Such an inference is consistent with
the general principle of evidence law that the factfinder is
entitled to consider a party’s dishonesty about a material fact
as ‘affirmative evidence of guilt.’” Reeves, 530 U.S. at 147
(quoting Wright v. West, 505 U.S. 277, 296 (1992)). A
reasonable jury could view Tillery’s shifting explanations of
the selection decision as evidence of pretext to cover up his
gender bias.

One final point regarding Tillery’s alleged manipulation of
the selection process is worth mentioning. It concerns Tillery’s
assessments of Higgins during the interview process. Appellant
convincingly explains why Higgins’ inclusion as a finalist in
the selection process neither negates the other inferences of
discrimination that are apparent in this case, nor diminishes the
evidence suggesting that Tillery manipulated the selection
process and used it as a pretext to discriminate against Stoe.
Here is what Appellant says:

Tillery’s top-place scoring of Higgins was
suspicious given his subsequent testimony that he
found that her interview performance was “very
disappoint[ing],” her relevant experience “did not
come across as well as it could,” and she “struggled
with the leadership questions.” [JA 1093-94]. This
testimony is almost impossible to square with
Tillery’s perfect scores for Higgins on the conformity
assessment/ICSP-related question and both questions
on supervisory abilities. [JA 634]. When asked to
 37
explain the apparent inconsistency, Tillery
inexplicably responded that Higgins provided “better
examples” with better “substance” than Stoe, but Stoe
“did a better presentation job than Ms. Higgins.” [JA
1106-07]. Cf. Figueroa [v. Pompeo], 923 F.3d [1078,
1094 (D.C. Cir. 2019)] (“evaluators essentially are
grading candidates on absolute terms and against one
another along a curve, and they therefore should be
able to explain why one candidate’s grade is lower
than others.”). Tillery’s testimony, combined with
Swineford’s recollection that he never even
mentioned Higgins in post-interview deliberations,
[see JA 1018-19], suggests Tillery’s first-place
scoring of Higgins was intended to mask his gender
bias against Stoe. Were that his plan, it was successful
vis-à-vis the District Court, which found Tillery’s
“high score for Higgins [tended] to negate Stoe’s
assertion that Tillery discriminated against her.”
[Stoe, 324 F. Supp. 3d at 199]. But a reasonable juror
could infer that Tillery’s scoring was a sham, and Stoe
outperformed both Higgins and Greene.

Br. for Appellant at 42 n.22. We agree with Appellant that
“[s]ubstantial evidence exists for a reasonable juror to find
Tillery did not score the candidates based on an honest
assessment of their interview performance, and his scoring
was biased against Stoe.” Id. at 42.

DOJ contends that Tillery’s support for Higgins,
including his equal scoring of Higgins and Greene, “negates
any weak inference that Tillery discriminated against Stoe.”
Br. for Appellee at 18. We strongly disagree. It is true that
when a plaintiff claims discrimination after being denied a
position, if the position was filled by another person who is
within the same protected class as the plaintiff, this normally
 38
“cuts strongly against any inference of discrimination.”
Murray v. Gilmore, 406 F.3d 708, 715 (D.C. Cir. 2005). But
see Stella v. Mineta, 284 F.3d 135, 146 (D.C. Cir. 2002)
(holding that “a plaintiff in a discrimination case need not
demonstrate that she [lost out on a promotion to] a person
outside her protected class in order to carry her burden of
establishing a prima facie case”). However, the “same
protected class” caveat cited in Murray has no application in
this case because neither Stoe nor Higgins was hired to fill the
Division Director position.

****

Given the caliber and quantity of the evidence offered by
Stoe in opposition to the motion for summary judgment, we
have no doubt that a reasonable jury could find that DOJ’s
proffered nondiscriminatory reason for denying Stoe the
promotion that she sought was pretextual and that
discrimination was the real reason. The matters at issue in this
case must be decided by a jury.

III. CONCLUSION

We reverse the grant of summary judgment in favor of
DOJ and remand the case to the District Court so that it may
proceed to trial on Stoe’s claims of gender and age
discrimination.
 39
APPENDIX

Interview Questions

(1) One of the major duty assignments associated with
this position is to provide program advice and guidance. As
noted in the vacancy announcement, if you are selected you
will serve as one of the two alternate standards executives for
the Department of Justice. In that regard, describe a situation
in which you identified a problem related to conformity
assessment activities, and evaluated the alternatives to make a
recommendation or decision. What was the problem and who
was affected? How did you generate and evaluate your
alternatives? What was the outcome?

(2) Another of the major duty assignments associated
with this position is program planning and management. As
noted in the vacancy announcement, if you are selected for this
position you will be expected to oversee NIJ’s efforts to
coordinate federal policy as it relates to technology applied to
criminal justice purposes. Please describe situations in which
you provided program management advice and assistance
related to technology policy. As it relates to the above,
describe a situation in which you applied analytical and
evaluative methods and techniques related to program policies
and activities as they relate to major agency programs. What
were the methods and/or techniques you used? Who was
affected by your decision? What was the outcome of your
decision?

(3) Another of the major duty assignments associated
with this position is program planning and management. As
noted in the vacancy announcement, if you are selected for this
position you will be expected to oversee OST’s business
processes as they relate to NIJ’s science and technology
 40
programs. Describe a situation in which you advised or
consulted on program management matters, finance issues, or
data, as they related to grants and/or contract management,
and/or management of agreements; particularly wih [sic] for-
profit entities. What was the problem and who was affected?
How did you generate and evaluate the alternatives to make a
recommendation? What was the outcome?

(4) The fourth major duty assignment associated with this
position is supervisory and/or management responsibilities.
Please describe a situation in which you performed a
leadership role and/or motivated others toward the
accomplishment of a goal. What was involved, what did you
do, and what was the outcome?

(5) Describe a situation in which you dealt with a variety
of stakeholders. Who was involved, what was the issue or
objective, and what was the outcome[?] How did you interact
with the group[?]