NOT PRECEDENTIAL

                       UNITED STATES COURT OF APPEALS
                            FOR THE THIRD CIRCUIT
                                 ___________

                                       No. 16-3899
                                       ___________

                        SHAWN BULIFANT; GARY HUGHES;
                       DANIEL LOPER; JAMES MCCLINTOCK;
                            CHRISTOPHER VERNON,
                                         Appellants

                                             v.

                      DELAWARE RIVER & BAY AUTHORITY

                       ____________________________________

                     On Appeal from the United States District Court
                              for the District of New Jersey
                               (D.N.J. No. 1-14-cv-07384)
                       District Judge: Honorable Noel L. Hillman
                      ____________________________________

                       Submitted under Third Circuit LAR 34.1(a)
                               on Friday, June 16, 2017

            Before: JORDAN, KRAUSE, and GREENBERG, Circuit Judges

                               (Opinion filed: July 7, 2017)


                                        OPINION*




       *
        This disposition is not an opinion of the full Court and pursuant to I.O.P. 5.7
does not constitute binding precedent.
KRAUSE, Circuit Judge.

       Shawn Bulifant, Gary Hughes, Daniel Loper, James McClintock, and Christopher

Vernon appeal the District Court’s grant of summary judgment in favor of Appellee, the

Delaware River & Bay Authority (DRBA), related to their claims under the Age

Discrimination in Employment Act (ADEA). For the reasons that follow, we will affirm

in part, vacate in part, and remand for further proceedings on Hughes’s and McClintock’s

age discrimination claims.

 I.    Background

       Appellants are crew members who have worked as seasonal employees for the

DRBA’s ferry services at various points between 2006 and the present. Appellants

applied for full-time positions with the DRBA in response to job postings in February

2012, September 2012, and January 2013. With limited exception,1 Appellants received

interviews each time they applied, but ultimately were not selected for any of these full-

time positions.

       In making its hiring decisions for these positions, the DRBA employed a

standardized approach in which the same panel of four DRBA employees interviewed

every candidate for a given position using the same preset questions that focused on four

core competencies—functional and technical skills, safety, customer service, and peer

relationships. Based on the candidates’ answers, each panelist assigned the candidates a

numeric score in each competency. The scores of the four panelists were then added

       1
        Vernon did not receive an interview for the February 2012 position, and did not
apply for the January 2013 position.

                                             2
together, and the candidates were ranked in order of their total scores.2 These rankings,

as well comments from the panel on each candidate, were then submitted to human

resources and the managing director for the position in question.

       Although there is no record of what occurred when the rankings were submitted

for the specific positions at issue in this case, the DRBA’s executive director testified that

the rankings are always used as an “important guide” in the ultimate selection. App. 790.

He also testified that while managing directors and human resources “have … the ability

to deviate somewhat from the strict numerical rankings in order to achieve other goals . . .

such as diversity or other specific goals,” App. 788, “a record is made of why [the

DRBA] picked who [it] picked” and an “explanation” is given when such deviation

occurs, App. 791.

       For the February 2012 position, the DRBA followed its rankings to a T, turning

Appellants down in favor of those who ranked above them numerically.3 For the

September 2012 position, however, the DRBA deviated from its standard practice, hiring

the first-through-fourth ranked candidates, ages 52, 52, 24, and 52, but skipping over

Hughes and McClintock, ages 61 and 53, and ranked fifth and sixth, in favor of the


       2
         The record reflects that the competencies were not given equal weight when
calculating a candidate’s total score but does not disclose how they were weighted.
       3
         We reference the February 2012 position for its relevance to reviewing the
DRBA’s general hiring practices and not as the basis for a claim in and of itself. The
District Court held that any claims with respect to that position were time barred because
Appellants missed the deadline to file a related charge with the Equal Employment
Opportunity Commission—a prerequisite to filing an age discrimination claim in federal
court, see 29 U.S.C. § 626(d)(1)—and Appellants do not dispute this ruling on appeal.

                                              3
seventh, eighth and ninth-ranked candidates, ages 35, 26, and 33.4 Despite its executive

director’s testimony that he would “be surprised” if a written explanation addressing this

deviation did not exist, App. 792, the DRBA has not produced any contemporaneous

records documenting the rationale behind the decision. Thereafter, for the January 2013

position, the DRBA resumed its strict adherence to its rankings, rejecting Appellants in

favor of the two highest-rated candidates.

       Appellants filed suit, asserting that they were not hired for these positions due to

their age in violation of the ADEA.5 In addition, because Appellants submitted

complaints to the DRBA regarding this alleged age discrimination after their first round

of unsuccessful applications, Appellants also asserted they were not hired for the two

later positions in retaliation for their complaints—likewise a violation of the ADEA. The

District Court granted summary judgment in favor of the DRBA on the ground that, even

assuming Appellants had established prima facie cases of discrimination and retaliation

under the ADEA, they had not established that the DRBA’s articulated legitimate reasons

for their hiring decisions were pretextual. This appeal followed.

II.    Discussion6

       4
           The remaining, lower-ranked, Appellants also were not offered a position.
       5
         In their complaint, Bulifant, Loper and Vernon also assert they were unlawfully
excluded from applying for additional positions in November 2011 and January 2012.
The District Court concluded that these positions, which were open to internal applicants
only, “d[id] not form the basis for [Appellants’] claims,” App. 8 n.3, and Appellants have
not challenged that finding.
       6
        The District Court had subject-matter jurisdiction pursuant to 28 U.S.C. § 1331,
and we have jurisdiction pursuant to 28 U.S.C. § 1291.

                                              4
       We review the District Court’s grant of summary judgment de novo. Faush v.

Tuesday Morning, Inc., 

, 215 (3d Cir. 2015). Summary judgment is

appropriate where the moving party has established that “there is no genuine dispute as to

any material fact” and, viewing the facts in the light most favorable to the non-moving

party, “the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a);

Burton v. Teleflex Inc., 

, 425 (3d Cir. 2013). Because we conclude that a

dispute of material fact remains only with respect to Hughes’s and McClintock’s age

discrimination claims, we will vacate the District Court’s grant of summary judgment in

the DRBA’s favor on those claims and will affirm on all others.

       A.     Appellants’ Age Discrimination Claims

       To prevail on their ADEA discrimination claims, Appellants must establish by a

preponderance of the evidence that age was the “but-for” cause of the DRBA’s decision

not to hire them. Gross v. FBL Fin. Servs., Inc., 

, 177–78 (2009). Because

Appellants rely on circumstantial evidence to prove their case, we evaluate their claim

using the three-part McDonnell Douglas burden-shifting framework.7 See Willis v.

UPMC Children’s Hosp. of Pittsburgh, 

, 644 (3d Cir. 2015). That


       7
         In addition to arguing that they have carried their burden under McDonnell
Douglas, Appellants also contend that the McDonnell Douglas framework is not
applicable because they have offered “direct evidence” of discrimination—that is,
“evidence which, if believed, would prove the existence of [discrimination] without
inference or presumption.” Torre v. Casio, Inc., 

, 829 (3d Cir. 1994) (citation
and quotation marks omitted). The District Court carefully parsed Appellants’ allegedly
“direct” evidence, and we agree with its conclusion that none of this evidence proves
discrimination, and much of it, when considered in context, does not even support such
an inference.

                                             5
framework requires a plaintiff opposing summary judgment to satisfy the initial burden of

establishing a prima facie case of discrimination; if he does so, the burden of production

then shifts to the defendant to articulate a legitimate, nondiscriminatory reason for its

hiring decision; and if the defendant articulates such a reason, the burden shifts back to

the plaintiff to show by a preponderance of the evidence that the defendant’s proffered

explanation was pretextual. Smith v. City of Allentown, 

, 691 (3d Cir. 2009).

If the plaintiff succeeds in “demonstrating sufficient evidence to allow a finder of fact to

discredit the employer’s proffered justification,” summary judgment is not appropriate.

.

       To state a prima facie case of age discrimination at the first step of the McDonnell

Douglas framework, a plaintiff must demonstrate “(1) the plaintiff is at least forty years

old; (2) the plaintiff suffered an adverse employment decision; (3) the plaintiff was

qualified for the position in question; and (4) the plaintiff was ultimately [not hired in

favor of] another employee who was sufficiently younger so as to support an inference of

a discriminatory motive.” 

. All five Appellants satisfy this initial

burden.8 At the second step of the McDonnell Douglas framework, however, the

rankings establish a legitimate, nondiscriminatory reason for the DRBA’s failure to hire

Bulifant, Loper, and Vernon for any position, and its failure to hire Hughes and

McClintock for the positions other than the one posted in September 2012, as discussed


       8
          Appellee does not appear to dispute Appellants’ qualifications for the relevant
positions, and the comments accompanying Appellants’ interview rankings generally
reflect that the DRBA considered them to be qualified applicants.

                                              6
further below. We agree with the District Court’s conclusion that, to the extent the

DRBA followed its own rankings system, its process was “formal, open, objective, and

documented,” App. 29, and, thus, Appellants have not demonstrated for these positions

that the DRBA’s strict adherence to this system was a pretext for discrimination.9

       The DRBA cannot rely on its rankings, however, to explain its failure to hire

Hughes and McClintock for the September 2012 position. Instead, although no

contemporaneous explanation was documented, the DRBA now offers the non-

discriminatory explanations that (1) the comments accompanying the rankings justified

its hiring decisions; and (2) its hiring of lower-ranked applicants was necessary to

promote diversity in its workforce. Although these explanations were enough to shift the

burden of production back to Appellants, Hughes and McClintock have offered enough

competing evidence that a reasonable jury could conclude these explanations were

pretextual.



       9
         Appellants contend that the DRBA’s reliance on its rankings is pretextual
because the rankings are “entirely subjective in nature.” Appellants’ Br. 33. While we
have cautioned that “low evaluation scores may be a pretext for discrimination,” Tomasso
v. Boeing Co., 

, 706 (3d Cir. 2006), Appellants have offered nothing more
than their own self-serving assessments that they were more qualified than higher-ranked
applicants and an unsubstantiated hearsay statement to support their contention that they
were scored less favorably than other applicants due to their age. Moreover, while
interviewers were given considerable flexibility when making their assessments, the
process was not as opaque as Appellants claim. Each interviewer rated each applicant in
four specific job-related categories, and, unlike the case of Alvarado v. Texas Rangers,

(5th Cir. 2007) on which Appellants rely, the interviewers supported their
numeric rankings with specific comments reflecting their impressions of the strengths
and weaknesses of the individual candidates. See 

(noting
“[Appellant’s] score sheets contain no notes or comments on her interview
performance”).
                                             7
       One way an ADEA plaintiff may demonstrate pretext is by “point[ing] to evidence

that would allow a factfinder to disbelieve the employer’s reason for the adverse

employment action.” 

. Such evidence “must indicate such

weaknesses . . . in the employer’s proffered legitimate reasons” that a reasonable fact-

finder could “conclude the employer’s actions could not have been for nondiscriminatory

reasons.” 

(internal quotation marks omitted).

       Hughes and McClintock have met this burden by identifying three significant

weaknesses in the DRBA’s proffered justifications. First, they point to the DRBA’s

deviation from its rankings system in favor of three significantly younger applicants with

no contemporaneously documented explanation.10 This evidence is compelling, as the

DRBA consistently followed its rankings when hiring for the other positions, and has

acknowledged that it presumptively follows the rankings—so much so that it has a policy

of creating a written record documenting its reasons in the event of a deviation. Thus,

what occurred here—with three applicants in their twenties and thirties leapfrogging two

applicants in their fifties and sixties with no documented explanation as to why—is

significant evidence of pretext in itself.




       10
          The DRBA asserts that because Hughes and McClintock could not identify at
their depositions the ages and qualifications of the lower-ranked candidates selected
ahead of them, they cannot use these candidates as “comparators” for their discrimination
claims. Appellee’s Br. 24 n. 30. Appellants have consistently identified these younger,
lower-ranked applicants who were hired over them in their briefing, and their inability to
identify their comparators by name and to expound on their backgrounds while being
deposed is immaterial to Appellants’ claims.

                                             8
       Second, Hughes and McClintock cast doubt on the DRBA’s supposed reliance on

the comment sheets accompanying the applicants’ numeric rankings. Hughes and

McClintock are correct that the comments sheets are overwhelmingly positive about all

five applicants in question, and thus do little to explain why the three younger applicants

were selected and Hughes and McClintock were not.

       Third, Hughes and McClintock undercut the DRBA’s post-hoc explanation that its

deviation from its rankings was necessary to promote diversity in its workforce. While

the seventh and ninth-ranked applicants may provide diversity that Hughes and

McClintock do not, the eighth ranked applicant—the youngest of the five by almost ten

years—is, like Hughes and McClintock, a white male. Moreover, although the DRBA’s

executive director, when deposed as part of this lawsuit, identified diversity as a potential

reason why the DRBA may have deviated from its rankings, he also testified that a

written record explaining that deviation normally would have been generated. As noted

above, no such record was generated here.

       The District Court did not explicitly rely on either of these justifications as its

basis for granting summary judgment in the DRBA’s favor, but instead held that because

the first, second, and fourth-ranked applicants who were hired were also over forty, “age

does not appear to the reason why” McClintock and Hughes were not hired. App. 30.

The District Court reasoned that, had age been the reason Hughes and McClintock were

not offered positions, “it would follow” that the DRBA would have “skipped over” these

older applicants as well. App. 29.



                                               9
       While we commend the District Court’s careful assessment of the factual record,

we conclude its reasoning misconceives Appellants’ burden under the ADEA. The

ADEA does not require that Appellants prove the DRBA had a blanket policy of not

hiring any crew member over a certain age, but only that Appellants specifically were not

hired because of their age. See El v. SEPTA, 

, 239–40 (3d Cir. 2007) (“Title

VII operates not primarily to the benefit of racial or minority groups, but to ensure that

individual applicants receive the consideration they are due”); Pivirotto v. Innovative

Sys., Inc., 

, 353 (3d Cir. 1999) (“Even if the plaintiff was replaced by

someone within her own class, this . . . does not establish that the employer did not fire

the plaintiff on the basis of her protected status.”). Thus, while evidence of the ages of

other hired applicants may be relevant to whether age was the “but-for cause” of the

DRBA’s decision, 

, it does not establish that the DRBA’s

proffered justifications were legitimate, and not pretextual, as a matter of law.

       In sum, Hughes and McClintock have offered “sufficient evidence to allow a

finder of fact to discredit the employer’s proffered justification.” 

. Appellants will bear the heavier burden at trial of “convinc[ing] the factfinder that

not only was the employer’s proffered reason false, but the real reason [they were not

hired] was impermissible discrimination.” 

. At this stage,

however, they have offered enough evidence of pretext to put their case before a jury.11


       11
          The DRBA makes much of the fact that three interviewers on the September
2012 interview panel were older than forty. Even accepting the non-binding opinions
cited by the DRBA for the proposition that the age of the decision-maker is relevant in an
ADEA case, the members of the interview panel were not the relevant decision-makers
                                             10
       B.     Appellants’ ADEA Retaliation Claim

       We also evaluate Appellants’ retaliation claims under the McDonnell Douglas

framework. Fasold v. Justice, 

, 188 (3d Cir. 2005). To establish a prima

facie case of retaliation, a plaintiff must show “(1) that s/he engaged in a protected

employee activity; (2) that s/he was subject to adverse action by the employer either

subsequent to or contemporaneous with the protected activity; and (3) that there is a

causal connection between the protected activity and the adverse action.” 

not made this prima facie showing because, even assuming they

satisfied the first and second prongs, they have failed at the third prong to establish a

causal connection between their protected activity and the DRBA’s decision not to hire

them. That is, the DRBA has posited Appellants were not hired due to their lower

rankings, and, apart from Hughes’s and McClintock’s retaliation claims with respect to

the September 2012 position, Appellants have not demonstrated to the contrary.

       In contrast to their discrimination claims, however, Hughes and McClintock also

have not demonstrated the “causal connection” necessary to support a retaliation claim

vis-a-vis the September 2012 position. Such a connection is established when the

“proffered evidence, looked at as a whole” supports an inference of causation. Kachmar

v. SunGard Data Sys., Inc., 

, 177 (3d Cir. 1997). The evidence on which



for purposes of Hughes’s and McClintock’s claims. In fact, it is clear from the record
that the interviewers themselves did not discriminate at all, as they ranked Hughes and
McClintock more favorably than the eventually-hired, younger applicants. Thus, the ages
of the interviewers here do not “mitigate[] any inference of pretext from [the] DRBA’s
bypass of Hughes and McClintock.” Appellee’s Br. 25.

                                             11
Hughes and McClintock rely consists of three documents: (1) a May 29, 2012 letter to the

DRBA’s executive director protesting, among other things, the organization’s prior age

discrimination; (2) a March 25, 2013 EEOC discrimination charge filed by Loper; and (3)

a March 31, 2013 letter to the DRBA’s diversity manager protesting its past age

discrimination. None of this evidence, however, gives rise to an inference of causation.

       Two of these documents, the May 29th letter to the DRBA and Loper’s EEOC

charge, fail because Loper is the only signatory to either document, and Hughes and

McClintock have not offered any evidence that reflects the DRBA knew they were

involved in preparing these complaints prior to its decision not to hire them.12 And

although the March 31st letter is signed by all five Appellants, the DRBA, by that date,

had already finalized its hiring for all but one of the September 2012 positions, with the

one remaining hire finalized the following day. Thus, while “temporal proximity”

between a complaint and an adverse action may sometimes create an inference of

retaliation, 

, the inference from this record, if anything, goes the

other way, as Hughes and McClintock would be hard pressed to show (and indeed, have

offered no evidence) that, in the one-day gap at issue, the DRBA based even that final

adverse decision on Appellants’ letter. Without more, Appellants have not created the

inference of causation necessary to establish a prima facie case of retaliation under the

ADEA.


       12
           Although all five Appellants testified that they signed a different version of the
May 29th letter, no such letter has been produced by either party, and there is no evidence
that this version was ever received by the DRBA.

                                             12
III.   Conclusion

       For the foregoing reasons, we will affirm in part, vacate in part, and remand to the

District Court for further proceedings consistent with this opinion.




                                            13