UNPUBLISHED

UNITED STATES COURT OF APPEALS

FOR THE FOURTH CIRCUIT

EMILIE S. FRAZIER,
Plaintiff-Appellant,

v.
                                                                     No. 95-1290
LLOYD M. BENTSEN, Secretary of
Treasury,
Defendant-Appellee.

Appeal from the United States District Court
for the Eastern District of Virginia, at Alexandria.
Claude M. Hilton, District Judge.
(CA-94-791-A)

Argued: January 31, 1996

Decided: August 8, 1996

Before RUSSELL and HAMILTON, Circuit Judges, and BLAKE,
United States District Judge for the District of Maryland,
sitting by designation.

_________________________________________________________________

Affirmed by unpublished per curiam opinion.

_________________________________________________________________

COUNSEL

ARGUED: John F. Karl, Jr., MCDONALD & KARL, Washington,
D.C., for Appellant. Theresa Carroll Buchanan, Assistant United
States Attorney, Alexandria, Virginia, for Appellee. ON BRIEF:
Michaele Snyder Battles, KIBLAN & BATTLES, McLean, Virginia,
for Appellant. Helen F. Fahey, United States Attorney, Alexandria,
Virginia, for Appellee.
Unpublished opinions are not binding precedent in this circuit. See
Local Rule 36(c).

_________________________________________________________________

OPINION

PER CURIAM:

Emilie S. Frazier, a white, Jewish woman, applied for a competi-
tive position with the Internal Revenue Service ("IRS"). When she did
not receive the position, she brought an action against the Secretary
of the Treasury under Title VII of the Civil Rights Act of 1964, 42
U.S.C. § 2000e, et seq. She appeals from the district court's granting
of summary judgment in favor of the defendant. For the following
reasons, we affirm the judgment.

I.

Frazier was an employee of the IRS from October 1978 until Sep-
tember 1985, when she resigned to raise her children. At the time of
her resignation, she held the position of Computer Specialist and drew
salary at the GS-12 level. She was an exemplary employee. Her final
evaluation, dated January 11, 1985, resulted in an"Outstanding" per-
formance appraisal. Furthermore, she received two awards during her
tenure with the IRS: a Special Achievement Award in October 1983
and a Sustained Superior Performance Award in 1985.

When her youngest child reached school age, Frazier sought re-
employment with the IRS. In May 1992, she applied for the position
of Program Analyst in its Information Systems Development Office.
Her application included her final evaluation and the awards she had
earned.

The position for which Frazier applied was open to current and for-
mer employees of the government and was subject to the collective
bargaining agreement between the National Treasury Employees
Union and the IRS, known as NORD III. The hiring procedure
required the IRS to appoint a "ranking official," who evaluated the
applications for the position according to a scoring system provided

                    2
in article 13, section 5(J) of NORD III.1 The ranking official then
placed the top six candidates on the "best qualified" list, which was
forwarded to the selecting official. The selecting official interviewed
these "best qualified" candidates and chose one to fill the open posi-
tion.

Pamela J. Sweeney, a white woman, was the ranking official who
conducted the initial scoring of the applications for the Program Ana-
lyst position. When considering Frazier's application, Sweeney gave
Frazier full credit for her "outstanding" performance appraisal and for
her Sustained Superior Performance award. As a result, Sweeney
placed Frazier on the "best qualified" list. When Sweeney completed
her evaluations, she returned the list to the personnel office.

Fayette Forbes, an African-American woman, was the personnel
staffing specialist who reviewed the accuracy of the evaluations. For-
bes discovered a number of errors with respect to Frazier's applica-
tion. She noted that Frazier should not have received full credit for
her "Outstanding" performance appraisal. The vacancy announcement
for the position, as well as § 0335.263(3)(b) of the Internal Revenue
Manual, required applicants to submit a performance appraisal that
was completed within the last twelve months. Section 4I of the Hand-
book provided that if an applicant failed to submit a current perfor-
mance appraisal, he or she is presumed to have average performance
rankings. Because Frazier's "Outstanding" performance evaluation
was more than seven years old, Forbes concluded that Frazier should
have received scores for average performance instead of outstanding
performance. Forbes also noted that Frazier should not have received
credit for her seven-year-old performance award because, under the
NORD III scoring system, an applicant may receive points only for
performance awards received within the past three years.

After consulting her supervisor, Joyce Sanders, an African-
American woman, Forbes called Sweeney and informed her of the
errors. When Sweeney was informed of the requirements in NORD
_________________________________________________________________
1 Rules governing the application and rating process were also con-
tained in the Contract Administration Handbook (the"Handbook") that
administered NORD III, as well as the Internal Revenue Manual and the
Federal Personnel Regulations.

                    3
III, the Handbook, and the Internal Revenue Manual, she agreed to
the changes lowering Frazier's score. Forbes also noticed that another
candidate, Karen Rutledge (also an African-American woman), had
not received credit for three performance awards that she had received
within the last three years. When Forbes pointed out this inaccuracy,
Sweeney agreed to that Rutledge's score should be revised upward.
As a result of these changes, Frazier was removed from the "best
qualified" list; Rutledge, who had not previously scored among the
top six applicants, was placed on the list.

The selecting official for the Program Analyst position, Hugh
Davis, an African-American man, received the revised"best quali-
fied" list and interviewed the six candidates. Davis did not interview
Frazier for the position because she was not on the"best qualified"
list. Davis selected Linda Perryman, an African-American woman, to
fill the position, in part, because he had viewed her work previously
during a temporary assignment.

On June 13, 1994, Frazier filed a complaint in the United States
District Court for the Eastern District of Virginia against Lloyd Ben-
tsen, the Secretary of the Treasury (the "Secretary"). In Count I, she
alleged that the IRS's rules precluding the use and consideration of
old performance appraisals and performance awards have a disparate
impact upon white women because they are more likely to be out of
the work force for more than one year while they raise their children.
In Count II, Frazier alleged that Forbes discriminated against her
based on race, color, religion, and national origin when she lowered
her score and removed her from the "best qualified" list.2 After the
close of discovery, the Secretary filed a motion for summary judg-
ment, which the district court granted. Frazier filed a timely appeal.

II.

We review the district court's grant of summary judgment de novo.
Evans v. Technologies Applications & Serv. Co. , 

, 958
_________________________________________________________________
2 Frazier's complaint originally contained three other counts, but the
parties stipulated to their dismissal on November 17, 1994.



                    4
(4th Cir. 1996). To prevail on a motion for summary judgment, the
moving party must demonstrate the absence of a genuine issue of any
material fact such that the party is entitled to judgment as a matter of
law. Fed. R. Civ. P. 56(c); see Anderson v. Liberty Lobby, Inc., 

, 247-48 (1986). To determine whether a genuine issue of
material fact exists, we construe all facts and reasonable inferences
drawn therefrom in favor of the non-moving party. Matsushita Elec.
Indus. Co. v. Zenith Radio Corp., 

, 587 (1986). The non-
moving party may not rest on its pleadings alone, but must show that
specific, material facts exist that give rise to a genuine triable issue.
Celotex Corp. v. Catrett, 

, 324 (1986). The non-moving
party may not "create a genuine issue of material fact through mere
speculation or the building of one inference upon another." Beale v.
Hardy, 

, 214 (4th Cir. 1985). Although courts should be
cautious when considering a summary judgment motion in a discrimi-
nation case because motive is often the critical issue, summary judg-
ment remains appropriate if the plaintiff cannot prevail as a matter of
law. 

. Thus, summary judgment should be
granted where the record taken as a whole could not lead a rational
trier of fact to find for the non-moving party, such as where the non-
moving party has failed to make a sufficient showing on an essential
element of the case that the non-moving party has the burden to
prove. See 

.

A.

We turn first to Frazier's disparate treatment claim. To meet her
burden of proof on summary judgment, Frazier may proceed under
ordinary standards of proof or under the burden-shifting scheme set
forth in McDonnell Douglas Corp. v. Green, 

, 802
(1973). See 

; EEOC v. Clay Printing Co., 

, 940 (4th Cir. 1992). To satisfy ordinary standards of proof,
Frazier must offer direct or circumstantial evidence that the defendant
acted with discriminatory motive. 

. The record
reveals no direct or indirect evidence that Forbes acted with discrimi-
natory intent when she lowered Frazier's application score and
removed her from the best qualified list. In fact, Frazier admitted dur-
ing her deposition that she filed her suit based solely on her personal,

                    5
subjective feeling that she had been discriminated against.3 The courts
have created the alternate burden-shifting scheme set forth in
McDonnell Douglas precisely because it is so difficult to prove dis-
criminatory motive. See Goldberg v. B. Green & Co., 

849 (4th Cir. 1988).

Under the McDonnell Douglas framework, the plaintiff bears the
burden of proving by a preponderance of the evidence a prima facie
case of discrimination. If the plaintiff succeeds in proving a prima
facie case, the burden shifts to the defendant to articulate some legiti-
mate, non-discriminatory reason for its employment action. Should
the defendant carry this burden, the plaintiff has the opportunity to
prove by a preponderance of the evidence that the legitimate reasons
offered by the defendant were not its true reasons, but were a pretext
for discrimination. Texas Dept. of Community Affairs v. Burdine, 

, 252-53 (1981) (citing McDonnell 

). Although the McDonnell Douglas test shifts the burden of pro-
duction, "[t]he ultimate burden of persuading the trier of fact that the
defendant intentionally discriminated against the plaintiff remains at
all times with the plaintiff." 

the summary judgment
stage of the proceedings, "assessing the burden of production helps
the judge determine whether the litigants have created an issue of fact
to be decided by the jury." 

The shifting burdens of pro-
duction "is intended progressively to sharpen the inquiry into the elu-
sive factual question of intentional discrimination." 

prima facie case, the plaintiff must show by a pre-
ponderance of the evidence (1) that she is a member of a protected
_________________________________________________________________
3 In her deposition, the following exchange occurred:

          BY MS. BUCHANAN:

          Q: But you don't have anything specific to support your feel-
          ing that you were being discriminated against individually
          on the basis of race, religion, color, national origin, or sex.
          It's just your own personal feeling that you were.

          BY MS. FRAZIER:

          A: Yes.

Joint Appendix ("J.A.") at 88.

                     6
class; (2) that she applied for and was qualified for the position; (3)
that, despite her qualifications, she was denied the position; and (4)
that, after her rejection, the position remained open and the employer
continued to seek applicants with the plaintiff's qualifications.
McDonnell 

; Williams v. Cerberonics, 

, 455 (4th Cir. 1989). The burden of establishing a prima
facie case is "not onerous," 

, but it requires
the plaintiff to demonstrate that her rejection did not result from the
two most common legitimate reasons for rejecting a job applicant: a
lack of qualifications and the absence of a vacancy in the job sought.
International Brotherhood of Teamsters v. United States, 

, 358 n.44 (1977).

The McDonnell Douglas test was never meant to be an inflexible
formulation of the elements of a prima facie case. The Supreme Court
recognized that "[t]he facts necessarily will vary in Title VII cases,
and the specification . . . of the prima facie proof required from [a
plaintiff] is not necessarily applicable in every respect to differing
factual situations." McDonnell 

n.13. The
McDonnell Douglas test sets forth the elements of a prima facie case
for a plaintiff who claims that he or she was not hired for a job
because of discrimination. Frazier, however, does not claim that she
was the victim of discrimination because she was not hired for the
position of Program Analyst; she claims that, but for discrimination,
she would have remained on the "best qualified" list and would have
been considered for the position by the selecting official. She does not
dispute that the selecting official could still have hired another appli-
cant, even if she had remained on the "best qualified" list.

Because Frazier alleges that discrimination caused her removal
from the "best qualified" list and not her failure to receive the job, we
cannot apply the McDonnell Douglas factors strictly. To make out a
prima facie case of discrimination, therefore, Frazier must show that
she was qualified to be on the "best qualified" list, not simply that she
was qualified for the position of Program Analyst. Frazier cannot
raise the inference that her removal from the "best qualified" list was
the result of discrimination if she cannot first negate the possibility
that she was removed from the list because she was unqualified.

Frazier fails to establish a prima facie case of discrimination
because she was never qualified to be on the "best qualified" list.

                     7
Under the rules for scoring job applications set forth in NORD III and
related regulations, Frazier could not receive credit for her outstand-
ing performance appraisal because it was more than twelve months
old. Furthermore, she could not receive credit for either of her perfor-
mance awards because they were more than three years old. The
applicable rules required the ranking official to presume that Frazier
had only average performance rankings. Because the ranking official
could not use Frazier's old performance evaluation or the old perfor-
mance awards to boost her score, Frazier did not place among the top
six candidates and therefore was not qualified to a spot on the "best
qualified" list.

Even if Frazier could state a prima facie case of discrimination, the
burden would simply shift to the defendant to articulate a legitimate,
non-discriminatory reason for Frazier's removal from the "best quali-
fied" list. The defendant's burden is one only of articulation; it "need
not persuade the court that it was actually motivated by the proffered
reason[ ]." 

. The Secretary can easily articu-
late a legitimate, non-discriminatory reason for Frazier's removal
from the list. When Forbes recalculated Frazier's score, she simply
followed the regulations, which did not allow the ranking official to
consider a performance appraisal review that was more than one year
old or a performance award that was more than three years old.
Because Frazier's outstanding performance review and her perfor-
mance award were seven years old, Forbes had a legitimate, nondis-
criminatory reason to discount Frazier's score.

Thus, the burden would shift back to Frazier to demonstrate by a
preponderance of the evidence that the legitimate, non-discriminatory
reason for removing Frazier from the "best qualified" list was actually
a pretext for discrimination. Frazier cannot meet this burden. Other
than her own speculations, Frazier failed to submit any evidence sug-
gesting Forbes was not motivated by a desire to enforce the rules
when she lowered Frazier's score and removed her from the "best
qualified" list.

Accordingly, we conclude that the district court correctly granted
summary judgment in favor of the defendant on the disparate treat-
ment claim.

                     8
B.

We next turn to Frazier's disparate impact claim. She challenges
the IRS' practice, when evaluating applications, of not considering
performance appraisals that are more than one year old or perfor-
mance awards that are more than three years old. Although the IRS's
practice is neutral on its face, Frazier contends that it has a disparate
impact on women who have been absent from the work force. She
also maintains that the practice affects white women more than black
women because white women tend to leave and re-enter the work
force more often than black women.

To establish a prima facie case of disparate impact, Frazier must
(1) identify a "specific employment practice that is challenged," and
(2) show causation. Watson v. Fort Worth Bank and Trust, 

, 994 (1988). To show causation, Frazier must"offer statistical
evidence of a kind and degree sufficient to show that the practice in
question has caused the exclusion of applicants for jobs or promotions
because of their membership in a protected group." 

offer sufficient statistical evidence to show causa-
tion. Frazier submitted a report by her expert witness, Dr. Lance W.
Saberhagen, supporting her disparate impact claim. However, Dr.
Saberhagen admitted during his deposition that his report showed that
there was no statistical underrepresentation of white women--
Frazier's protected class4--in the IRS. After the deposition, Frazier
submitted a second report prepared by Dr. Saberhagen that provided
additional statistics to support Frazier's disparate impact claim. The
district court refused to consider this evidence because it was submit-
ted after the pretrial conference.
_________________________________________________________________
4 Frazier argues on appeal that the operative class is not white women,
but white women who seek to re-enter the work force after taking mater-
nity leave for more than one year. Cf. Philips v. Martin Marietta Corp.,

(1971) (recognizing for the first time a "sex plus" theory
of discrimination). The plaintiff's expert, however, did not offer any sta-
tistical evidence relating to white women who have re-entered the work
force after taking time off to raise children. In compiling his statistics,
the expert accounted for only race and gender; he did not consider any
other factors.

                     9
Frazier contends that the district court erred in refusing to consider
Dr. Saberhagen's second report. Generally, the district court enjoys
broad discretion in deciding whether or not evidence is admissible,
and we will not reverse its rulings absent an abuse of discretion.
Distaff, Inc. v. Springfield Contracting Corp., 

, 111 (4th
Cir. 1993). Frazier attempted to submit Dr. Saberhagen's second
report after the defendant had successfully discredited his original
report. The second report was an obvious attempt to introduce addi-
tional statistics at the last minute to resurrect Frazier's disparate
impact claim. Frazier submitted the second report three days before
the hearing on the summary judgment motion and two weeks before
trial, giving the defendant little, if any, time to study the new evidence
and to depose Dr. Saberhagen before the hearing on the summary
judgment motion. The district court did not abuse its discretion in
refusing to consider the expert's second report. 5
_________________________________________________________________
5 Frazier also argues that the district court held the parties to different
standards on procedural matters. On September 1, 1994, the district court
entered a Scheduling Order, which directed the parties to complete dis-
covery by November 11, 1994, and set a pretrial conference for Novem-
ber 17, 1994. The order provided that a party could not use as evidence
any exhibit that was not submitted at the pretrial conference. It also
required the parties to file any pretrial motions before the pretrial confer-
ence. Frazier complains that the district court enforced its scheduling
order only against her: the district court refused to consider Dr. Saberha-
gen's second report because it was filed after the pretrial conference, but
it considered the defendant's motion for summary judgment even though
it was filed on December 16, 1994, one month after the pretrial confer-
ence.

The district court did not act with a double standard. Dr. Saberhagen
was not able to complete his first report by the close of discovery. The
district court agreed to grant Dr. Saberhagen additional time to complete
his report, as long as the defendant would have an opportunity to depose
him. Dr. Saberhagen submitted his report on November 17, 1994, and the
defendant scheduled a deposition for December 8, 1994. After the depo-
sition revealed the weakness of Dr. Saberhagen's testimony, the defen-
dant filed a motion for summary judgment within eight days of the
deposition. Given Frazier's delay in submitting her expert's report, it was
appropriate for the district court to consider the defendant's motion for
summary judgment even though it was filed after the pretrial conference.

                    10
Because Frazier did not present sufficient statistical evidence dem-
onstrating that the IRS' hiring policies have a disparate impact on
white women, the district court properly granted summary judgment
in favor of the defendant on Frazier's disparate impact claim.

III.

For the foregoing reasons, we affirm the district court's granting of
summary judgment in favor of the defendant.

AFFIRMED

                    11