JAMES McGIRR KELLY, District Judge.

The court has considered the testimony presented in this case and is now prepared to make its Findings of Fact and Conclusions of Law and decision.

FINDINGS OF FACT

1. The plaintiff’s claim arises under 42 U.S.C. § 2000e-16, alleging discrimination based on race (Caucasian) and color (white) in federal employment.

2. The plaintiff exhausted all administrative remedies within the Defense Logistics Agency (“DLA”), an agency within the Department of Defense, and timely filed his Complaint in the United States District Court for the District of New Jersey. The case was transferred to this Court by stipulation.

3. Jurisdiction is conferred by 28 U.S.C. §§ 1331 and 1343, and by 42 U.S.C. § 2000e.

4. The plaintiff is employed as a chemist in the Clothing and Textiles (“C & T”) Directorate at the DLA’s Defense Person *110 nel Support Center (“DPSC”) in Philadelphia.

5. The DLA is responsible for the procurement of equipment, food, clothes and supplies for military personnel stationed throughout the world. It employs nearly 5,000 people at the DPSC in Philadelphia.

6. Prior to the selection at issue, the DLA had adopted a Multi-Year Affirmative Employment Program Plan for the DPSC for the fiscal years 1988-92 (the “EEO Plan”). The EEO Plan applied to all of the DPSC’s nearly 5,000 employees.

7. According to the EEO Plan, its primary goal was “to achieve a fully integrated workforce in all grade levels and all occupational series.” The EEO Plan further provided:

All merit promotion referral lists are annotated with underrepresentation data to insure selecting officials are aware of the underrepresentation for the particular occupational series and grade. If an underrepresented applicant is referred and not selected the selecting official must prepare a letter providing justification for the selection.

Exhibit D-27, EEO Plan, at 11.

8. In February, 1988, the plaintiff, who is white, applied for a merit promotion to the position of Supervisory Chemist, Series 1320, Grade 13. He was one of ten applicants referred by the DPSC’s Office of Civilian Personnel to the selecting official, Algie Manuel, a black male, for consideration. The Office of Civilian Personnel rated each applicant according to the qualifications required in the Job Announcement Bulletin, called KASOs, and only referred those applicants it considered qualified. To be qualified, an applicant had to be rated 70 or above.

9. At the time of the selection, Manuel was told that “high-grade” positions (GS-13 rank and above) in the C & T Directorate were considered by the DPSC’s Equal Employment Opportunity Office (“EEO Office”) to be underrepresented in the categories of minority and female employees.

10. The EEO Office made the determination that the high-grade positions in the C & T Directorate were underrepresented with the following analysis. They examined the racial composition of all high-grade jobs in the C & T Directorate, which encompassed 26 administrative job positions and one professional job position, and determined that 14.8% of the jobs were filled by minority candidates. See Exhibit P-2, Tab 4, (Dec. 18, 1989). They then examined the statistics for the general civilian workforce in Philadelphia, which was 18.2% minority workers. See Exhibit P-5, Equal Employment Opportunity Office, Defense Personnel Support Center, Participant’s Handbook for Supervisory Equal Employment Opportunity Awareness Training 34 (Jan. 1984) [hereinafter Participant’s Handbook ]. The parity index was determined by taking the ratio of the minorities in the C & T Directorate to the general non-professional civilian workforce. The parity index for the C & T Directorate was 81.3. The goal of the EEO Plan was to achieve a parity index of 100 or more. Because the parity index for all high-grade positions in the C & T Directorate was less than 100, the job position for which the plaintiff applied, GM 1320, GS 13, was considered underrepresented.

11. The professional workforce in Philadelphia is composed of 12.2% minorities. See Participant’s Handbook, at 35.

12. The occupational series 1320, which is considered a professional classification by the Office of Personnel Management, see Exhibit P-4, Tab 12, FPM Letter 720-2, at 10 (Sept. 19, 1979), at all grades is composed of 43.75% minorities, 37.5% blacks, and 25% black males. See Exhibit D-l, Tab 6 (June 30, 1988 statistics).

13. The high-grade positions in occupational series 1320 are 100% minority. See id.

14. The entire distribution in DPSC professional jobs are 17% minority, and for professional and administrative jobs together, minorities are 22.6% of the total composition. See Exhibit P-4, Tab 13, EEOC Form 569 (Aug. 1987).

15. During February and March of 1988, Manuel reviewed the applicants’ SF-171 application forms and their responses *111 to the KASO standards, and interviewed each of the applicants personally. He was not provided with the ratings of the applicants that were prepared by the Office of Civilian Personnel.

16. On March 29, 1988, Manuel selected one of the ten qualified applicants, Paul Conrad, for the position. Because Conrad is white, Manuel wrote a justification letter in accordance with the EEO Plan. However, the Supervisory Quality Assurance Specialist within the C & T Directorate, Jean Anne Grandinetti, refused to approve the selection of Conrad because he had no first-line supervisory experience.

17. Almost immediately thereafter, on March 31, 1988, Manuel selected the plaintiff for the position of Supervisory Chemist, and prepared a justification letter for that selection. This selection was approved within the C & T Directorate by three others: Grandinetti, Lieutenant Colonel Joseph Kernodle who was Deputy Chief of the Quality Assurance Division, and the Logistics Management Officer, Paul Zebowski. Both Manuel and Grandinetti had held the position of Supervisory Chemist, GM 1320, GS 13, previously.

18. The selection of the plaintiff was reviewed within the EEO Office by an EEO specialist, Margaret Haberman. Haber-man prepared an inaccurate grid to compare the qualifications and experience of the plaintiff with the qualifications and experience of the applicants from the allegedly underrepresented groups who were also qualified for the position.

19. The grid substantially understated the supervisory qualifications of the plaintiff, misstated the performance rating of the plaintiff, and omitted some of the awards of the plaintiff.

20. The grid also erroneously stated that the minority candidate Joseph Speight, who was eventually selected for the position, held a masters degree in chemistry. The plaintiff was the only candidate who did hold a masters degree in chemistry; this degree was not required for selection to the position.

21. Haberman used this grid as a guide when reviewing the selection of the plaintiff with Margaret Crumley, the EEO Office Manager at DPSC, who decided that a stronger justification letter on behalf of the plaintiff would be required to justify the plaintiff's selection over the allegedly underrepresented candidates.

22. On April 8, 1988, Crumley prepared a memorandum to Manuel, noting that the C & T Directorate as a whole was “below parity in the employment of minorities and females in high grade positions,” and that two of the underrepresented applicants had more supervisory experience than the plaintiff based on the grid. See Exhibit 22. The EEO Office returned the plaintiff’s selection package, including the grid, to Manuel for additional review and further support for the selection of the plaintiff.

23. According to Crumley, the EEO Office was concerned about the sufficiency of the memorandum in the event a claim of racial discrimination was raised by an applicant from one of the underrepresented groups.

24. Manuel did not feel that he could write a stronger justification letter when faced with the instructions in the memorandum from the EEO Office. He instead selected another of the qualified applicants, Speight, a black male, for the promotion. Manuel decided that although Speight’s qualifications in general were not as high as the plaintiff’s, he had 15 years of supervisory experience instead of the seven years of the plaintiff. 1 Speight was officially approved for the position on June 7, 1988.

CONCLUSIONS OF LAW

1. In order to make out a claim of racial employment discrimination under Title VII, the plaintiff must first make a *112 prima facie showing of discrimination. Second, if the plaintiff succeeds in proving the prima facie case, the burden shifts to the defendant to articulate some legitimate, nondiscriminatory reason for the employee’s rejection. Third, if the defendant does articulate such a reason, the plaintiff must then have an opportunity to prove by a preponderance of the evidence that this reason is not the real reason for the promotion, but only pretextual. See Chipollini v. Spencer Gifts, 814 F.2d 893, 897 (3d Cir.1987).

2. In cases concerning affirmative action plans, if the plaintiff has shown a prima facie case of discrimination, the employer may use the affirmative action plan as the legitimate, nondiscriminatory reason for rejecting the plaintiff. See Johnson v. Transportation Agency, 480 U.S. 616, 626, 107 S.Ct. 1442, 1448, 94 L.Ed.2d 615 (1987).

3. The plaintiff must then show that the use of the affirmative action plan was improper or the plan invalid. The plaintiff may do so in one of two ways that are relevant to this case: the plaintiff must show that a “manifest imbalance” between the job at issue and the proportion of minorities in the labor population did not exist to justify the use of the plan, or that the plan “unnecessarily trammeled” on the rights of the white employees. Id. at 626, 637-38, 107 S.Ct. at 1448, 1454-55.

4. In order to determine if a manifest imbalance between the employer’s jobs and the labor market exists, if “a job requires special training, ... the comparison should be with those in the labor force who possess the relevant qualifications.” Id. at 632, 107 S.Ct. at 1452, citing Hazelwood School District v. U.S., 433 U.S. 299, 308, 97 S.Ct. 2736, 2742, 53 L.Ed.2d 768 (1977) (holding that the court must compare the “racial composition of Hazelwood’s teaching staff and the racial composition of the qualified public school teacher population.”). This requirement avoids an undue burden being placed on the employer to fill skilled positions with minority employees when not enough skilled minority applicants exist to fill the positions. See id.

5. The plaintiff here has succeeded in making out the prima facie case of discrimination because he has shown that the reason he was not chosen for the job was that he was white and the “selectee” was black.

6. The plaintiff has shown that a manifest imbalance did not exist in this case.

a. First, as indicated in the EEO Plan itself and discussed in Johnson, the occupational series and grade at issue should be examined to determine if that particular series and grade were underrepresented. In this case, there is only one position in that occupational series and grade. It cannot be determined what the racial composition of GM 1320, GS 13 is because only one position exists in that description, and the position was vacant. The EEO Office decided that it should examine the entire high-grade C & T Directorate, since it could not get a valid statistical base for comparison from examining GM 1320, GS 13 statistics. This grouping is 96% administrative positions, according to the classifications from the Office of Personnel Management. Therefore, this grouping is not the best statistic for comparison to the GM 1320, GS 13 position, because that position is classified as professional by the Office of Personnel Management. Although an administrative element does exist in the GM 1320, GS 13 position, a better sample would be other professional positions that also have administrative elements.

Some statistics that would have been more probative of the racial composition of GM 1320, GS 13 type jobs would be: 1) all grades of the position of Chemist, that is, all grades of the GM 1320 positions, had a minority composition of 43.75%; 2) the high-grade 1320 occupational series positions are 100% minority employees; 2 and 3) the DPSC professional positions are 17% minority employees and for DPSC administrative combined with professional employees, the group is 22.6% minority employees.

*113 b. Second, the EEO Office did not compare the statistics relating to the job at issue, Chemist, to the “labor force who possesses relevant qualifications.” Johnson, 480 U.S. at 632, 107 S.Ct. at 1452. The EEO used the percentage of minorities in the nonprofessional Philadelphia civilian workforce for comparison. The correct statistic would be to examine the professional workforce statistics in Philadelphia, which is 12.2% minority workers.

c. Even if the percentage used by the EEO Office, 14.8% in all C & T Directorate professional and administrative positions, were used to compare to the correct portion of the Philadelphia workforce, 12.2%, the C & T Directorate shows no imbalance between the racial composition of the at-issue jobs and the racial composition of the skilled Philadelphia workforce. Obviously if any of the other statistics mentioned in paragraph 6a supra were used, no imbalance would exist either.

7. Therefore, the plaintiff has succeeded in showing that the reason for the promotion of Speight over the plaintiff, the execution of an affirmative action plan, was merely a pretext because the manifest imbalance required to be present before the use of an affirmative action plan is permissible did not exist. The plaintiff, then, has satisfied his burden of showing impermissible employment discrimination due to race according to the requirements of Title VII.

. According to the grid made by Haberman, the plaintiff only had nine months of supervisory experience. She made this determination by looking only at the job titles possessed by the plaintiff in previous positions and if these job titles did not state "supervisor,” she did not consider them supervisory. In reality, the plaintiff had seven years of supervisory experience based on the actual job descriptions of his previous jobs.

. This sample is only a few positions, though, so it probably would be too small for statistical validity.