SMITH, J.

The Boston Police Superior Officers Federation (federation) brought an action in the nature of certiorari (G.L.c. 249, § 4) in the Superior Court for judicial review of a decision of the Civil Service Commission (commission). That decision reversed a ruling by the personnel administrator (administrator) of the Department of Personnel Administration (DPA) with respect to the 1987 promotional examination for lieutenant within the Boston police department (BPD). A Superior Court judge allowed the commission's motion for judgment on the pleadings (treated as a motion for summary judgment) and denied the federation's motion for summary judgment. The federation has appealed from the judgments entered in the Superior Court.

We summarize the facts which form the background of the appeal. These facts are taken from the motion judge's memorandum of decision. The judge's facts were taken from the findings of fact made by the commission. In 1987, the BPD, under the direction of the DPA, administered a promotional examination for lieutenant. It was the first examination for the position of lieutenant given by the BPD in over ten years. The examination consisted of three components: (1) a multiple choice examination component; (2) a performance component; and (3) a training and experience component, based on a candidate's academic and work history. It is the performance component that is at the heart of the dispute. The performance component consisted of three parts: (1) an in-basket exercise; (2) a video performance exercise; and (3) a leaderless group exercise.

The first part, the in-basket exercise, consisted of thirty questions about the handling of paperwork that might appear on the desk of a police lieutenant. It was not intended to be used or graded independently of the video performance exercise but rather was designed to serve as a pass-fail screening device for proceeding forward with the video and leaderless group exercises and to be used for possible adjustment of a final grade in the video exercise. The video performance exercise presented brief scenarios designed to raise certain management and supervision issues which the candidates were asked to identify and discuss. In the third part of the performance component, the leaderless group exercise, a group of candidates discussed a particular situation together. The three parts of the performance component were intended to function as an integrated, practical assessment of a candidate's supervisory skills.

After the examination had been given but before it was graded, the administrator learned that one Sweeney, a BPD superintendent who served as head of the BPD's bureau of professional standards, had coached a study group preparing for the examination. Sweeney also took part in evaluating and grading the examination, in particular the performance component.

The administrator initiated an investigation into Sweeney's conflicting roles. The investigation included meeting with Sweeney, the police commissioner, members of the BPD, DPA staff members, and representatives from the Attorney General's office. The administrator determined that there had been a breach of the security of the performance component, in particular the video performance and in-basket exercises, because of Sweeney's dual role as coach to test takers and as consultant in the preparation of the video performance exercise and in the grading of the in-basket exercise. The administrator, after consulting with a public safety examination expert, concluded that an examination consisting only of the multiple choice examination component and the training and experience component constituted a fair promotional examination for the position of BPD lieutenant. Thereafter, on January 27, 1988, an eligible list for lieutenant was established from the scores of the combined multiple choice examination and training and experience components. As a result of the administrator's decision, fifteen candidates whose names did not appear on a preliminary eligible list were on the January 27 list. The BPD promoted thirty-three individuals to lieutenant from that list.

Five BPD sergeants filed individual requests with the administrator for a review of whether the examination was a fair test of their fitness to perform the duties of a lieutenant. G.L.c. 31, § 22. They claimed, among other things, that, without the performance component, especially the video performance exercise, the examination was not a fair test. The administrator denied their claims, and the candidates appealed to the commission. G.L.c. 31, § 24. The federation and the BPD intervened in support of the administrator's ruling.

The commission held fourteen days of hearings during which fifteen witnesses, including five expert witnesses, testified. On December 21, 1989, the commission issued its decision in the form of a memorandum that contained its findings of fact and conclusions of law. The commission determined that the security of only one part of the performance component, the video performance exercise, had been compromised. It ruled, contrary to the administrator's ruling, that the multiple choice examination component and training and experience component, without the performance component, did not constitute a fair test for the position of a BPD lieutenant, because those components did not adequately test for supervisory skills. Although the commission declined to order retroactive relief, it gave the administrator alternative methods to solve the problem. He could order a new three-part performance component examination, or he could reinstate the grades of the two parts of the original performance component (the in-basket and leaderless group exercises) that had not been compromised and establish a new eligibility list based on that reconstituted examination. The administrator selected the latter alternative, reinstated the in-basket and leaderless group exercises as graded components, and issued a new eligibility list.

On appeal, both parties agree that, because the federation sought review of the commission's decision pursuant to G.L.c. 249, § 4, the Superior Court judge properly limited his review to correcting "substantial errors of law that affect material rights and are apparent on the record." Debnam v. Belmont, 388 Mass. 632, 635 (1983). The federation, however, claims that the judge committed error in ruling that the commission, in reviewing the fairness of the promotional examination, had broad authority in its quasi judicial capacity to hold evidentiary hearings and make findings of fact and conclusions of law. The federation argues that the commission, instead of conducting a de novo review of the fairness of the examination, should have applied a narrow standard: whether the administrator had committed an error of law or abused his discretion.

We start our analysis with a review of G.L.c. 31, which governs the administration of the civil service system in the Commonwealth. That statute devotes considerable attention to promotional examinations and their formulation and grading, and courts interpreting it have recognized the different responsibilities the administrator and the commission have in the statutory scheme in ensuring that such examinations fairly test the applicants' fitness to perform the primary or dominant duties of the position sought.

General Laws c. 31, § 16, as amended by St. 1981, c. 767, § 16, delegates the determination of the "form, method and subject matter" of promotional examinations to the administrator. In performing that function, the administrator must consult with labor representatives and professionals in the field that is the subject of a promotional examination. Ibid. As a limit on the administrator's discretion, G.L.c. 31, § 16, requires that "[e]xaminations shall fairly test the knowledge, skills and abilities which can be practically and reliably measured and which are actually required to perform the primary or dominant duties of the position for which the examination is held." The administrator, therefore, "is the skilled professional" who is authorized, among other things, "to decide technical matters such as the scoring and interpretation of examinations." Ash v. Police Commr. of Boston, 11 Mass. App. Ct. 650, 652 (1981). The commission is thus relieved of "a substantial amount of detail" and is left "free to formulate policy and to perform quasi legislative and quasi judicial functions." Moore v. Civil Serv. Commn., 333 Mass. 430, 434 (1956).

The administrator's preparation of an examination is expressly subject to further review on the question of fairness. Applicants may request that the administrator review whether an examination constituted a fair test of their fitness to perform the primary or dominant duties of the position sought. G.L.c. 31, § 22. Upon receiving such a request, the administrator must conduct a review, render a decision, and notify the requesting applicant of that decision. G.L.c. 31, § 23. In conducting a review, the administrator does not hold an adversary hearing but rather reconsiders the examination in light of the challenge to its fairness. See Black's Law Dictionary 1320 (6th ed. 1990) where a "review" is defined as a "reconsideration, second view or examination; revision; consideration for purposes of correction."

The Legislature was undoubtedly aware that, in conducting a "review" of the challenged examination, the administrator would be reviewing the very examination that he (and others) had prepared. Therefore, the Legislature, in order to effectuate its goal of fair promotional examinations, provided a further level of appeal if the administrator rejected the challenge — an appeal to the commission. G.L.c. 31, § 24. In doing so, the Legislature specifically provided that the commission must hold a "hearing" in regard to the challenge. Ibid. A "hearing," in contrast to a "review," is a "proceeding of relative formality (though generally less formal than a trial), generally public, with definite issues of fact or of law to be tried, in which witnesses are heard and evidence presented." Black's Law Dictionary 721 (6th ed. 1990). When conducting such a hearing, the commission, acting in its quasi judicial capacity, has the authority to reverse a ruling of the administrator and request that a new examination be held where the examination did not afford the applicants a fair opportunity to demonstrate their abilities. DiRado v. Civil Serv. Commn., 352 Mass. 130, 134 (1967).

In view of the Legislature's goal that promotional examinations fairly test the applicants' fitness to perform the primary or dominant duties of the position sought, as reflected in the statutory scheme, we hold that the judge did not commit error in ruling that the commission was authorized to conduct an evidentiary hearing and make findings of fact in regard to the fairness of the 1987 BPD's lieutenant examination. It is clear that, when challenges to the fairness of promotional examinations reach the commission, the Legislature intended the commission acting in its quasi judicial capacity, and not the administrator, to be the fact finder relative to the fairness of the examination.

On review of the commission's decision, we are concerned only with whether it was "legally tenable and supported by substantial evidence on the record as a whole." Commissioner of Health Hosps. of Boston v. Civil Serv. Commn., 23 Mass. App. Ct. 410, 411 (1987). Substantial evidence is "such evidence as a reasonable mind might accept as adequate to support a conclusion." New Boston Garden Corp. v. Assessors of Boston, 383 Mass. 456, 466 (1981).

There was substantial evidence before the commission to support its finding that the administrator committed error in deciding that the examination was fair without the performance component. Obviously, supervisory skills and abilities represent a significant element of fitness to perform the primary duties of a Boston police lieutenant. The commission heard expert testimony on the matter. The commission is the sole judge of the credibility of the witnesses before it and of the weight to be given to the evidence presented. See, e.g., Martin v. Director of Div. of Employment Security, 347 Mass. 264, 268 (1964). The commission properly found that the multiple choice and training and experience components alone failed to constitute a fair test of supervisory skills and ability.

Finally, the commission's finding that there had been no breach of the security of the in-basket exercise was supported by substantial evidence.

Judgment affirmed.