This is a sex discrimination suit involving Dr. Louis Schwartz, a male university teacher at Florida State University. Dr. Schwartz sued FSU in 1978, complaining that he was being inequitably treated because of his sex by not being allowed to participate in the legislatively mandated salary equity studies being conducted at FSU and that as a result he was being inequitably paid in comparison to comparable female faculty members.

A settlement agreement was reached under which adjustments were made to Dr. Schwartz's 1979-80 base salary, and the University agreed that he could participate in any future salary equity study if men were allowed to participate (i.e., that he could make a claim, with supporting data, that he was entitled to a salary adjustment). He gave a general release, and the suit was dismissed. Later Dr. Schwartz filed this suit, asserting pay disparity based on sex and race for periods after 1979-80. There was no evidence of racial discrimination, and such claims have dropped out of the case. The case involves construction of the settlement agreement and also differing statistical methods for calculating salary to which Dr. Schwartz was entitled after 1979-80. The district court found he had suffered no pay disparity after May 1980 and entered judgment for defendants. We reverse.

Concern at FSU in the late 1960's and early 1970's about equalizing the salaries of female faculty members with those of male faculty members led to two task forces that issued reports finding inequities in the salaries of female faculty members and recommending adjustments to the salaries of a number of women. In addition, in 1976 the Florida legislature enacted a statute requiring universities to conduct annual salary studies to determine which female faculty members were being paid less in comparison with their male counterparts, and to adjust their salaries accordingly. Fla.Stat. 241.735 (1976).1 Dr. Schwartz, a full professor in the Department of Childhood, Reading and Special Education of the College of Education, was denied the opportunity to participate in these studies. He filed a grievance with FSU complaining that the attempt to correct prior disparate treatment between males and females had gone too far and that the cure itself was discriminatory.

After his grievance was denied, Schwartz filed an action in district court against the defendants or their predecessors in office, complaining that he was being inequitably treated because of his sex by not being allowed to participate in the salary equity studies and that as a result he was being inequitably paid. In 1979 the Florida legislature amended the statute discussed above by eliminating its operative effect to females only, requiring that salary studies include men, and that discrimination in the granting of all salaries be eradicated. Fla.Stat. 240.247 (1979). Schwartz then settled his suit. He gave the defendants a general release

of and from all manner of action and actions ... which against them I ever had, now have, or which my heirs ... can, shall or may have ... from the beginning of the world to the day of the date of these presents; pertaining to or related to [the 1978 action] ... and matters raised therein or that could have been raised therein and any matter relating to or arising out of my contractual relationship at Florida State University including, but not limited to, claims of sex discrimination, equal protection, and due process of the law through June 30, 1980.

In return for the release, and a stipulation of dismissal of the action with prejudice, Schwartz received a lump sum payment of $11,000; an agreement that his 1980-81 salary would be computed as if his base salary for 1979-80 had been $24,459 (i.e. a salary raise of $1,240 in his 1979-80 base pay), with specific instructions that it would in no way preclude him from being awarded any other additional pay increases to which he might be entitled for the 1980-81 year; and the unqualified right to participate in any salary equity study conducted after July 1, 1980, if men were allowed to participate in such study, with the proviso that it was recognized that by settling the case all issues arising from his initial employment at FSU through the end of the school year 1979-80 had been resolved to his satisfaction.

Despite the settlement agreement Schwartz was denied the right to participate in the 1980-81 salary equity study on the grounds that proviso language of the General Appropriations Bill and Letter of Intent for the 1979-81 biennium, passed by the legislature during the same session in which it amended the statute, still mandated that the studies be limited to females. In 1981 FSU announced that both male and female faculty members were eligible to participate in the yearly salary equity studies. During each of the years 1981-84 Schwartz prepared and submitted claims for salary adjustments. In addition, in 1982 the Board of Regents disseminated to FSU a regression analysis that used a number of objective criteria to identify persons whose salaries deviated significantly from the predicted salary level. The 1982-83 and 1983-84 regression analyses performed by FSU identified Schwartz as having a salary that deviated significantly from his predicted salary. Schwartz's claims were denied each year by FSU on the ground that the discrepancy between his actual and expected salary was not caused by sexual discrimination.

Schwartz's final complaint alleged violations of various statutory and constitutional provisions prohibiting pay disparity based upon sex. After the submission of several motions by both parties and the filing of a second and third amended complaint by Schwartz, defendants moved for partial summary judgment. The district court granted summary judgment on several issues, leaving Schwartz with the following causes of action against the regents and the state officials in their official capacities: a Title VII claim based on the 1980-81 pay equity study; Equal Pay Act claims for the academic years 1980-81 through 1983-84; and 42 U.S.C. Secs. 1981, 1983, and 1988 claims for those same years.

At trial both sides presented expert statistical testimony. Schwartz's expert, Dr. McClave, did a linear regression analysis of the salaries of seven female full professors in the College of Education at FSU with whom Schwartz had compared himself in his claims for salary adjustments. He found that among the myriad of factors that might explain salary and salary increases at FSU, the three factors that produced the greatest degree of reliability were discipline, rank, and years in rank.

Dr. McClave developed the following statistical formula for comparing the salaries of full professors in the College of Education. He compared the salaries of each of the seven female professors, from which he derived a base salary for each year in question; the base salary is the amount a full professor who has zero to one year in that rank would earn. He developed for each year a multiplier factor representing the value of one year in that rank. The base salary and the multiplier factor changed each year to keep up with inflation.2 To arrive at the salary a professor should have received in a particular year (this is what Dr. McClave refers to as the professor's "projected salary"), Dr. McClave multiplied the value of one year in rank for that year by the number of years the subject had held the rank of full professor and added this to the base salary. Based upon his analysis Dr. McClave testified that Schwartz was paid significantly less than he should have been during the years 1980-85: in 1980-81 he was more than three standard deviations below the salary projected by the formula and was more than six below in 1984-85. According to Dr. McClave's analysis, none of the females was more than two standard deviations from the salary she should have received.

The district court found that in the settlement agreement Schwartz had "agreed that he no longer suffered from any disparity in pay." In light of this the district court found Dr. McClave's formula to have one inherent limitation: because an initial discrepancy in salary will continually widen, even if one receives equal treatment thereafter, the formula would demonstrate an ever-increasing pay disparity. In an effort to correct this the district court adjusted the formula to incorporate an assumption that as of the settlement agreement there was no pay disparity. The district court's formula worked as follows. To determine the amount that Schwartz's salary should have increased from 1980-81 to 1981-82,3 the court subtracted the 1980-81 base salary from the 1981-82 base salary and added to that figure the 1980-81 value of one year in rank. It then added this figure to Schwartz's actual salary for 1980-81 to arrive at the salary Schwartz should have received in 1981-82 to be on a par with his colleagues.4 The same formula was used for each subsequent year by inserting the base salary, value of one year in rank, and actual salary for the year at issue.5

Using the revised formula to calculate the salary Schwartz should have earned, and using Dr. McClave's original formula to calculate the salaries to which the seven female full professors chosen as Schwartz's comparators were entitled, the district court compared the actual salaries of Schwartz and of the comparators to the salaries projected by the formulae. The court found that the amount of variation of Schwartz's actual salary from the formula's projection was well within the accepted range set in calculating the salaries of Schwartz's chosen comparators and that both Schwartz's salary and the salaries of his comparators fell within the accepted range of deviation from the salaries projected for them by the McClave formulae. The district court therefore ruled that Schwartz had proven no disparity in pay and thus had failed to prove that he suffered any discrimination. Accordingly, it ordered judgment for defendants on all claims. After the district court denied Schwartz's motion for a rehearing or new trial, this appeal was filed.

The parties are in agreement as to the legal principles governing the construction of a settlement agreement but disagree as to the proper construction of this settlement agreement. The construction of the settlement agreement is a question of law subject to de novo review by this court. Turner v. Orr, 759 F.2d 817, 821 (11th Cir.1985), cert. denied, --- U.S. ----, 106 S. Ct. 3332, 92 L. Ed. 2d 738 (1986). A settlement agreement is a contract and, as such, its construction and enforcement are governed by principles of Florida's general contract law. Wong v. Bailey, 752 F.2d 619, 621 (11th Cir.1985). Words in a contract are to be given their plain and ordinary meaning, and it is not for the court to add or subtract any language from the face of a clearly worded agreement. Robin v. Sun Oil Co., 548 F.2d 554, 557 (5th Cir.1977); Quesada v. Director, Federal Emergency Management Agency, 577 F. Supp. 695, 697 (S.D.Fla.1983), aff'd on other grounds, 753 F.2d 1011 (11th Cir.), reh'g denied, 761 F.2d 698 (11th Cir.1985), citing Equitable Life Assurance Society of the U.S. v. Pinon, 344 So. 2d 880 (Fla. 3d DCA), cert. denied, 354 So. 2d 984 (Fla.1977), and Emmco Insurance Co. v. Southern Terminal & Transport Co., 333 So. 2d 80 (Fla. 1st DCA 1976), cert. dismissed, 354 So. 2d 986 (Fla.1978). Nor is the court to add to a settlement terms that were not contemplated by the parties. Florida Education Ass'n, Inc. v. Atkinson, 481 F.2d 662, 663 (5th Cir.1973). The court's role is to determine the intention of the parties from the language of the agreement, the apparent objects to be accomplished, other provisions in the agreement that cast light on the question, and the circumstances prevailing at the time of the agreement. J & S Coin Operated Machines, Inc. v. Gottlieb, 362 So. 2d 38, 39 (Fla. 3d DCA 1978).

Schwartz challenges the district court's finding that in his settlement with defendants, Schwartz "agreed he no longer suffered from any disparity in pay." He argues that the agreement was only a compromise to settle past claims and that in no way did he agree that it established salary parity or that, by signing it, he agreed to waive his rank, years in rank or any other benefits of employment. Defendants argue that the language regarding Schwartz's right to receive future pay adjustments was only meant to protect his right to receive any raises he might be entitled to in the 1980-81 year for reasons other than the salary equity study process.

The intent of the parties in entering the settlement agreement was clearly to end the litigation with regard to the issues then in dispute and to preclude any further suits on those issues or any matters related to Schwartz's employment at FSU prior to July 1, 1980. The settlement agreement says nothing, however, about waiving claims of discrimination arising after July 1, 1980. To the contrary it guaranteed Schwartz the right to participate in the salary equity study process--and thereby to complain about salary inequities--beginning in 1980 should men be allowed to participate. It is not for a court to read terms into the settlement agreement that are not explicitly stated, particularly when such terms would be contrary to the purposes of the statute. There can be no prospective waiver of an employee's rights under Title VII, Alexander v. Gardner-Denver Co., 415 U.S. 36, 51, 94 S. Ct. 1011, 1021, 39 L. Ed. 2d 147 (1974), or under the Fair Labor Standards Act, a portion of which is the Equal Pay Act,6 because "this would 'nullify the purposes' of the statute and thwart the legislative policies it was designed to effectuate." Barrantine v. Arkansas-Best Freight System, Inc., 450 U.S. 728, 740, 101 S. Ct. 1437, 1445, 67 L. Ed. 2d 641 (1981) (quoting Brooklyn Savings Bank v. O'Neil, 324 U.S. 697, 707, 65 S. Ct. 895, 902, 89 L. Ed. 1296, reh'g denied, 325 U.S. 893, 65 S. Ct. 1189, 89 L. Ed. 2005 (1945)). Schwartz is correct that the settlement agreement did not constitute a waiver of his future Equal Pay Act or Title VII claims.

There is no specific statement in the settlement agreement that at the time of the agreement Schwartz's salary had been made equal or was equal. The rights guaranteed to Schwartz in the settlement agreement to participate in future salary equity studies and to receive any future pay adjustments are evidence that the settlement agreement did not establish permanent salary parity. Schwartz is entitled to appropriate systems of salary adjustment that are accorded to other employees. The settlement agreement entitled Schwartz to any salary increases he might be entitled to in the 1980-81 year for reasons other than the salary equity process, and he could question whether he had been discriminatorily denied such raises through his participation in the 1980-81 salary equity study. Any salary inequities determined by that study would be remedied through a salary increase in 1981-82. Schwartz did, however, agree to the 1979-80 base salary established in the settlement agreement. In any subsequent claim for equal pay Schwartz must be bound by his agreement to this salary and cannot be heard to complain that it was inequitable. Any other interpretation would be untenable for it would allow the parties to renegotiate the terms of the settlement agreement, and employers will have no incentive to enter settlement agreements. FSU is safe from liability so long as it does not discriminate against Schwartz after the period covered by the settlement agreement.7II.

To make out a prima facie case under the Equal Pay Act, a plaintiff must show "that an employer pays different wages to employees of opposite sexes 'for equal work on jobs the performance of which requires equal skill, effort, and responsibility, and which are performed under similar working conditions.' " Corning Glass Works v. Brennan, 417 U.S. 188, 195, 94 S. Ct. 2223, 2228, 41 L. Ed. 2d 1 (1974) (quoting 29 U.S.C. Sec. 206(d)(1)); see also Brock v. Georgia Southwestern College, 765 F.2d 1026, 1032 (11th Cir.1985). "Once plaintiff has established a prima facie case, the burden shifts to the employer to prove that the difference in pay is justified by one of the four exceptions to the Equal Pay Act: (i) a seniority system; (ii) a merit system; (iii) a system which measures earnings by quantity or quality of production; or (iv) a differential based on any factor other than sex." Brock, 765 F.2d at 1036 (citing 29 U.S.C. Sec. 206(d)(1)). Defendants must show that the factor of sex provided no basis for the wage differential, and the district court's determination on this issue is a finding of fact that will be overturned on appeal only if it is clearly erroneous. Id. We may reverse the district court only when on the entire evidence we are "left with the definite and firm conviction that a mistake has been committed." U.S. v. Gypsum Co., 333 U.S. 364, 395, 68 S. Ct. 525, 542, 92 L. Ed. 746, reh'g denied, 333 U.S. 869, 68 S. Ct. 788, 92 L. Ed. 1147 (1948). We are convinced that such a mistake has been made in this case.

The district court was correct that Schwartz is bound by the salary agreed to in the settlement agreement and therefore reasonably tried to adjust Dr. McClave's formula to take that into account. Because Dr. McClave's formula relied entirely on projected figures, it did not incorporate the fact that Schwartz had agreed to his 1979-80 base salary regardless of whether that salary would have been found equitable when compared to the salary projected by the formula. The court erred however in its development of the revised formula that it applied to Schwartz: it was incorrect in using Schwartz's actual salary for every year as the means of adjusting the formula to take the settlement agreement salary into account.8

The court would have been correct in using Schwartz's actual 1979-80 salary--the settlement agreement salary--in its calculations of the salary that Schwartz should have earned and was entitled to in 1980-81, but its method of computing the salary Schwartz should have earned in each subsequent year by using Schwartz's actual salary for the previous year as the base to which to add the amount that a salary should have increased from one year to the next was clearly erroneous. The revised formula used by the court in effect incorporated an assumption that the actual 1980-81 and 1981-82 salaries were equitable--not something Schwartz agreed to or even could have agreed to in the settlement agreement. After computing the salary that Schwartz should have received in 1980-81 by using the 1979-80 settlement agreement salary,9 the district court should have determined what salary Schwartz should have received in succeeding years based upon what he should have gotten in the prior year rather than upon his actual salary for the prior year. It should have added the amount that Schwartz's salary should have increased from one year to the next to the salary that Schwartz should have earned and was entitled to in the former year.10

Should the district court on remand find a pay disparity, it should also address Schwartz's Title VII and Sec. 1983 claims that were implicitly denied by the district court's holding "that final judgment enter in favor of defendants and that plaintiff take nothing by this action."

REVERSED.