To say that the importance of legal issues in performance appraisal has skyrocketed in recent years would be something of an understatement. When I began my research for this chapter by searching various computer databases, I found almost 500 published judicial and arbitration decisions from just the last several years that involve performance appraisals in one form or another! Many of these decisions turned out merely to contain evidence of favorable performance, offered to show that an individual was qualified for a particular job, and to raise an inference that the reason for refusing to hire, promote, or retain that person must have been discriminatory (see McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973)). However, the sheer number of cases underscores a critical reality for today’s industrial-organizational psychologist or human resource practitioner: it is almost inevitable that one or more elements of your organization’s performance appraisal system will attract legal scrutiny at some point in time. This likely scrutiny is particularly worrisome when considering the potential for jury trials, compensatory and punitive damages, and other burdens imposed in discrimination cases under the Civil Rights Act of 1991.

In this chapter, I offer a foundation for helping to recognize aspects of performance appraisals that are likely to wind up in litigation, and for modifying those that have caused problems for employers in a variety of legal disputes. I begin with an overview of performance appraisals as they relate to the nature of the employment relationship. I then examine specific appraisal processes and potential liability under Title VII of the Civil Rights Act of 1964 (Title VII), the Civil Rights Act of 1991 (CRA 1991), the Age Discrimination in Employment Act (ADEA), the Americans with Disabilities Act (ADA), the Equal Pay Act, and tort theories such as negligence (breach of duty to conduct appraisals with due care), defamation (disclosure of untrue unfavorable performance information that damages the reputation of the employee), and misrepresentation (disclosure of untrue favorable performance information that presents a risk of harm to prospective employers or third parties). Next, I take a closer look at substantive and procedural aspects of performance appraisals and the legal defensibility of employment decisions. I also examine discipline and related performance issues that arise under union and other employment contracts. I conclude with a discussion of emerging legal issues in performance appraisal such as those related to flexible job designs, security and privacy, and workplace violence. Although I do not offer legal advice (knowledgable local counsel should be consulted when specific questions arise), I do provide practical suggestions for the design and implementation of legally sound performance appraisal systems throughout the chapter.

Initial selection tests typically become involved in legal disputes because a desired employment relationship does not emerge as a result of a given test. Performance appraisals, on the other hand, become central to disputes that arise after an employment relationship has been established. Performance and other ratings are used to select present employees for merit pay, promotion, training, retention, transfer, discipline, demotion, or termination. The nature of the employment relationship, as well as the nature of the employment decision, must be considered to determine the potential for performance evaluations to fuel discrimination and other types of lawsuits such as those under the Fair Labor Standards Act (FLSA) and the Family and Medical Leave Act (FMLA).

Table 1 presents a summary of selected laws and legal principles that relate to performance appraisals and the nature of the employment relationship. Further details and examples from the cases are provided below.

Where there are no express contract terms that govern performance appraisals, courts have allowed employers a good deal of latitude to determine how to evaluate their employees (Gomez-Mejia, Balkin, & Cardy, 1995). However, even in an employment at will relationship, under which either the employer or the employee may terminate the employment relationship at any time, an employer’s discretion is not entirely without limits. As Koys, Briggs, and Grenig (1987) have pointed out, courts vary by jurisdiction in their recognition of the at will doctrine as a matter of state contract and labor law. Moreover, courts have found employers liable under numerous exceptions to the doctrine. These exceptions include implied contract (as where an obligation to terminate only for just cause arises based on verbal promises or those contained in an employee handbook) and violation of public policy (as where an employee is terminated after complaining of harassment or accusing the employer of other misconduct). Potential liability for defamation or negligence also may restrict, in practical terms, the manner in which employers can manage and appraise performance. As the following examples illustrate, performance appraisals also may figure in determining the very nature of the employment relationship. This can in turn determine the effect of laws such as the FLSA, the ADEA, Title VII, and the FLMA.

Implied Contracts and Public Policy. Mathewson v. Aloha Airlines, 152 LRRM 2986 (Haw. 1996) provides an interesting example of both the implied contract and public policy exceptions to the at will doctrine. In that case, Mathewson, an Aloha Airlines pilot, disputed his termination, which occurred just 2 weeks

Summary of Selected Laws and Legal Principles That Relate to Performance Appraisals and the Nature of the Employment Relationship

before a one-year at will probationary period would have expired. The termination was supposedly based on poor performance ratings, particularly those in peer evaluations by fellow pilots. However, it turns out that Mathewson had been "blacklisted" by the pilots’ union for having worked as a "scab" for another airline during a union strike. The Hawaii Supreme Court held that the employer had violated both an implied contract to provide fair, unbiased evaluations (based on parts of the company’s Flight Operations Manual), and public policy against discrimination based on non-membership in a labor organization (a right protected under state and federal labor law).

In another case that illustrates the importance of state law in this area, a California appellate court found it necessary to overturn a lower court ruling on the at will issue in Haycock v. Hughes Aircraft, 10 IER Cases 1612 (Cal. App. 1994). Even though California labor law creates a presumption of at will employment, that presumption can be overcome by an implied agreement not to terminate except for just cause. Evidence suggested that, because of such an agreement, Hughes had fraudulently altered favorable evaluations of a 25-year employee, without his knowledge, to justify a "for cause" termination rather than a layoff under which he would have been protected by the company’s appraisal review mechanisms and seniority preferences. For an analogous case in the whistleblowing context, see Roberton v. Alabama Department of Economic and Community Affairs, 4 AD Cases 1749 (D. Ala. 1995), which upheld, on public policy grounds, Robertson’s claim that the employer had lowered her appraisal results in retaliation for supporting a co-worker’s sex discrimination complaint, and for accusing the employer of misapplying federal funds.

Performance Appraisals and the Existence of an Employment Relationship. The very fact that performance is evaluated can influence the determination that an employment relationship exists, as well the nature of that relationship. In EEOC v. Johnson & Higgins, 71 FEP Cases 818 (2d. Cir. 1996), Johnson & Higgins (J&H) appealed from a lower court ruling that its mandatory retirement policy for directors reaching age 62 violated the ADEA. J&H argued that the ADEA did not apply because their directors were not "employees" within the meaning of the Act, but instead maintained a "fundamentally entrepreneurial relationship" similar to a partnership. The Court of Appeals held in favor of the Equal Employment Opportunity Commission (EEOC), however, finding an employment relationship (and thus an ADEA violation) based on the fact that J&H’s senior Board members reviewed directors’ performance annually, and the results determined directors’ compensation to a great extent.

In a somewhat different context, the fact that appraisals were conducted led the court in Sturm v. TOC Retail Inc., 2 WH Cases 2d 628 (D.C. Ga 1994), to dismiss Sturm’s claim for overtime pay under the FLSA. Because Sturm, a convenience store employee, was responsible for evaluating the performance of two or more employees in order to determine their raises and continued employment, the Court found that Sturm’s duties were primarily managerial, and that he therefore was exempt from the overtime provisions of the Act.

Fair performance evaluations have been held to be tangible job benefits, and interference with those benefits may give rise to liability for harassment under Title VII (see, e.g., Ton v. Information Resources Inc., 70 FEP Cases 355 (N.D. Ill. 1996), in which Ton was issued an "indefensibly harsh" performance evaluation in retaliation for rejecting his manager’s sexual advances). However, an unfavorable review, without more, is not actionable. In Smart v. Ball State University, 71 FEP Cases 495, 498 (7th. Cir. 1996), Smart, a tree surgeon trainee, filed an EEOC complaint against the University claiming gender discrimination under Title VII. She later claimed that the University had retaliated against her for that complaint by applying more formal and stringent evaluation procedures, which resulted in less favorable reviews. Nevertheless, Smart successfully completed her training, and got a job with the University as a full-fledged tree surgeon. In upholding dismissal of Smart’s lawsuit, the Court of Appeals noted that Smart had failed to establish a causal connection between her negative evaluations and any sort of adverse employment action. The Court acknowledged that "adverse employment action" has been defined broadly, and is not limited to tangible losses such as discharge or reduction in pay. However, the Court explained further that "not everything that makes an employee unhappy is an actionable adverse action. Otherwise, minor and even trivial employment actions ... could form the basis of a discrimination suit." This language illustrates that most courts are reluctant to pass judgment on the general fairness of an organization’s employment practices, unless discrimination or some other improper outcome can be shown (see generally Barrett & Kernan, 1987).

Issues involving performance appraisals and the employment relationship also have arisen under the FMLA. In Patterson v. Alltel Information Services, 3 WH Cases 2d. 406 (D. Maine 1996), Patterson claimed that Alltel violated the FMLA when, after returning from medical leave for work-related stress, he was relegated to "special projects," and received unfavorable evaluations that eventually led to his layoff as part of a company-wide reduction in force (RIF). The Court, however, dismissed Patterson’s claim that he had suffered a post-leave change in his "position of employment" (an FMLA violation) that led to poor performance and his eventual layoff. Alltel was able to establish that Patterson’s performance had been rated poorly prior to any change in position, and that he would have been laid off even had he not gone on leave.

Practical Suggestions for Limiting Performance Appraisals’ Unwanted Effects on the Employment Relationship. As discussed in greater detail later in this chapter, timely and consistent documentation of performance information in accordance with established practices can be critical to the successful defense of a variety of legal disputes. However, as the cases digested above demonstrate, performance appraisals may be construed by courts to affect the nature and existence of employment relationships in ways that were neither contemplated nor desired by the employer. In order to limit the extent to which this can occur, employers should consider drafting job offers, employee handbooks, and other employment documents to include express at will language that refers explicitly to performance appraisals (see Table 2).

This language might state, for example, that employment is understood to be at will, that the employer expressly reserves the right to discharge the employee at any time for any reason with or without cause and with or without notice, and that nothing in the employer’s policies, practices, or procedures, including performance appraisals, should be construed to confer any right upon the employee to continued employment. Management also could expressly reserve the right to unilaterally alter the terms and conditions of employment, including the manner in which performance may or may not be appraised.

Further, unless limited by explicit contract language to the contrary, employment documents should make it understood that the employer is under no obligation to appraise performance, and that neither the fact that appraisals are or are not conducted, nor the manner in which they may be conducted, should be construed to give rise to a "just cause" requirement for terminating the employment relationship. In summary, employment documentation, signed by the employee, should set forth clearly the express intention and understanding of both the employer and the employee that performance appraisals and other

______________________________________________________________________________

Consider drafting job offers, employee handbooks, and other employment documents, for signature by the employee, to include express at will language that refers explicitly to performance appraisals. Such language might state:

(1) that employment is understood to be at will;

(2) that the employer expressly reserves the right to discharge the employee at any time for any reason with or without cause and with or without notice;

(3) that nothing in the employer’s policies, practices, or procedures, including performance appraisals, should be construed to confer any right upon the employee to continued employment;

(4) that the employer expressly reserves the right to unilaterally alter the terms and conditions of employment, including the manner in which performance is or is not appraised;

(5) that the employer is under no obligation to appraise performance;

(6) that neither the fact that appraisals are or are not conducted, nor the manner in which they may be conducted, should be construed to give rise to a "just cause" requirement for terminating the employment relationship;

(7) that performance appraisals and other evaluation procedures should in no way be considered in any other manner in determining the existence or nature of any employment relationship that may be found to exist between the parties.

______________________________________________________________________________

evaluation procedures should in no way be considered in determining the existence or nature of any employment relationship that may be found to exist between the parties. Such language should, at a minimum, make it more difficult for disgruntled current and former employees to rely on performance appraisals and related procedures to the detriment of the employer.

Both the theory used to prove discrimination (disparate treatment or disparate impact), and the type of employment decision (e.g., compensation, promotion, layoff, or discharge), are likely to determine when and how performance appraisals will figure in discrimination cases. Previous reviews (Burchett & De Meuse, 1985; Martin & Bartol, 1991; Martin, Bartol, & Levine, 1986; Veglahn, 1993) digest a good deal of earlier caselaw and arbitration decisions that address these issues. I therefore focus on recent developments.

Table 3 presents a summary of selected laws and legal principles that can relate to performance appraisals and potential liability for employment discrimination. For those who may not be familiar with this area of the law, I provide a more detailed summary of discrimination analysis under disparate treatment and disparate impact theories, which may be applied to any of the substantive laws--e.g., Title VII, ADEA, ADA--under which employers are often accused of discrimination. I also discuss briefly how the employer’s defense strategy will differ with the nature of the employment decision in question. Examples from the cases follow.

Disparate Treatment. Most discrimination cases that involve performance appraisal allege disparate treatment. In these cases, employees claim that they were intentionally treated differently because of their gender, race, ethnic background, national origin, age, disability, or other status protected under State or Federal law. An employee can establish a prima facie case

(raise an inference) of discrimination either by presenting direct evidence of discrimination (e.g., racist or sexist remarks that appear to have influenced

the employment decision), or by showing that he or she was qualified and available for a position, but was rejected under circumstances that suggest unlawful discrimination (Texas Department of Community Affairs v. Burdine, 450 U.S. 248 (1981)). This most commonly is done according to McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973), by showing that the employee (1) belongs to a protected class; (2) applied for and was qualified for a job opening; (3) was rejected; and (4) the job remained open while the employer continued to seek similarly qualified applicants or hired an individual who is not a member of the same protected class. Under the now well-accepted burden shifting process of McDonnell Douglas, the employer then must articulate a legitimate, non-discriminatory reason (typically a performance-related one) for its action. If the employer can do so, the employee then must show that the employer’s stated reason was a "pretext" for a discriminatory decision. In so-called "mixed motive" cases (those where decisions are based partly on discriminatory motives but also on performance), an employer still can avoid liability for compensatory and punitive damages, reinstatement, back pay, and similar relief otherwise available under substantive law, if it can convince the fact-finder that it would have made the same decision even had it not taken improper factors (e.g., gender) into account (Price Waterhouse v. Hopkins, 109 S.Ct. 1775 (1989); CRA 1991, section 107(a)). This defense is not available, however, in cases where protected class status (e.g., age) is clearly used as a limiting criterion in the employment decision (see, e.g., EEOC v. Johnson & Higgins, where J&H unsuccessfully tried to defend its age-based mandatory retirement policy by claiming that it enabled the company to plan its succession in an orderly fashion).

Disparate (Adverse) Impact. Performance appraisals figure less prominently in disparate impact cases, in which a seemingly neutral employment practice may have an unintentional but nonetheless discriminatory effect. In such cases, employees must demonstrate a causal connection between a specific employment practice (e.g., performance appraisals) and a discriminatory result, unless the elements of the employer’s decision making process are not capable of separation for purposes of analysis (Civil Rights Act of 1991; Wards Cove Packing Co. v. Atonio, 490 U.S. 642 (1989)). Appraisal results then can be used to rebut plaintiffs’ (usually statistical) evidence of an improper disparity in promotion, layoff, or other employment decisions (EEOC v. Texas Instruments, Inc., 72 FEP Cases 980 (5th. Cir. 1996); Griggs v. Duke Power Co., 401 U.S. 424 (1971); Watson v. Fort Worth Bank & Trust Co., 108 S.Ct. 2777 (1988)). Here, the burden shifting process parallels that of disparate treatment cases; the employer must show that the challenged practice bears a "manifest relationship" to job performance consistent with "business necessity." A common articulation of this burden is to show that the challenged practice "is significantly related to the legitimate goals of the employer," Wards Cove, 490 U.S. at 659. (The Wards Cove case, while subject to attack during the legislative process, was not expressly overturned by CRA 1991, and is still considered by leading commentators to provide viable guidance in this context; see Cathcart & Snyderman, 1997). The employee then may establish pretext if he or she can show that other appraisal practices would have served the employer’s interests without such a discriminatory effect (Albemarle Paper Co. v. Moody, 422 U.S. 405 (1975).

Nature of the Employment Decision. The employer’s defense strategy typically will differ depending on the nature of the employment decision (Martin and Bartol, 1991). In failure-to-promote cases, employers use appraisals to show that the person selected for the promotion is likely to perform in the higher position better than the plaintiff. In layoff or discharge cases, employers use appraisals to show that the plaintiff was already performing poorly in comparison either to peers (layoff cases) or to some minimum standard of acceptable performance (discharge cases). In merit pay cases, employers compare the plaintiff’s performance with specific compensation criteria.

I now present examples of recent cases that involve Title VII, the Equal Pay Act, the ADEA, the ADA, and other theories of recovery. Because of the rapidly growing number of cases that involve performance appraisals, I only had room to include a limited number of cases. I therefore decided to offer a mix of appellate, trial court, and arbitration decisions to provide the reader with a feel for both the legal and factual issues that tend to be in dispute. I omitted cases in which favorable appraisal results were used solely to establish a prima facie case, because these usually were not very instructive for practitioners. Instead, I focused on cases where the employer’s defense relied on poor performance appraisals to demonstrate a legitimate, non-discriminatory reason for its actions, or that it would have taken the same action irrespective of possible discrimination in a mixed motive case. I also included cases where favorable appraisals were used to show pretext, which is often the determinative issue in contemporary discrimination litigation.

Title VII of the Civil Rights Act of 1964 makes it unlawful to discriminate against any individual with respect to compensation, terms, conditions, or privileges of employment, because of race, color, religion, sex, or national origin. State Fair Employment Practices Acts, which vary by jurisdiction, contain similar prohibitions that provide at least as much protection as federal law, and sometimes carry procedural or strategic advantages for plaintiffs. The Equal Pay Act provides protection against wage discrimination on the basis of gender "for equal work on jobs the performance of which requires equal skill, effort, and responsibility, and which are performed under similar working conditions," 29 U.S.C. Section 206 (d)(1). Violations of the Act are inferred if the employer pays different wages to an employee of the opposite sex for substantially equal work. The employer then may defend by showing that any wage discrepancy resulted from (1) a bona fide seniority system; (2) performance or merit-based distinctions; (3) piece-rate or other production-based systems; or (4) some legitimate factor other than gender (Corning Glass Works v. Brennan, 417 U.S. 188 (1974)).

Several recent cases in this area involve professionals such as college professors, attorneys, and engineers. In Fisher v. Vassar College, 70 FEP Cases 1155 (2d. Cir. 1995), Fisher, a biology professor, was denied tenure at Vassar and claimed discrimination based on sex, marital status, age discrimination, and violation of the Equal Pay Act. Vassar based its decision on the biology department’s recommendation, outside evaluations, and student comments, all of which were reviewed by the college dean and a college-wide faculty committee. The department’s evaluation focused on scholarship, service, teaching ability, and leadership (all of which were found wanting), but also noted Fisher’s 8-to-9-year hiatus to raise a family before returning to work and coming up for tenure review. After trial in a factually complex case, the District Court concluded that Fisher had met the recognized qualifications for tenure, that the department’s evaluations were biased and pretextual, and awarded her more than $600,000 in damages.

On appeal, a panel of the Second Circuit reversed. The Court agreed that Fisher had established a prima facie case of discrimination, and that Vassar had articulated a legitimate, non-discriminatory reason for denying her tenure (the department’s negative evaluation of Fisher’s performance). It also seemed inclined to agree that Vassar’s tenure review was pretextual, in that its tenure standards were unclear and unspecified, and the tenure committee had selectively excluded favorable information about Fisher’s performance. However, citing St. Mary’s Honor Ctr. v. Hicks, 113 S.Ct. 2742, 2752 (1993), the Court pointed out that it is not enough merely to show pretext; the plaintiff must prove that the employer’s articulated reason is a pretext for discrimination. The Court found the evidence insufficient to establish that Vassar’s denial of tenure was discriminatory. It also rejected Fisher’s Equal Pay Act claim because the higher paid male to whom Fisher compared herself had been on the tenure track several years longer than Fisher, which established a legitimate differential based on factors other than gender.

Jiminez v. Mary Washington College, 67 FEP Cases 1867 (4th. Cir. 1995), is another case that illustrates judicial reluctance to second-guess the merits of promotion decisions, particularly when the performance being evaluated lies within the decision makers’ substantive expertise. In Jiminez, the District Court found that the College had discriminated in denying Jiminez tenure based on race (black) and national origin (Trinidad). As in Fisher, tenure decisions were based on evaluations of service, scholarship, and teaching effectiveness. Although Jiminez’ service was appraised highly, he had failed to publish adequately or complete his doctoral dissertation in 6 years with the College. He also had received consistently low ratings from students. However, Jiminez argued that these ratings were due to a concerted effort by a group of white students to have him removed. The Court found that Jiminez had established a prima facie case of discrimination, and that the College had rebutted it (based on poor performance), but that Jiminez had shown pretext (the College’s failure to disregard "racially tainted" teaching evaluations).

A panel of the Court of Appeals reversed. Noting that courts typically "review professional employment decisions with great trepidation," the Court determined (as in Fisher) that, even had the College’s reliance on factors such as "untainted" evaluations and Jiminez’ failure to defend his dissertation been pretextual, he still had failed to establish discrimination.

Byrd v. Ronayne, 68 FEP Cases 769 (1st. Cir. 1995) illustrates a similar analysis in the law firm context. Byrd, a law firm associate, was terminated after client complaints about the quality of her work. She sued, accusing the firm of Title VII sex discrimination, an Equal Pay Act violation, and retaliatory discharge based on her previous filing of an EEOC complaint. Byrd had received bonuses and favorable reviews notwithstanding some acknowledged performance difficulties, but the District Court found that she had failed to establish a prima facie case under McDonnell Douglas, because her documented poor performance indicated she was not qualified for her job.

The Court of Appeals agreed, but went further. It ruled that, even had Byrd been able to make out a prima facie case, she had failed to raise sufficient evidence of pretext (via the bonuses and favorable reviews) because her performance clearly was poor at the time of her discharge. The Court also upheld dismissal of Byrd’s Equal Pay Act claim, because she failed to show that her duties were similar to a more highly paid male associate, or that this associate (who also generated more billings) had a similar record of client complaints and performance problems.

For an analogous case in which the employer also prevailed, see EEOC v. Lousiana Network, 1 WH Cases 2d 435 (D. La 1992). In that case, the court found that a radio station did not violate Title VII or the Equal Pay Act when it paid a female news anchor less than a male anchor who had superior experience and on-the-air presentation skills, as documented in written performance appraisals. A note of caution is in order here, however. Although the employers prevailed in these cases, the facts underlying the pretext claim in Byrd illustrate the danger of trying to be "nice" to a poor performer by continuing to give good evaluations in the face of known performance problems. This issue is discussed in greater detail in later portions of this chapter.

Amirmokri v. Baltimore Gas & Electric Co., 68 FEP Cases 809 (4th. Cir. 1995), presents another situation where the employer successfully defended a Title VII claim, here for failure to promote based on national origin. In this case, the employer relied on evidence that the promoted employee had performance evaluations superior to Amirmokri’s. Although Amirmokri, an engineer of Iranian descent, made out a prima facie case under McDonnell Douglas, the employer was able to show that the person promoted was more qualified, based on objective performance evaluations and subjective factors such as interpersonal skills and ability to lead a team. The District Court also rejected Amirmokri’s pretext claim, and dismissed his discrimination case altogether. However, based on evidence that Amirmokri was called names including "the ayatollah," "the local terrorist," and "camel jockey," the Court of Appeals allowed the case to go to trial on the issue of hostile environment harassment (EEOC and Supreme Court rulings have interpreted Title VII to prohibit harassment based on national origin as well as gender and other protected status). The Court also ordered a trial on Amirmokri’s claim for constructive discharge, which alleged that the hostility of the working environment was so intolerable that a reasonable person would have felt compelled to leave their job. The Court focused on evidence that Amirmokri’s supervisor had intentionally embarrassed him by assigning him impossible tasks and telling co-workers that he was incompetent, which may have negatively affected both his performance and its evaluation. The case illustrates that defensible performance appraisals alone may not ensure success in court, because conduct related to the evaluation process can give rise to liability under more than one legal theory.

Kerr-Selgas v. American Airlines, 69 FEP Cases 944 (1st. Cir. 1995) is another Title VII harassment case. In this one, a male supervisor gave his female employee a negative performance evaluation shortly after she rejected his sexual advances. Although the case deals primarily with procedural issues unrelated to performance appraisals, the employer’s failure to take proper and timely action against the supervisor led a jury to award $1.2 million for discrimination, retaliation, and invasion of privacy (see also Ton v. Information Resources, a same-sex harassment case in which the plaintiff’s performance was rated poorly after he refused his manager’s sexual advances).

Finally, affirmative action policies were at the heart of a white male’s complaint of reverse discrimination in Whalen v. Rubin, 71 FEP Cases 1170, 1174 (7th. Cir. 1996). Whalen, an IRS appeals officer who was both a lawyer and a CPA, was passed over for a promotion, which went to a woman named Price. Price was ranked as the most qualified among seven candidates, while Whalen tied for last on the list. An IRS panel of interviewers unanimously chose Price as the best person for the job, based on her substantial experience as an acting supervisor and her completion of a pre-management training course. Whalen countered that Price’s selection was tainted, because the IRS maintained an affirmative action plan and provided bonuses to employees based on their "success" with equal employment opportunity. The Court of Appeals rejected this argument, and held that Whalen had failed to establish a direct causal link between the IRS’ affirmative action policy and his inferior performance ratings: "the mere existence of an affirmative action policy is ... insufficient to prove that the IRS actually intentionally discriminated against Whalen." For an analogous holding under New Jersey state law, see Wachstein v. Slocum, 67 FEP Cases 587 (N.J. App. 1993).

As the average age of the American worker continues to increase, so does the number of age discrimination lawsuits, including those claiming age-related biases in performance appraisals (see Miller, Kaspin, & Schuster, 1990; Perry, Kulik, & Bourhis, 1996). The ADEA as originally passed protected employees between the ages of 40 and 65 against discrimination "with respect to compensation, terms, conditions, or privileges of employment," or classification that would "tend to deprive any individual of employment opportunities or otherwise adversely affect [their] status as an employee," 29 U.S.C. sections 623(a)(1)-(2). The Act was later amended to raise the upper limit to 70, and amended again (in 1986) to remove the upper limit entirely. Although employers have been able to defend quite a few of these cases successfully, employees continue to succeed where they can show that performance evaluations were manipulated to justify age-based discrimination.

For example, in Starceski v. Westinghouse Electric Corp., 67 FEP Cases 1184 (3rd. Cir. 1995), the Court of Appeals affirmed a jury verdict for Starceski, a 63 year old senior engineer and 36-year Westinghouse employee, based on evidence that a supervisor had directed another employee to "doctor" Starceski’s evaluations to reflect poor performance. Starceski was able to show that managers were told to transfer jobs away from older workers in anticipation of an RIF, that 5 of 6 workers laid off were over 40, that the average age of the work group after these layoffs was under 40, and that several of those who were not laid off had lower performance evaluations than he did. Under these circumstances, the Court rejected Westinghouse’s claim that it would have laid Starceski off even had it not considered his age.

A similar result often occurs where a long history of good performance is suddenly followed by the bad evaluation of an age-protected worker just before a layoff. In Woodman v. Haemonetics Corp., 67 FEP Cases 838, 840 (1st. Cir. 1995) the Appellate Court reversed dismissal of Woodman’s ADEA claim, in part because of a supervisor’s statement that "these damn people [management]--they want younger people here." The Court found that this statement and other evidence warranted a trial on the issue of pretext and discriminatory intent. See also Brewer v. Quaker State Refining Corp., 69 FEP Cases 753, 759 (3rd. Cir. 1995), in which Brewer’s receipt of a bonus shortly before his discharge, and a performance memorandum stating that plaintiff "is 53 years old, which presents another problem," amounted to circumstantial evidence of age discrimination that required a trial on the issue of the "corporate culture in which the employment decision to discharge Brewer was made."

The employer fared better in Thomas v. IBM, 67 FEP Cases 270 (10th. Cir. 1995). In that case, Thomas’ ADEA and fraudulent evaluation claims were dismissed by the District Court, a decision upheld on appeal. Thomas claimed, but could not demonstrate, that IBM gave her undeservedly low performance evaluations because of her age to induce her to accept early retirement or an extended leave of absence. The case contains an interesting discussion of IBM’s performance ranking system, through which employees ranked low in terms of their "relative contribution to IBM’s business" find their positions at risk. In upholding dismissal of Thomas’ claims, the Court noted that IBM’s appraisal system appeared to contain safeguards designed to minimize possible bias (these aspects of appraisals are discussed later on).

EEOC v. Texas Instruments, Inc., 72 FEP Cases 980 (5th. Cir. 1996) presents a case in which the employer found itself in the unusual position of having to argue that its appraisal system was essentially worthless for evaluating performance in determining who to lay off during an RIF. The Court ruled in favor of Texas Instruments (TI), notwithstanding evidence of age-related comments and disregard of its own seniority system and appraisal results. Unlike Woodman v. Haemonetics Corp., the Court in this case found remarks to the effect that "his age got him" and that TI "had to make room for some of the younger supervisors" to be "stray remarks" that were insufficient to rebut TI’s stated reasons for discharging six supervisors who were in their fifties. In an age of rapidly changing technology, the Court accepted as non-pretextual TI’s claim that it disregarded both seniority preferences and favorable appraisals because to adhere to their use would have caused TI to retain highly paid supervisors whose willingness or ability to learn new technology was questionable. TI’s evaluations and "Key Personnel Assessment" rankings apparently were not designed to make fine distinctions among employees; the former clustered ratings tightly around group medians, whereas the latter merely reflected departmental pay differentials among employees.

The Court also accepted TI’s argument that favorable performance ratings on present jobs were not necessarily relevant to determining ability to perform post-RIF duties that required mastery of sophisticated computer software and other new resources in a rapidly changing technological environment. However, it is not clear that these arguments would prevail in all jurisdictions. Practitioners would be well advised to keep job descriptions, performance criteria, and appraisal processes updated to the greatest extent practicably possible.

Finally, Fisher v. Vassar College, discussed earlier in the Title VII context, also contained an ADEA claim that was dismissed without trial. Despite evidence that eight other tenured faculty at Vassar were at least 9 years younger than Fisher when they were tenured, and a colleague’s statement that Fisher "was too old to ever become tenured," the Court of Appeals upheld dismissal of her case because the sample of younger tenured faculty was too small to permit meaningful inferences about discrimination, and because Fisher’s extended hiatus from academia prior to entering the tenure track explained any age discrepancy without the need to infer discrimination.

The ADA prohibits an employer from discriminating in employment, based on a known physical or mental impairment, against a qualified individual with a disability. A "qualified individual with a disability" is defined as an individual substantially limited in one or more major life activities who, with or without a reasonable accommodation, can perform the essential functions of the position that the individual desires or holds; 42 U.S.C. Section 12111; 29 C.F.R. Section 1630.2(m). The ADA also prohibits discrimination based on a record of impairment or perception of impairment (so-called "regarded as" cases). The Rehabilitation Act contains similar proscriptions for employers with federal funding or federal contracts, and tends to be used primarily in cases where the alleged discrimination took place before the effective date (July, 1992) of the ADA.

Growing recognition that a wide variety of conditions may be construed as disabilities (particularly in the area of stress-related and other mental impairments) has led to a rapid increase in the number of claims filed under these statutes, and perhaps to a judicial reluctance to expand the scope of liability in the courts. In the context of performance appraisals, these cases usually turn on whether the employer knew or had reason to know of the employee’s disability, whether the employee was qualified, or whether the employee was evaluated on something other than the job’s essential fuctions.

For example, in Hedberg v. Indiana Bell Telephone Co., 4 AD Cases 65, 69 (7th. Cir. 1995), the employer was undergoing an RIF based on relative performance rankings. It terminated Hedberg in part based on low rating scores on a written evaluation form, and in part based on appraisal comments about problems with interpersonal skills, not coming to work on time, and his "work ethic." Unknown to the company, Hedberg suffered from a condition called primary amyloidosis, a disease that can cause chronic fatigue symptoms. The District Court dismissed Hedberg’s ADA claim because it concluded that the disease was not a "known disability" within the meaning of the Act. The Court of Appeals affirmed, noting that, unlike most discrimination claims in which the protected characteristic of the plaintiff (e.g., race or gender) is obvious to the employer, there are situations in which an employer clearly does not know about an employee’s disability. The Court reasoned that an employer cannot fire an employee "because of" a disability of which it has no knowledge. Here, the employer maintained that it fired Hedberg primarily based on low written evaluations, but the Court offered that, even had they fired him for perceived lack of "work ethic," they would not have violated the ADA:

For a case in which the employer did use negative appraisals to show that the employee was not qualified for her job, see Demming v. Housing Authority, 4 AD Cases 1593 (8th. Cir. 1995), in which Demming claimed that her discharge occurred because she had been hospitalized for thyroid cancer, but could not establish that performance factors unrelated to her disability were not the real reason for her discharge.

Qualification for a job, essential functions of the job, and employer knowledge of the employee’s disability all were involved in Lewis v. Zilog, Inc., 4 AD Cases 1787 (N.D. Ga. 1995). In that case, Lewis was diagnosed with bipolar mood disorder, a condition she claimed resulted in part from a stressful meeting about a written performance evaluation that cited her need to improve relationships with others. Lewis had a history of yelling at coworkers and exhibiting attitude and mood swings. During the meeting, she jumped up and ran around the office, blamed others for various problems, and sliced a photo of her supervisor into small pieces. Lewis later claimed to be totally disabled, took the second of two medical leaves, filed for long-term disability benefits, and was terminated for exceeding the allowable maximum leave under company policy. She sued Zilog under Title VII and the ADA.

Zilog moved to dismiss Lewis’ ADA claims, conceding that Lewis had a disability within the meaning of the Act, but denying that it had been aware of any disability during Lewis’ employment, and denying that she was a "qualified individual with a disability" for ADA purposes. The Court granted Zilog’s motion, ruling that her benefit-related assertions that she was unable to work proved that she could not perform the essential functions of her job (Lewis’ psychiatrist was unable to suggest a reasonable accommodation that would have allowed her to return to work). The Court also found that Zilog’s termination of Lewis in accordance with its own medical leave policy was a legitimate non-discriminatory reason, and rejected any claim of pretext. Zilog had terminated other individuals in accordance with its policy, and the Court found no evidence that Zilog knew Lewis had bipolar mood disorder even though coworkers were aware that she suffered from stress and took the drug Prozac.

Olson v. General Electric Astrospace, 6 AD Cases 270, 275 (3rd. Cir. 1996) further illustrates how evidence offered to prove one element of an ADA claim (e.g., "qualified individual") can be used to negate a different element (e.g., "individual with a disability"). Olson had been laid off during an RIF after a history of hospitalization for depression and a possible sleep disorder. Olson’s supervisor, in a written performance evaluation prior to Olson’s layoff, questioned his work habits and commitment to his department because of "an unusual amount of time off for personal illness reasons." Olson later was interviewed for rehire by the same supervisor, who recommended a different candidate to a superior that was responsible for making the actual hire decision. This superior accepted the recommendation to hire the other candidate, and Olson sued, claiming an ADA violation.

Olson’s ADA claim initially was dismissed by the District Court, because evidence that he was able to perform his duties despite any psychological problems (i.e., that he was qualified for the job) "ironically establish[ed] that he was not substantially limited in a major life activity." The Court of Appeals agreed with this analysis as far as it went, but ordered a trial on the issue of whether Olson had been "regarded as" having a disability. Even though the hiring decision was made by a person who claimed to have no knowledge of Olson’s medical history, the fact that this person’s decision was influenced by the recommendation of a supervisor who did have such knowledge raised a factual issue that needed a trial to resolve. This case suggests that supervisors should be better trained to recognize potential "regarded as" liability, and that review mechanisms and other safeguards should be installed, to ensure that potentially tainted recommendations are not involved in hiring, promotion, and related employment decisions.

Two recent cases involving Human Immunodeficiency Virus (HIV) further illustrate the nuances of ADA actions where the disability is not obvious or clearly known to the employer. In R.G.H. v. Abbott Laboratories, 4 AD Cases 289 (N.D. Ill. 1995), R.G.H., a research biochemist, tested positive for HIV after accidental exposure to the virus in one of Abbott’s labs. After a number of favorable appraisals and raises, he was turned down for two promotions, the first because a better qualified applicant was selected, and the second because of deficient performance evaluations. R.G.H. alleged an ADA violation, but the Abbott employees who made these decisions established that they were ignorant of his HIV status, and he was unable to present evidence as to the role played by any perceived or actual disability in either outcome.

The HIV-infected plaintiff in Runnenbaum v. Nationsbank of Maryland, 5 AD Cases 1602 (4th. Cir. 1996), faced a similar problem (dismissal) at the District Court level, and appealed. A panel of the 4th Circuit, in a split decision, reversed and remanded for trial. The majority found accusations that Runnenbaum failed to complete certain "training tasks" or present a "professional image" to be potentially pretextual, because the bulk of his performance appraisals were positive or outstanding. The dissent seemed unwilling to interfere with the professional judgment of the employer.

Finally, Borkowski v. Valley Central School District, 4 AD Cases 1264 (2d. Cir. 1995), involved a number of performance-related ADA issues, including reasonable accommodation and essential functions, in the context of a teacher tenure denial. Borkowski, an elementary school teacher, had sustained serious head injuries in a motor vehicle accident. Although she recovered substantially, she continued to experience difficulties with memory, concentration, and dealing with multiple stimuli simultaneously. These problems led to difficulties controlling students in her classes, as noted by an observer during an unannounced classroom visit. Overall, Borkowski received mixed performance reviews, and was denied tenure. She unsuccessfully sought reconsideration due to her disability, and sued under the Rehabilitation Act.

On appeal, while acknowledging the School District’s discretionary authority to make tenure decisions, the Court ruled that the District had acted improperly in refusing to consider whether Borkowski could have performed the essential functions of her position with a reasonable accommodation, such as providing a teaching assistant to help her control her classes (the Court also questioned whether classroom control really was an essential function of the job). The Court further noted that Section 504 of the ADA requires an employer to ensure that its evaluative techniques measure the job-related skills of an individual with a disability, not the disability itself. This raises a further issue as to whether the practice of using unannounced classroom visits to evaluate Borkowski’s performance might have been an ADA violation in and of itself.

The Court’s language in Borkowski makes clear that employers need to make reasonable accommodations not only with respect to how jobs are to be performed, but also with respect to the manner in which performance is to be evaluated. Employers must take care to ensure that only the essential functions of the job are evaluated, and that only performance is used to make employment-related decisions. In addition, although the ADA prohibits discrimination on the basis of "known" disabilities, and it is generally the responsibility of the employee to request a reasonable accommodation, there is a trend toward requiring the employer to initiate an interactive process with the employee to provide an accommodation even without such a request (Fram, 1997). It therefore would seem beneficial to establish mechanisms through which disabled individuals can make the employer aware of their disability on a confidential basis. Such steps might help to avoid potential invasions of privacy or liability for discrimination based on perception of a disability.

Taking such steps also might increase both the chances of indentifying problematic situations proactively, and of defending discrimination claims in court with a showing of good faith efforts to comply with the requirements of the ADA. For example, the employer might post a notice setting forth its intention, in accordance with "federal law", to make necessary "modifications" to break down workplace barriers to effective job performance, and that employees who know of any such barriers should identify them in confidence to a designated individual. Using this sort of language would help to avoid drawing attention to the employee’s impairment and the disability-based nature of the notice. Supervisors also should be trained to identify, during appraisal processes and otherwise, employees whose performance suggests that reasonable accommodation of a physical or mental impairment might be in order. Supervisors then should explore with the employee on an interactive basis the need for and nature of the accommodation.

Performance evaluation processes also can cause problems for employers under state tort law theories such as infliction of emotional distress, negligence, defamation, or misrepresentation. Emotional distress cases usually involve fact-specific disputes over the outrageousness of conduct such as harassment, and are not discussed further here. Negligence cases have declined in frequency, so I address them only briefly. Defamation actions have become more common, and misrepresentation is a relatively new player in the appraisal context. Taken together, these types of cases underscore the need to perform appraisals and communicate their results with a great deal of care.

Negligence. Negligence claims require a showing that the employer owed a duty to conduct performance appraisals with due care, and that this duty was violated. Because of the employment at will doctrine, such a duty, where it exists, is usually found in some form of employment contract (see, e.g., Mathewson v. Aloha Airlines). Interestingly, cases in this area may involve complaints about favorable evaluations, as where an employer fails to notify an employee through the appraisal process that performance is inadequate, and terminates the employee without providing an opportunity to improve (e.g., Mann v. J.E. Baker Co., 52 FEP Cases 1111 (D. Pa. 1990). Liability also may arise where the employer communicates favorable but untrue information about a former employee’s performance, and causes a substantial risk of harm to others by doing so (see Randi W., a Minor, v. Muroc Joint Unified School District et al., 97 Cal. Daily Op. Serv. 614 (1997), discussed more fully below).

Some courts formerly allowed cases to proceed on both negligence and breach of contract theories simultaneously (e.g., Schipani v. Ford Motor Co., 30 FEP Cases 361 (Mich. App. 1981)). However, courts in most jurisdictions have refused to recognize a tort claim for negligent evaluation (which might draw a larger damage award) unless the appraisal-related conduct can be distinguished from breach of a duty imposed by contract (e.g., Mooneyham v. Smith Kline & French Labs, 55 FEP Cases 1777 (W.D. Mich. 1990); Haas v. Montgomery Ward & Co., 43 FEP Cases 188 (6th. Cir. 1987)). For practical reasons, most of these cases now tend to involve the employer’s alleged failure to follow its own performance management or progressive discipline procedures, as specified in a collective bargaining or other employment contract. These types of cases are addressed later in this chapter in connection with the legal defensibility of appraisal procedures. As discussed more fully later on, it is a good idea to ensure that poorly performing employees are notified of work-related problems, so they cannot later claim that they would have improved but for the employer’s failure to properly manage their performance.

Defamation. To establish defamation arising out of a negative performance appraisal, an employee must prove (1) that the appraisal contained a false assertion of fact (rather than a subjective opinion incapable of objective verification; e.g., "Smith missed important deadlines on two occasions", rather than "Smith is difficult to deal with"); (2) that the assertion was "published" (i.e., communicated to some third party), and (3) that the published falsehood injured the employee’s reputation (Isaacson v. Keck, Mahin & Cate, 61 FEP Cases 1145 (N.D. Ill. 1993). Because statements that disparage a person’s profession or occupation have been held to be defamatory per se (i.e., not requiring proof of actual injury), the employer’s defense usually relies on lack of publication or a claim of privilege. In general, there is a qualified privilege to communicate, in good faith (without malice), accurate appraisal results, if the employer is under a duty to do so or shares a common interest with the recipient in the subject matter.

For example, in Thompto v. Coburn’s Inc., 10 IER Cases 263 (N.D. Iowa 1994), Thompto, a deli manager, was fired because she failed to meet performance goals for profits, labor costs, and customer complaints. She sued for wrongful discharge and defamation after Coburn’s gave information about her dismissal to Job Service, civil rights investigators, and department managers. The Court rejected Thompto’s claims, however, ruling that Coburn’s had a qualified privilege because it was required to cooperate with state agencies to establish the basis for Thompto’s termination, and because the other recipients of the information held a shared interest in its subject.

Misrepresentation: Walking The Tightrope Between Negligence and Defamation. A recent case illustrates the difficult balancing act employers must perform with regard to job performance information. In Randi W., a Minor, v. Muroc Joint Unified School District et al., the California Supreme Court held that, although an employer is not often accountable for failing to disclose negative information about a former employee, it may be held liable if a recommendation affirmatively misrepresents that a former employee’s performance was favorable when such a misrepresentation might present a substantial and foreseeable risk of harm to a prospective employer or third party.

Randi W., a 13-year old female middle school student, accused her vice-principal, Gadams, of sexual molestation. Gadams had obtained his position through a college placement office, which had received letters from Gadams’ previous employers that described him as "an upbeat, enthusiastic administrator who relates well to the students," and contained glowing praise for Gadams’ "genuine concern" for students, "outstanding rapport" with everyone, and contribution to achieving "a safe, orderly, and clean environment for students and staff." One of the former employers recommended him "without reservation;" the other energetically concluded that it "wouldn’t hesitate to recommend Mr. Gadams for any position!" However, evidence indicated that both employers knew that Gadams had had previous performance problems that included hugging female junior high school students, giving them massages, and making sexual overtures. One of the former employers even had disciplined Gadams in connection with sexual harassment charges that included "sexual touching" of female students, and had pressured Gadams to resign.

Lawyers for Randi W. sued based on negligent misrepresentation and related theories. In the trial court, defendants successfully moved to dismiss on the grounds that they owed no legal duty to Randi W. The Court of Appeals reinstated the case, however, and the Supreme Court agreed. The Court reasoned that, although employers have no duty to say anything about former employees’ performance, when they choose to communicate such information, there is a duty to use reasonable care in doing so. The Court held that defendants should have foreseen that future prospective employers would rely on these favorable letters and decide to hire Gadams, placing students such as Randi W. at risk. The Court was not persuaded by the employers’ arguments that the threat of tort liability would unduly restrict the flow of information and impede job applicants from finding new employment, and that no reasonable person would assume that a letter of recommendation contains the whole truth about a candidate’s performance background and character. However, perhaps wary of the growing number of employers who have reluctantly adopted uniform "no comment" policies (many now refuse to provide any references whatsoever even for stellar former employees), the Court carefully limited its ruling to situations that present "a substantial foreseeable risk of physical injury to the prospective employer or third persons," and which in fact result in such harm. The potential applicability of this holding to a broader range of workplace violence cases is discussed later on in this chapter.

Practical Suggestions for Limiting Discrimination and Related Legal Liability in the Context of Performance Appraisals. A review of the cases digested above discloses a number of ways in which performance appraisals can give rise to unexpected employer liability for discrimination, harassment, constructive discharge, and related theories of liability. Some suggestions for limiting such liability are summarized in Table 4.

Performance appraisals sometimes have been treated as tests in employment litigation, and are theoretically subject to the Uniform Guidelines on Employee Selection (41 CFR 60, et seq.; see Brito v. Zia Co., 478 F.2d 1200 (10th. Cir. 1973); Albemarle Paper Co. v. Moody, 422 U.S. 405 (1975)). However, Barrett and Kernan (1987) have argued that performance appraisals are better understood as criteria correlated with tests, given that there are usually no other performance standards whose correspondence with appraisal results can be independently assessed. In practice, courts have appeared not to pay much attention to formal notions of reliability or validity (Beck-Dudley & McEvoy, 1991), and these issues have rarely been discussed explicitly in reported cases. However, they do appear implicitly in many instances. For example, in a sample of 295 federal appellate discrimination opinions through 1995, Werner and Bolino (1997) found support for the importance of job analysis, written instructions, employee review of results, and triangulation among raters in predicting case outcomes. These and other notions of fairness, accuracy, validity, and due process continue to appear in many legal opinions.

Practical Suggestions for Limiting Discrimination and Related Legal Liability in the Context of Performance Appraisals.

A number of review articles have examined performance evaluations in the caselaw, and have offered recommendations that appear to relate favorably to the legal defensibility of employment decisions (e.g., Ashe & McRae, 1985; Barrett & Kernan, 1987; Beck-Dudley & McEvoy, 1991; Bernardin, Kane, Ross, Spina, & Johnson, 1995; Burchett & De Meuse, 1985; Cascio & Bernardin, 1981; Lubben, Thompson, & Klasson, 1980; Martin & Bartol, 1991; Martin, Bartol, & Levine, 1986; Veglahn, 1993). These reviews represent a consensus regarding substantive and procedural recommendations for legally sound evaluation practices that remain valuable today. Substantive recommendations are summarized in Table 5, while procedural recommendations are summarized in Table 6 a bit later on. Examples from the caselaw follow.

Cases Involving Substantive Aspects of Performance Appraisal

Recent cases concerning the content of performance appraisals have addressed subjectivity, job-relatedness, specificity, and consistency. These cases illustrate both good and not-so-good examples for practice.

Subjectivity and Job-Relatedness of the Criteria. In Eldred v. Consolidated Freightways, 71 FEP Cases 33 (D. Mass. 1995), Eldred, a female dispatcher, was denied a promotion based on her male supervisor’s "gut feeling" that she lacked "aggressiveness" and was too "soft" with Consolidated’s drivers. In ruling for Eldred on her sex discrimination claim, the Court noted that this "gut feeling" was neither reliable nor accurate (the person selected for the promotion turned out to perform poorly), and that the employer’s stated reliance on subjective criteria that were not shown to be related to job performance supported a finding of gender bias.

Similarly, in City of Indian Harbor Beach, 103 LA 634, 637 (Arb. 1994), the City denied one of its police officers a merit pay increase despite

Appraisal criteria:

(1) should be objective rather than subjective;

(2) should be job-related or based on job analysis;

(3) should be based on behaviors rather than traits;

(4) should be within the control of the ratee;

(5) should relate to specific functions, not global assessments;

(6) should be communicated to the employee. ________________________________________________________________________

Note: Recommendations summarized above are drawn in part from Ashe & McRae, 1985; Barrett & Kernan, 1987; Beck-Dudley & McEvoy, 1991; Bernardin, Kane, Ross, Spina, & Johnson, 1995; Burchett & De Meuse, 1985; Cascio & Bernardin, 1981; Lubben, Thompson, & Klasson, 1980; Martin & Bartol, 1991; Martin, Bartol, & Levine, 1986; Veglahn, 1993; and recent cases. Recommendations were extracted, consolidated across articles, and supplemented by the author.

overall performance ratings that were high enough to warrant a raise. The City argued that it never intended for merit increases to be based solely on performance ratings, because evaluations are performed by the immediate supervisor, who typically belongs to the same union as the employee. The arbitrator, however, emphatically rejected any suggestion that management should be allowed to consider other, more subjective factors in its merit pay determinations. In sustaining the officer’s grievance, the arbitrator found that the City had not shown its merit pay determinations to be "based upon evidence and standards that are reasonable, demonstrable, and objective."

On the other hand, in Thomas v. IBM, 67 FEP Cases at 276, IBM was able to successfully defend accusations of age discrimination, in part bacause its appraisal system "contained safeguards to minimize the possibility for unlawful bias or discrimination, such as a written performance plan, objective criteria, a requirement that the supervisor specify in writing the reasons for each rating, and an independent review of the supervisor’s evaluation by the second-level manager" (emphasis added).

Interestingly, in Amirmokri, the court found it permissible to use subjective criteria such as interpersonal skills and team leadership to justify promoting another employee over Amirmokri, and stated that it would have reached the same conclusion even if Amirmokri’s education and experience had been objectively superior to the employee selected. Thus, it appears possible that courts may be more willing to accept the favorable use of subjective criteria to justify a positive employment decision than they are to accept the unfavorable use of such criteria to justify a negative one. Employers should take care to document both types of decisions in any event.

Specificity and Consistency of Performance Criteria and their Application. In addition to Thomas v. IBM, which lauded the use of specific reasons for individual facet ratings, several other cases have dealt with the specificity of criteria and the consistency of their application across individuals or over time. For example, in Fisher v. Vassar College, the District Court had examined Vassar’s "unclear and unspecified" tenure criteria (teaching, scholarship, service, and leadership), and found them to be "disingenuous" and pretextual as applied. However, the Court of Appeals’ reversal in favor of Vassar suggests that in promotion cases involving academics and other professionals, such criteria may be deferred to because these decisions are inherently subjective. Fisher had been able to point out inconsistencies in the way performance criteria were applied to other professors ahead of her on the tenure track, but to no avail. The Court even commented that "it is difficult to conceive of tenure standards that would be objective and quantifiable" (70 FEP Cases at 1165).

Woodman v. Haemonetics Corp. also involved the use of general, global criteria that were inconsistently applied. In that case, performance criteria were changed, in response to a directive from a recently promoted executive, to emphasize "flexibility" (e.g., susceptibility to cross-training and multiple responsibilities) and "participation" (e.g., capacity to provide suggestions for efficiency improvements). After 10 years of good reviews, Woodman’s performance was rated unsatisfactory, and he was terminated in an RIF. The court focused closely on changes in Woodman’s performance ratings after the switch to these new criteria to find that Woodman had made a showing of pretext that was sufficient to go to trial on his ADEA claim.

Cases Involving Procedural Aspects of Performance Appraisal

Recent cases concerning performance appraisal have addressed the presence or absence of many of the procedural safeguards recommended by previous authors, as summarized in Table 6. The absence of such safeguards

Appraisal procedures:

(1) should be standardized and uniform for all employees within a job group;

(2) should be formally communicated to employees;

(3) should provide notice of performance deficiencies, and opportunities to correct them;

(4) should provide access for employees to review appraisal results;

(5) should provide formal appeal mechanisms that allow for employee input;

(6) should use multiple, diverse, and unbiased raters;

(7) should provide written instructions and training for raters;

(8) should require thorough and consistent documentation across raters that includes specific examples of performance based on personal knowledge;

(9) should establish a system to detect potentially discriminatory effects or abuses of the system overall.

________________________________________________________________________

Note: Recommendations summarized above are drawn in part from Ashe & McRae, 1985; Barrett & Kernan, 1987; Beck-Dudley & McEvoy, 1991; Bernardin, Kane, Ross, Spina, & Johnson, 1995; Burchett & De Meuse, 1985; Cascio & Bernardin, 1981; Lubben, Thompson, & Klasson, 1980; Martin & Bartol, 1991; Martin, Bartol, & Levine, 1986; Veglahn, 1993; and recent cases. Recommendations were extracted, consolidated across articles, and supplemented by the author.

generally has not been fatal to the employer unless some other impropriety (e.g., discrimination) or harm to the employee can be shown. On the other hand, where such safeguards do exist, both the employer and employee have been expected to observe and abide by them. Frankly, it is difficult to discern a noticable pattern in this area, and it appears that the legal viability of appraisal procedures will depend on the facts and circumstances of each case. Examples illustrating the success or failure of appraisal procedures follow.

Standardized Procedures and Rater Training. In Kelly v. Drexel University, 5 AD Cases 1101 (E.D. Pa. 1995), the Court rejected Kelly’s argument that his position was eliminated because of age or disability discrimination. Despite the employer’s failure to systematically evaluate employees’ performance, or to provide equal employment opportunity training for raters, the Court could find no evidence that the employer’s stated reasons--lack of computer proficiency and inability to adapt to a heavier workload--were not the real reasons for Kelly’s dismissal.

Procedures and Performance Deficiencies Communicated to the Ratee. In Salt Lake City VA Medical Center, 103 LA 285 (Arb. 1994), the arbitrator rejected a housekeeper’s grievance based on the employer’s failure to inform him of the requirements for a higher performance rating. The grievant’s overall rating was "fully successful," and he had been provided with the standards for that rating under the applicable collective bargaining agreement. On the other hand, in Peoples Natural Gas, 105 LA 37 (Arb. 1995), an employee terminated for unsatisfactory performance in a new job was reinstated, based on the employer’s failure to follow through on its agreed plan of action to provide adequate training and monitor the employee’s level of improvement.

Ratee Access/Input to Appraisal Results, and Review Mechanisms. In Demming v. Housing Authority, Demming’s failure to take advantage of notice and opportunity to respond to an explicit, 49-point performance evaluation was used to support her termination and reject her ADA claim. Similarly, in Town of Bedford, 106 LA 967 (Arb. 1996), a police sergeant’s refusal to cooperate in his own performance evaluation by submitting information that he wanted the employer to consider proved fatal to his grievance to upwardly amend his appraisal results. On the other hand, in Haycock v. Hughes Aircraft, the

employer’s failure to submit a fraudulently altered performance appraisal to Haycock for his signature pursuant to company policy supported his claim of pretext, and helped to sustain his implied contract and defamation claims.

Multiple Raters (Self, Peers, Clients, Others). In Whalen v. Rubin, the employer’s use of multiple raters who independently decided that an employee other than Whalen was most qualified for a promotion helped to negate Whalen’s reverse discrimination claim. In Byrd v. Ronayne, client comments and an associate attorney’s candid self-evaluation that acknowledged performance deficiencies helped to sustain the firm’s decision to terminate her. In McLee v. Chrysler Corp., 73 FEP Cases 751 (2nd. Cir. 1997), a warehouse supervisor’s own admissions conceding the truth of negative performance ratings also helped to undermine any claim that his termination was racially motivated or prextual. And in Golder v. Lockheed Sanders, Inc., 71 FEP Cases 1425 (D.N.H. 1996), peer evaluations helped to support Golder’s performance-based dismissal during an RIF over his unsubstantiated objections that they were unreliable, overly subjective, generally unfair, and a pretext for age discrimination.

On the other hand, in Mathewson v. Aloha Airlines, peer evaluations were found to be pretextual and retaliatory against a pilot who had crossed picket lines during the pilot union’s strike against a different airline. And in Jimenez v. Mary Washington College, potentially collusive and racially motivated student ratings led the District Court to sustain Jimenez’ claim of discriminatory tenure denial (although this decision was reversed on appeal because the College’s performance-based reasons could not be shown to be pretextual). Conversely, in Hampton v. Tinton Falls Police Dept., 72 FEP Cases 101 (3rd. Cir. 1996), a police department’s disregard of a black lieutenant’s recommendation to promote a black sergeant raised a factual issue as to whether favorable performance information was improperly ignored because the rater and ratee were the same race.

In sum, these latter cases suggest that where peers and other untrained or potentially biased evaluators are used, the results should be weighed with caution against more systematic criteria, so that trait-based ratings are avoided, and objectivity, job-relatedness, and other substantive safeguards are preserved. For example, procedures could be developed to ensure that appraisal results that lead to negative employment decisions are reviewed systematically (or at least spot-checked) by I/O psychologists or HR professionals for evidence that age, gender, ethnicity, national origin, union status, or other improper factors may have influenced the outcome. Of course, where such review procedures do exist, they should be followed; see Woodson v. Scott Paper Co., 73 FEP Cases 1237 (3rd. Cir. 1997), in which the employer’s failure to have its corporate review committee examine discrepancies between a black manager’s low RIF ranking and his favorable past evaluations supported a jury’s determination that Woodson was terminated in retaliation for filing a discrimination complaint with the State human rights commission.

Documentation, Consistency, and Rater Knowledge of the Ratee’s Performance. A supervisor’s failure to document in writing purported performance deficiencies to support his "gut feeling" that another employee was better qualified for a promotion led the court in Eldred v. Consolidated Freightways to conclude that stated deficiencies were a pretext for gender discrimination. In Woodman v. Haemonetics, inconsistencies over time in Woodman’s evaluations led the court to conclude that a sudden reduction in his ratings five days before his termination during an RIF was pretextual.

Conversely, consistency over time and across raters led to conclusions that low performance ratings for Dennis, a black transportation planner, and Plummer, a black hotel security director, were not racially motivated; Dennis v. County of Fairfax, 67 FEP Cases 1681 (4th. Cir. 1995); Plummer v. Marriott Corp., 71 FEP Cases 945 (La. App. 1995). And, although the arbitrator in Town of Bedford found no problem with having a police sergeant evaluated by a lieutenant who worked a different shift and had only recently become his supervisor, the courts in Cook v. Arrowsmith Shelburne, Inc., 69 FEP Cases 392 (2nd. Cir. 1995), Henderson & Bryan v. A.T. & T. Corp., and Woodson v. Scott Paper Co. relied partly on raters’ lack of personal knowledge to sustain employees’ claims that negative performance ratings came as a result of age, race, or gender discrimination.

If it is important to observe substantive and procedural safeguards in appraising performance generally, then it is critical to do so when those safeguards are mandated by the terms of an express or implied contract. Alleged violations of contractual appraisal provisions necessarily involve the analysis of case-specific language, so I do not deal with them at length in this chapter. However, some issues in this context are relevant to the overall viability of any organization’s practices. Among these are seniority, the "just cause" standard, and progressive or positive discipline.

The seniority issue frequently arises during an RIF that attempts to accommodate both longevity preferences and performance criteria. For example, in Houston Lighting & Power Company, 103 LA 179 (Arb. 1993), a union contract provided that "in the event of a permanent reduction in force, where ability, skill, and qualifications are equal, seniority shall govern." A number of laid off employees filed grievances over individual decisions, but an arbitrator upheld the employer’s overall system, by which supervisors used a special performance profile to classify individuals within a particular job category as marginal, average, or above average, and applied seniority criteria within these classifications. This system also was upheld by the Court of Appeals in a related case, in which the Court found the system "reasonably calculated to fairly accomplish the determination of utility, skill, and qualifications" using criteria that were "within the scope of contractually allowed factors," 151 LRRM at 2023 (5th. Cir. 1995).

Failure to adhere to an informal seniority system during an RIF is not necessarily fatal to the employer if legitimate reasons for departing from prior practice are credibly put forth. In EEOC v. Texas Instruments, Inc., the Court found that the need to retain employees who could "satisfy and adapt to a rapidly changing technological environment" justified laying off older supervisors over younger ones who possessed college degrees, superior computer skills, and other technological expertise (72 FEP Cases at 982). As mentioned earlier, however, it would be best to keep job descriptions and performance criteria updated to the greatest extent practicably possible.

Express or implied agreements not to terminate an employee except for just cause are among the most commonly litigated exceptions to the doctrine of employment at will. For example, the court’s decision to uphold the employee’s reinstatement in Mathewson v. Aloha Airlines was based in large part on an implied agreement that Mathewson would be evaluated only on his merits and qualifications, rather than on biased peer appraisals influenced by his perceived "scab" status. And in Postville School District v. Billmeyer, 152 LRRM 2401 (Iowa 1996), the Iowa Supreme Court refused to overturn an arbitrator’s decision that failed to find evidence of the just cause that would have been required to support the School District’s termination of an employee for alleged sexual misconduct outside the workplace. On the other hand, the arbitrator in Linconview Local Board of Education, 103 LA 854 (Arb. 1994), found no basis to apply a just cause standard in the Board of Education’s decision to non-renew the limited one-year contract of a teacher, so long as it abided by its procedural obligations with respect to its performance appraisal processes. These obligations, with which the Board had complied, included evaluating the teacher twice a year based on actual classroom observation, conferring with the teacher and preparing written reports of evaluations, and providing copies of evaluations to the teacher.

Both progressive and positive discipline procedures involve a series of steps that allow a problematic employee to improve performance before being discharged. Progressive discipline involves warnings and punishments of increasing severity, while positive discipline involves counseling and interventions of increasing intensity (Gomez-Mejia et al., 1995). In general, where either system is present, adhering to stated procedures will support the employer’s discharge of the employee, whereas failure to do so will support the employee’s allegation of improper motivation on the part of the employer.

For example, in Hanchard v. Facilities Development Corp., 10 IER Cases 1004 (N.Y. App. 1995), the Facilities Development Corp. (FDC) employee handbook set forth a disciplinary plan with both positive and progressive features by which an employee of 5 or more years was to be provided performance counseling and a work plan to measure and guide improvement. The handbook also provided for the right to submit documentation and request a hearing prior to termination, and the right to be discharged only for "materially deficient work performance." Hanchard, an architect, had declining performance evaluations, became disruptive and insubordinate, was behind on most of his projects, and alienated both clients and co-workers. In a split decision, the Court rejected Hanchard’s claim that his discharge was arbitrary and capricious, finding that the employer had "substantially complied" with its disciplinary procedures (the Court also found that Hanchard had failed to take advantage of the protections afforded to him). The dissent felt that FDC’s compliance with its procedures had not been "substantial" enough.

For other cases in which the employer’s adherence to its stated procedures led to a successful defense of its position, see Gipson v. KAS Snacktime Co., 71 FEP Cases 1677 (E.D. Mo. 1994) (employee’s failure to cooperate in personal development program offered by the employer negated claim of pretext in race discrimination case); R.G.H. v. Abbott Laboratories (employee’s failure to respond to performance improvement plan over six month period negated claim of pretext in disability discrimination case); and Samaritan Health System, 106 LA 927 (Arb. 1996) (employer’s substantial compliance with progressive discipline and performance evaluation procedures, coupled with employee’s failure to challenge negative performance appraisal, supported employee’s discharge for poor performance).

However, also see Chertkova v. Connecticut General Life Insurance, 71 FEP Cases 1006 (2nd. Cir. 1996). In that case, a supervisor’s abuse of the "performance improvement plan" process, in which so-called "coaching sessions" allegedly involved berating Chertkova behind closed doors and yelling comments such as "there is no chance! You are not going to be here!" was found to warrant reinstatement of her previously dismissed claims for constructive discharge and pretext in the context of sex discrimination. This case further underscores that disregard of proper appraisal procedures, whether by the employer or the employee, can cause problems in litigating discrimination and other kinds of legal disputes.

It has been estimated that almost 70% of the American populace now can claim membership in one or more protected classes under various anti-discrimination laws. This number can be expected to grow as increased immigration, greater longevity due to medical advances, and the widening scope of allowable mental health disabilities modify current demographics. As global competition and technology continue to drive RIFs and other restructuring, protected class status will continue to figure prominently in challenges to layoffs and discharges that employers claim were based on performance but employees claim were based on discrimination. The move toward "leaner, meaner" organizations, in which employees are expected to work harder and take on a wider range of responsibilities, also will increase the potential for performance appraisals to generate legal issues in emerging areas. Among these areas are flexible job designs, security and privacy, and workplace violence.

In the context of innovations such as team-based designs, Mirvis and Hall (1996) have noted the growing incidence of flexible firms, flexible jobs, and skill-based individual careers. Continuous learning, self-management, self-designed jobs, and flexible work roles are replacing rigid job descriptions and classification schemes. These changes carry implications for appraising performance, particularly with respect to demonstrating the job-relatedness or essential nature of performance criteria such as "interpersonal skills," "ability to lead a team," "flexibility," and "participation" (Amirmokri v. Baltimore Gas & Electric; Woodman v. Haemonetics Corp.). Although such criteria sometimes pass muster (as in Amirmokri), they also may form the basis for a claim of pretext (as in Woodman). This points out the need for employers to further define, refine, and validate flexible performance criteria, in order to be able to argue successfully in defense of discrimination claims that the use of such criteria serves a legitimate business function.

In re National Archives and Records Administration, 107 LA 123 (Arb. 1996) presents a case in which the high stakes often linked with favorable performance ratings (here, qualification for a promotion) led a National Archives employee to falsify his ratings by fraudulently altering one of his written appraisals. The employee, who had a top security clearance, was discharged for cause, in part because he had betrayed his position of trust with the agency. This points out the need for employers to develop appropriate security procedures to prevent similar occurrences, and to avoid compromising the privacy of other employees whose performance results also may be on file.

In Sturm v. TOC, security and surveillance actually were principal parts of a convenience store manager’s performance appraisal duties, which perhaps takes the recommendation that evaluations should be based on personal knowledge to the extreme. Sturm’s appraisal duties involved different forms of ongoing honesty testing, including "baiting" the cash register with extra cash, looking over employees’ shoulders, and viewing video surveillance tapes. Given the severe problem with employee theft faced by many companies, and the increasing deference afforded to private employers in areas such as drug testing, it is probable that alleged invasions of privacy based on conduct such as that in Sturm will arise with increasing frequency in challenges to such practices. Plummer v. Marriot Corp., a constructive discharge case, provides an example of such a challenge in connection with a polygraph test during an internal investigation of employee theft.

As noted earlier in connection with the Randi W. case, employers face increasing potential liability for non-disclosure of problematic employee performance, as when their affirmative misrepresentation that a former employee’s performance was favorable presents a foreseeable risk of harm to others. With appraisal results continuing to figure prominently in layoffs and discharges, employee reactions to appraisers and appraisal processes seem increasingly likely to lead to injury, violence, or death in and outside the workplace. News media have reported such instances, as where an employee committed suicide by sticking his head into a giant circular saw when told by his supervisor that his already heavy workload standards would be increased yet again ("Firm Blamed," 1995). When an employee exhibits signals during an evaluation that suggest the possibility that such problems might exist (e.g., symbolically decapitating one’s supervisor during a meeting to discuss threatening behavior toward co-workers; Lewis v. Zilog, Inc.), it would not be surprising to see an employer held liable for related violence on the grounds that it knew or should have known that serious harm to the employee or others might result.

Although it remains difficult at best to predict who among the many employees that exhibit stress-related symptoms will actually resort to violence, the ADA imposes a duty on employers to provide reasonable accommodations to employees who suffer from mental disabilities, including those that might lead to such conduct. Opportunities for reduced stress, flexible workloads, and professional counseling might be appropriate first steps toward such an accommodation. Of course, in extreme cases, as where an employee poses "a direct threat to the health or safety of other individuals in the workplace," 42 U.S.C. Section 12113(b), there may be no choice but to remove the employee from the workplace. Employers can defend resulting discrimination claims on the grounds that the employee was not qualified under the ADA because he or she could not perform the essential functions of their job. Employers have successfully argued in such situations that the essential functions of any job include the ability to deal with criticism from supervisors and coworkers in a civil manner, and to refrain from violent behavior (see Fram, 1997; Willis, 1997).

As our economy continues increasingly to move toward a service and information emphasis, it seems inevitable that flexible, subjective performance criteria will remain in use, particularly at the professional and managerial level. As global competition continues increasingly to demand a customer focus and total quality orientation, it also seems inevitable that multiple sources of feedback (clients, subordinates, peers) will remain in use, perhaps at all levels of employment. As we have seen, both subjective criteria and biased or untrained raters can spawn discrimination claims that are difficult to defend. Where such criteria or raters are used, they would be best used in conjunction with objective criteria and trained, unbiased raters whose input is given greater (perhaps determinative) weight. This practice would preserve the benefits of multiple criteria and information sources, while also providing defenses to discrimination claims in the form of legitimate, nondiscriminatory reasons for adverse employment decisions or the absence of pretext. This practice also would support the mitigating defense, in a mixed motive case, that the employer would have made the same decision notwithstanding the use of arguably discriminatory factors.

I close by summarizing some additional recommendations from a noted practitioner in the employment law arena that become particularly relevant in termination cases (Cathcart, 1996). Employers should review appraisal procedures to ensure (1) that employees are provided with an oral interview and a written statement of their appraisal, as well as the opportunity to acknowledge in writing receipt or review of the appraisal; (2) that managers are encouraged to support candor in appraisals, to avoid central tendency or "friendliness" errors that can make subsequent demotions or discharges difficult to defend; and (3) that appraisers are trained and "spot checked" from time to time, to identify and correct practices that might generate legal liability. Employers also should conduct a thorough pre-termination review of all information on potential candidates for discharge, and determine how other employees were treated who were subject to discretionary judgments of the same supervisor. Employers should consider the employee’s seniority, the importance and specificity of alleged performance deficiencies, past practices in similar circumstances, the impartiality of evaluators, any extenuating circumstances, and compliance with company discipline and discharge procedures. Finally, the employer should consider any potential for defamation or invasion of employee privacy, as well as the manner in which the employee (and anyone else) will be told about the adverse decision.

Hopefully, these recommendations will increase the chances that your appraisal practices manage to avoid legal scrutiny. If, as has frequently become the case, however, your appraisal practices do wind up in litigation, I offer words that Obi-Wan Kenobi might have said: "May the Courts be with you."

Ashe, R. L., & McRae, G. S. (1985). Performance evaluations go to court in the 1980’s. Mercer Law Review, 36, 887-905.

Barrett, G. V., & Kernan, M. C. (1987). Performance appraisal and terminations: A review of court decisions since Brito v. Zia with implications for personnel practices. Personnel Psychology, 40, 489-503.

Bernardin, H. J., Kane, J. S., Ross, S., Spina, D.S., & Johnson, D. L.

(1995). Performance appraisal design, development, and implementation. In G. R. Ferris, S. D. Rosen, & D. T. Barnum (Eds.), Handbook of Human Resource Management (pp. 462-493). Cambridge, Mass.: Blackwell.

Beck-Dudley, C. L., & McEvoy, G. M. (1991). Performance appraisals and discrimination suits: Do courts pay attention to validity? Employee Responsibilities and Rights Journal, 4, 149-163.

Burchett, S. R., & De Meuse, K. P. (1985). Performance appraisal and the law. Personnel, 62, 29-37.

Cascio, W. F., & Bernardin, H. J. (1981). Implications of performance appraisal litigation for personnel decisions. Personnel Psychology, 34, 211-226.

Cathcart, D. A. (1996). Employment termination litigation: Collateral tort theories and the multimillion dollar verdict: A 1994 Update. In Employment Discrimination and Civil Rights Actions in Federal and State Courts (pp. 675-779). Philadelphia: The American Law Institute.

Cathcart, D. A., & Snyderman, M. (1997). The Civil Rights Act of 1991. In P. M. Panken (Ed.), Employment and Labor Law (pp. 277-361). Philadelphia: The American Law Institute.

Firm blamed for suicide. (1995, November 3). The San Jose Mercury News, p. D1.

Fram, D. K. (1997). Expanded checklists for determining "disability," "qualified," and "reasonable accommodation" under the ADA. In National Employment Law Institute Human Resources Workbook: Larkspur, CA: NELI.

Gomez-Mejia, L. R., Balkin, D. B., & Cardy, R. L. (1995). Managing Human Resources. Englewood Cliffs, NJ: Prentice Hall.

Koy, D. J., Briggs, S., & Grenig, J. (1987). State court disparity on employment-at-will. Personnel Psychology, 40, 565-577.

Lubben, G. L., Thompson, D. E., & Klassen, C. R. (1980). Performance appraisal: The legal implications of Title VII. Personnel, May-June, 11-21.

Martin, D. C., & Bartol, K. M. (1991). The legal ramifications of performance appraisal: An update. Employee Relations Law Journal, 17, 257-286.

Martin, D. C., Bartol, K. M., & Levine, M. J. (1986). The legal ramifications of performance appraisal. Employee Relations Law Journal, 12, 370-395.

Miller, C. S., Kaspin, J. A., & Schuster, M. H. (1990). The impact of performance appraisal methods on Age Discrimination in Employment Act cases. Personnel Psychology, 43, 555-578.

Mirvis, P. H., & Hall, D. T. (1996). New organizational forms and the new career. In D.T. Hall (Ed.), The Career is Dead--Long Live the Career: A Relational Approach to Careers (pp. 72-101). San Francisco: Jossey-Bass.

Perry, E. L., Kulik, C. T., & Bourhis, A. C. (1996). Moderating effects of personal and contextual factors in age discrimination. Journal of Applied Psychology, 81, 628-647.

Veglahn, P. A. (1993). Key issues in performance appraisal challenges: Evidence from court and arbitration decisions. Labor Law Journal, October, 595-606.

Werner, J. M., & Bolino, M. C. (1997). Explaining U.S. Courts of Appeals decisions involving performance appraisal: Accuracy, fairness, and validation. Personnel Psychology, 50, 1-24.

Willis, S. G. (1997). Stress-related disability claims under the ADA. The Practical Lawyer, 43, 73-86.