Disparate Impact - Case Study Example

 Topic: Disparate Impact The complexity of discrimination law requires some basic understanding of the term “Disparate Impact”. Disparate Impact is a circumstantial method of substantiating discrimination. Taking an example to clarify the term “disparate Impact”, a regulation necessitating all janitors to possess high school diplomas doesn’t seem to be discriminatory from above but it may have “disparate impact” on certain sections of society with high rate of dropouts. In such a situation, a plaintiff may transfer the responsibility of the negative effect of not having high school diploma on the city by following the disparate impact theory of liability to prove its business requirement – not just a ploy to single out certain groups of society from getting employment (Lazarus, 2001). The Supreme Court first described the disparate impact theory in 1971, in Griggs v. Duke Power Co., 401 U.S. 424, 431-2 (1971): Title VII. It “proscribes not only overt discrimination but also practices that are fair in form, but discriminatory in operation. The touchstone is business necessity. . . . [G]ood intent or absence of discriminatory intent does not redeem employment procedures or testing mechanisms that operate as 'built-in headwinds' for minority groups and are unrelated to measuring job capability.” In 1989, the Supreme Court minimized the defendant’s burden of proving business necessity to a burden of producing proof of business requirement in the case of Wards Cove Packing Co. v. Antonio, 490 U.S. Later, the Civil Rights Act of 1991 annulled that part of the Wards Cove decision (HR Guide 2001). Disparate Impact 2 "Disparate impact" is a legal theory for proving unlawful employment discrimination. But in practice, “disparate treatment” theory is practiced. Disparate impact is a thought that some recruitment practices adversely impact a group or community of people than the others. In the example of US Supreme Court Title VII case on the issue of disparate impact, in a particular case of employing laborers, the applicants needed to be high school diploma holders. This condition weeded out more blacks than whites, although there was no such intention on the part of the employer to discriminate against blacks. But as a result of the condition, there was a disparate impact on a particular race (Runkel, 2006). According to the Supreme Court, if the employees raise such a concern, the responsibility of proving the usefulness of the high school diploma lies with the employer, having “a manifest relationship to the employment in question.” According to federal legislation acted out in 1991, in case such a disparate impact is felt by some employee, it is for the employer to show that the practice “is job related for the position in question and consistent with business necessity.” Such cases are difficult and costly, as they require both, the experts and delicate mathematical factors (Runkel, 2006). Through statistical comparisons, plaintiff can prove that it is a prima facie case having negative impact on a group of people. Then it is the liability of the employer to prove that the practice is “job-related for the position in question and consistent with business necessity.” Disparate Impact 3 Further, even if an employer proves a business necessity, a plaintiff can insist on the secondary choice to be applied, to which the employer has no issues regarding business necessity and it is not adversely impacting the plaintiff’s interests. See 42 U.S.C. § 2000e-2(k)(1)(A)(i)(ii) (HR Guide 2001). There are two ways of measuring adverse impact, either by following EEOC’s Uniform Guidelines on Employment Selection Criteria or via non-scored objective criteria. The courts take it as adverse impact if the difference between the number of members of favored class selected and the number that would be predicted in a random selection is more than two or three standard deviations. The defendant can challenge the prima facie case by showing that the scored test is job related and uniform with business necessity and the test is corroborated (HR Guide, 2001). The non-scored objective criteria: Other criteria of employee qualifications like education, experience and licensing needs also require Uniform Guidelines. In certain cases like clerical jobs, educational qualifications have been found to create a disparate impact, as has been the case of Griggs v. Duke Power Co., 401 U.S. 424 (1971). The condition of passing high school diploma was invalidated because only 12 percent African American males had the diploma in comparison to 34 percent white males and further, the defendant could not prove a connection between the high school diploma and job performance. It is observed that the burden of proving job relatedness is lower for the defendant if the professional qualifications of the candidate are higher or the company is hiring unskilled applicants. See, e.g., Briggs v. Disparate Impact 4 Anderson, 796 F.2d 1009, 1023 (8th Cir. 1986) (college degree in psychology is a valid requirement for counselor position); Aguilera v. Cook County Police & Corrections Merit Board, 760 F.2d 844, 848 (7th Cir.), cert. denied, 474 U.S. 907 (1985) (high school diploma requirement for police officers and corrections officers is valid) (HR Guide, 2001). The subjective decision making can be challenged under the disparate impact theory, specifically when employment decisions regarding white collar jobs have to be made. Watson v. Fort Worth Bank & Trust, 487 U.S. 977 (1988) (HR Guide, 2001). Standard-Deviation Report The difference between the proportion of the protected class Selected and the proportion of all Applicants Selected has a normal distribution with a mean and standard deviation. The statistic is shown below: (r / n) - p Disparate Impact 5 Table1. Analysis of proportion of Females Selected where: r = number of Females Selected. n = number of Selected (Females and Males). p = proportion of Applicants that are Females. q = proportion of Applicants Selected. Selected Not Selected Row Totals Males 2 10 12 Females 3 10 13 Column Total 5 20 25 r = 3 n = 5 p = 13 / 25 = 0.52 q = (3 + 2) / (13 + 12) = 0.2 Standard Deviation Statistic = 0.4 These results show that the proportion of Females Selected is 0.4 standard deviations above the proportion of Applicants Selected. A result of less than 2 standard deviations is generally considered non-significant. Table 2. Analysis of proportion of Minorities Selected where: r = number of Minorities Selected. n = number of Selected (Minorities and Non-Minorities). p = proportion of Applicants that are Minorities. q = proportion of Applicants Selected. Selected Not Selected Row Totals Non-Minorities 3 30 33 Minorities 5 45 50 Column Total 8 75 83 r = 5 n = 8 p = 50 / 83 = 0.602 q = (5 + 3) / (50 + 33) = 0.096 Standard Deviation Statistic = 0.137 These results show that the proportion of Minorities Selected is 0.137 standard deviations above the proportion of Applicants Selected. A result of less than 2 standard deviations is generally considered non-significant. Table 3. Analysis of proportion of Older Selected where: r = number of Older Selected. n = number of Selected (Older and Younger). p = proportion of Applicants that are Older. q = proportion of Applicants Selected. Selected Not Selected Row Totals Younger 10 11 21 Older 5 5 10 Column Total 15 16 31 r = 5 n = 15 p = 10 / 31 = 0.323 q = (5 + 10) / (10 + 21) = 0.484 Standard Deviation Statistic = 0.124 These results show that the proportion of Older Selected is 0.124 standard deviations above the proportion of Applicants Selected. A result of less than 2 standard deviations is generally considered non-significant. Disparate Impact 7 Table 4. Analysis of proportion of Disabled Selected where: r = number of Disabled Selected. n = number of Selected (Disabled and Non-Disabled). p = proportion of Applicants that are Disabled. q = proportion of Applicants Selected. Selected Not Selected Row Totals Non-Disabled 40 10 50 Disabled 2 3 5 Column Total 42 13 55 r = 2 n = 42 p = 5 / 55 = 0.091 q = (2 + 40) / (5 + 50) = 0.764 Standard Deviation Statistic = -2.007 These results show that the proportion of Disabled Selected is more than two standard deviations below the proportion of Applicants (Disabled plus Non-Disabled) Selected. Quantum of Business Necessity Proof Required It is to be observed that although it is the employer’s burden to prove the business necessity but the significance of the numerical result can’t be under-estimated. As a general rule, the severity of judicial checking of employer’s business necessity proof correlates with the quantum of disparity – greater disparity – extreme scrutiny. One court declared, “As a general principle, the greater the test's adverse impact, the higher the correlation [between test performance and job performance], which will be required”(Sauls, 1995). Disparate Impact 8 The Griggs v. Duke Power Co., 401 U.S. 424 (1971) is the originative case, causing duality of meaning: whether it is the claim of disparate impact (Griggs) or disparate treatment (McDonnell-Douglas), or both. Take the example of Wal-Mart, where women neither wanted nor were eligible for promotions but might be a part of financial remedy, which underlines the importance of statistics. The disparate impact analysis makes quota-based workforce management enticing but wrong. There is a positive side of the whole issue, which makes good business sense for other than legal reasons. It is applying “best practices” HR techniques throughout the process of scrutinize hiring, compensation, promotion and termination practices (Lenard, 2004). “Disparate Impact” Litigation: Pro and Con Conservatives have always criticized the disparate impact approach to discrimination, because it can be used as a tool to propagate against any entrance test. It underscores high standards, is a support of below-average performers, and puts a high cost on businesses and local bodies. It also divides society into different ethnic and racial groups, not an ordinary gathering of individuals. Conservatives find difference in the focus on rules producing disparate effects or rules reflecting discriminatory intent – the conservatives prefer the latter option. The liberals in the Congress retorted back by calling it bigotry, which was better hidden because it is not easy to prove discriminatory intent (Lazarus, 2001). Disparate Impact 9 It is agreed by all that the burden of proof issue once resolved by Wards Cove in favor of the employer, is resolved in favor of the plaintiffs by the Civil Rights Act of 1991. In Wards Cove, the Court put the liability of producing evidence, but not the burden of persuasion to demonstrate business necessity, on the employer. On the other hand, the 1991 Act asks the employer to “demonstrate” that a questioned practice is job-related. Certain issues are still ambiguous – how “business necessity” will be worked out in future disparate impact cases but the issue of burden of proof with regard to production and persuasion is clearly stated in the 1991 Civil Rights Bill (Pattison et al, 1991). The issue of “business necessity” with its accompanying relationship to validity evidence is central to personal researchers. The validation standard set in the Uniform Guidelines does not pay attention to high costs incurred by employers but the courts have taken a positive view of employers’ arguments of justifying negative impact of cost/benefit analysis. The courts also fear that employers would consider hiring quotas the best alternative strategy to a limited definition of business necessity. The Supreme Court has taken a flexible attitude on employer’s burden. There is need to make fruitful efforts by furthering the boundaries of validity research, matching with the spirit of the Uniform Guidelines. The issues at stake are not rigid and may take a new turn with the passing of time. Further evolution in this field over the next few years could alter the opinions on business necessity (Verca et al, 1993). Disparate Impact 10 References: HR Guide to the Internet. (2001). EEO: disparate impact. Retrieved Monday, February 19, 2007 from http://www.hr-guide.com/data/G702.htm. HR-Software.net. (1998). Disparate impact analysis. Retrieved Monday, February 19, 2007 from http://www.hr-software.net/EmploymentStatistics/DisparateImpact.htm. Lazarus, Edward. (2001, May 1). The case that roared: A limited “disparate impact” Holding that could have larger repercussions. Find Law Resources. Retrieved Monday, February 19, 2007 from http://writ.news.findlaw.com/lazarus/20010501.html. Lenard, George. (2004). Disparate impact law, part I – where it all began. George's Employment Blawg. Retrieved Monday, February 19, 2007 from http://employmentblawg.blogspot.com/2004/08/disparate-impact-law-par t-i-where-it.html. Pattison et al.(1991). The resurrection of the disparate impact theory? American Business Law Journal, 29(3). Retrieved Monday, February 19, 2007 from http://proquest.umi.com.myaccess.library.utoronto.ca/pqdlink?Ver=1&Exp=02-14-2012&FMT=7&DID=1450946&RQT=309&cfc=1. Disparate Impact 11 Runkel, Ross. (2006). Disparate impact #20. Law Memo. Retrieved Monday, February 19, 2007 from http://www.lawmemo.com/101/2006/01/disparate_impac_1.html. Sauls, John G. (1995, April). Proving business necessity: The disparate impact challenge. 'Lectric Law Library's stacks. Retrieved Monday, February 19, 2007 from http://www.lectlaw.com/files/emp35.htm. Verca et al.(1993). Evidentiary standards in employment discrimination: A view toward the future. Personnel Psychology, 46(2). Retrieved Monday, February 19, 2007 from the Community of Scholars™ database. . Read More