Adverse Impact Background  After Title VII was passed, concerns remained that tests were being used in industry that functioned to limit.

Download Report

Transcript Adverse Impact Background  After Title VII was passed, concerns remained that tests were being used in industry that functioned to limit.

Adverse Impact Background
 After Title VII was passed, concerns remained that tests were
being used in industry that functioned to limit the opportunities
for minorities to gain employment
 Such concerns were fostered by the exemption in the language
contained within Title VII regarding the use of psychological
tests:
“…nor shall it be an unlawful employment practice for an employer to
give and to act upon the results of any professionally developed ability
test provided that such test, its administration or action upon the results
is not designed, intended or used to discriminate because of race, color,
religion, sex or national origin.”
EEOC Statement (1966) regarding professionally developed tests:
The Commission accordingly interprets “professionally
developed ability tests” to mean a test which fairly measures
the knowledge or skills required by the particular job or class
of jobs which the applicant seeks, … The fact that a test was
prepared by an individual or organization claiming expertise in
test preparation does not, without more, justify its use within
the meaning of Title VII.” (35 Fed. Reg. 12333).
Griggs v. Duke Power (1971)
Some Key Questions:
• Can employment practices be in violation of Title VII if there is a lack
of discriminatory intent (or motive)?
• Can evidence that a practice disproportionately excludes members of
a protected group be used to form a prima facie case?
• What standard/evidence must exist to defend the use of devices used
for personnel decisions (e.g., selection, promotion)?
Griggs Background
Duke Power organized into 5 departments:
1) Labor (Blacks limited to these jobs; highest pay here was less than the lowest
wage in the other departments)
2) Coal Handling
3) Operations
4) Maintenance
5) Laboratory & Test
Promotions made on the basis of seniority within each department
Duke Power incrementally began adding requirements:
1955: Company required a HS diploma for jobs, except Labor
1965: Company a HS diploma was required to transfer from the Labor Dept.
(*White employees hired before HS requirement performed satisfactorily)
Later in 1965: Required passing scores on 2 “aptitude tests” (passing score
approximated the national average for HS students; implemented the day Title VII
became law!!!)
Griggs v. Duke Power (1971)
Company’s Promotion Requirements (Facially neutral)
1) Pass the Wonderlic Personnel Test and Bennett Mechanical Aptitude Test
2) Possess a high school diploma
Challengers provided evidence that the requirements disproportionately
excluded protected group members (i.e., Blacks)
 Tests eliminated roughly 94% of blacks as compared to 43% of whites.
 12% of Blacks possessed a high school diploma in NC versus 34% of Whites
*** Consists of 50 questions with a 12-minute time limit
Griggs v. Duke Power (cont.)
 Company defense: Argued that the use of "professionally developed
tests" was allowed under Title VII (technically correct)
 Company also said that they did not intend to discriminate against protected
group members by mandating their requirements (tests and diploma)
From the SC decision: “… Congress directed the thrust of the Act to
the consequences of employment practices, not simply the
motivation. More than that, Congress has placed on the employer the
burden of showing that any given requirement must have a manifest
relationship to the employment in question.”
“… The Act proscribes not only overt discrimination but also practices
that are fair in form, but discriminatory in operation. The
touchstone is business necessity. If an employment practice which
operates to exclude Negroes cannot be shown to be related to job
performance, the practice is prohibited.”
Major Points of the Supreme Court Decision in Griggs
• Employment discrimination may result from the effects of one's actions
• All tests/requirements must be job related and consistent with business
necessity
• Proof of an intention to discriminate is not required for discrimination to
exist
• All tests must meet standard professional guidelines for psychometric
worth
~ Albemarle Paper v. Moody (1975) ~
Albemarle’s Testing Requirements:
Pass the Beta Exam (a test of nonverbal intelligence) and the Wonderlic Test
Why use the Wonderlic?
... selected based on” the theory that a certain verbal intelligence was called for by the
increasing sophistication of the plant's operations.”
• No attempt to ascertain the job-relatedness of the Wonderlic
• No rationale for the use of the its cutoff score (national norm) for passing
~ Albemarle’s Validation Study ~
• A few months before the trial, Albemarle hired a consultant to validate its tests
• 10 job groupings, 9 lines of progression (sample consisted of those near the top of the
progression lines --- 105 employees; 4 Blacks)
• Use of concurrent validation approach (test scores correlated with supervisor rankings
(pair comparisons)
• Instructions to supervisors regarding assessing the performance of their subordinates:
"determine which ones they felt, irrespective of the job that they were actually
doing, but in their respective jobs, did a better job than the person they were rating
against . . .”
Findings --- A “patchwork” of correlations
Beta Exam: Significant correlations in 3 out of the 8 lines of progression
Wonderlic: Significant correlations for one form (not the other) obtained in 4 job
groups
No significant correlations obtained in 2 groups; tests significant within some lines of
progression but not others
Albemarle --- Summary of Key Points
>>> Test scores were correlated (rather haphazardly) with job
performance measures across several job categories, but no evidence
existed showing that the jobs were comparable
“The study in this case involved no analysis of the attributes of, or the particular skills needed
in, the studied job groups. There is accordingly no basis for concluding that "no
significant differences" exist among the lines of progression, or among distinct job
groupings within the studied lines of progression. Indeed, the study's checkered results appear to
compel the opposite conclusion.”
>>> Subjective supervisor evaluations of employees were based on unclear
performance criteria (e.g., supervisors were asked to assess which workers,
regardless of job category, were better than the individual to which they
were being compared)
There is no way of knowing precisely what criteria of job performance the supervisors were
considering
Albemarle --- Summary of Key Points (cont.)
>>> The sample used in the validation study was heavily comprised of
those who were white, in upper-level positions, and who had greater years
of experience, than the typical job applicant for whom the tests were being
administered
Albemarle's validation study dealt only with job-experienced, white workers; but the
tests themselves are given to new job applicants, who are younger, largely inexperienced, and
in many instances nonwhite. The APA Standards state that it is "essential" that "[t]he validity of
a test should be determined on subjects who are at the age or in the same educational or vocational
situation as the persons for whom the test is recommended in practice."
~ Connecticut v. Teal (1982) ~
["bottom line" defense]
Background --• State required the passing of a written test for promotion
• Those who passed the test were placed on a list indicating that they were
eligible for future promotion consideration
• Once on the list, other factors were evaluated in making promotion
decisions such as supervisor recommendations, work performance, and
job tenure.
Black candidates who failed the exam sued, saying that the test was unrelated to
the job and resulted in adverse impact (Multiple Hurdle approach used)
% of Whites who passed the test: 79%
% of Blacks who passed the test: 54% (or a passing rate for blacks that was 68%
less than that of whites). Clear violation of the 4/5 rule.
• But, overall, 22.9% of Blacks were promoted vs. 13.5% of Whites*
* Promotion decisions were made about a year after the suit was filed and about 30 days before the
trial.
Connecticut’s position:
Connecticut said they were in compliance with the Uniform Guidelines. That
is, 1607.4(C) of the Uniform Guidelines states:
[1] If … the total selection process for a job has an adverse impact, the
individual components of the selection process should be evaluated for adverse
impact.
[2] If … the total selection process does NOT have an adverse impact,
the Federal enforcement agencies … will not expect a user to evaluate
the individual components for adverse impact, or to validate such
individual components, and will not take enforcement action based
upon adverse impact in any component of that process
~ Supreme Court Decision in Teal ~
 Court decided in favor of the plaintiffs. It concluded that the law protects
individual employees and that discrimination can exist even though the
group as a whole fared well
• "Title VII does not permit the victim of a facially discriminatory policy to
be told that he has been wronged because other persons in his or her race or
sex were hired.” … Every individual employee is protected against both
discriminatory treatment [457 U.S. 440, 456] and "practices that are fair in
form, but discriminatory in operation. ... In sum, petitioners'
nondiscriminatory "bottom line" is no answer, under the terms of Title VII, to
respondents' prima facie claim of employment discrimination.
• It is clear that Congress never intended to give an employer license to
discriminate against some employees on the basis of race or sex merely
because he favorably treats other members of the employee’s group
~ Adverse Impact Process Before Wards Cove ~
Step 1: The challenger must identify a specific employment practice
that caused the discrimination in question (disproportionately
excludes protected group members)
Step 2: The company must demonstrate that the challenged practice
is job related and consistent with business necessity
Step 3: The challenger must prove that an equally valid, job-related
practice exists with less (or no) adverse impact
Watson v. Ft. Worth Bank and Trust (1988)
Some Key Points:
A)
Plaintiffs do NOT need to show intentional discrimination with regard to subjective
information used for employment decisions. That is, subjective information needs to
comply with the requirement set forth in Griggs regarding job-relatedness.
When an employer's "...undisciplined system of subjective decision-making has precisely
the same effects as a system perverted by intentional discrimination, it is difficult to see
why Title VII" should not apply"
Therefore, subjective employment practices can be challenged under disparate impact
rule
B) A statistical difference may not be enough to establish a prima facie case
C) Majority of the Court required the plaintiff to identify the SPECIFIC employment
practice that CAUSED the statistical disparity between "protected" and
"unprotected" group
D) Plurality lessened the burden of proof on the part of the company in demonstrating
business necessity to that of production
The Plurality Opinion in Watson (Justice O’Conner)
Watson v. Ft. Worth Bank and Trust (cont.)
"Our cases make it clear that employers are not required, even when defending standardized or
objective tests, to introduce formal 'validation studies' showing that particular criteria predict
actual on-the-job performance”
"...formal validation techniques endorsed by the EEOC in its Uniform Guidelines may sometimes not
be effective in measuring the job-relatedness of subjective selection procedures"
Not true; objective and subjective data require the same psychometric rigor (APA brief)
"...It is self-evident that many jobs, for example those involving managerial responsibilities, require
personal qualities that have never been considered amenable to standardized testing...courts are
generally less competent than employers to restructure business practices, and unless mandated to do
so by Congress, they should not attempt it"
Selection Methods Amendable to Validation Studies (SIOP Principles)
* Underlined factors were challenged by Watson
Wards Cove v. Antonio (1989)
Basic facts/background
Ward’s Cove:
• Salmon cannery company
• Operates only in summer
• Location of salmon runs vary as does the number of employees in each
site
Two general types of jobs
1) Cannery (mostly minority, hired via a union, live near job sites)
2) Noncannery (skilled position, mostly White, higher paid)
Separate facilities for each job category
Wards Cove v. Antonio
Key Points:
Evidence for adverse impact is useable to demonstrate a prima facie case (both lower
courts rejected a pattern or practice claim)
What comparison is allowed to show disparate impact?
Plaintiffs used the below disparities to form a prima facie case:
% minorities in unskilled skilled vs.
% nonminorities in skilled jobs
Discrepancies due to differences in
hiring procedures (e.g., nepotism,
word of mouth recruiting, walk-ins,
vague hiring criteria)
Supreme Court Decision in Wards Cove
A) Proper statistical comparison is:
• Racial composition of those in the noncannery (skilled) jobs with those in the
"relevant" labor market who are qualified and have an interest in performing the
job(s) in question
• Here, % majority vs. % minority in skilled jobs
Adverse Impact or Disparate Treatment???
• Why not consider Wards Cove as a disparate treatment case (i.e.,
pattern or practice)?
• In Wards Cove, Justice White cited Hazelwood as the prime example of
adverse impact ----- WRONG!
• In Wards Cove, the Court did not even mention Griggs or any other
adverse impact case!
Note: Recruitment is NOT considered to be a selection procedure (e.g., Uniform
Guidelines). Recruitment, specifically passive recruitment practices, were the reason for
the disparities in Wards Cove
B) Plaintiff must identify specific employment practice that caused the
statistical disparity (consistent with plurality in Watson v. Ft. Worth Bank,
1989)
"Here, respondents have alleged that several "objective" employment practices (e.g.,
nepotism, separate hiring channels, rehire preferences), as well as the use of
'subjective decision making' to select non-cannery workers had a disparate impact
on non-whites...Respondents will... have to demonstrate that the disparity they
complain of is the result of one or more of the employment practices that they
are attacking here, specifically showing that each challenged practice
has a significant disparate impact on employment opportunities."
Wards Cove v. Antonio (cont.)
C) Company recordkeeping responsibility
"...employers...are required to maintain...records or other information which will
disclose the impact which its tests and other procedures have upon employment
opportunities or persons by identifiable race, sex, or other ethnic groups...This
includes records concerning 'the individual components of the selection process'
where a statistical disparity in the selection rates of whites and
nonwhites...Plaintiffs as a general matter will have the benefit of these tools
to meet their burden of showing a causal link between challenged employment
practices and racial imbalances in the workforce..."
D) Burden of Proof
Regarding earlier court opinions on this issue: "...they should have been
understood to mean an employer's production -- but not persuasion burden."
Faced with a prima facie case the employer may now "articulate" rather than
"prove" that it had a legitimate business reason for its practice.
“The touchstone of this inquiry is a reasoned review of the employer’s
justification for his use of the challenged practice ... there is no
requirement that the challenged practice be ‘essential or
indispensable‘ to the employer’s business for it to pass muster; this
degree of scrutiny would be almost impossible for most employers to
meet.”
~ Adverse Impact Process After Wards Cove ~
Step 1: The challenger must identify a specific employment practice
that caused the discrimination in question (disproportionately
excludes protected group members)
Step 2: The company must demonstrate that the challenged practice
is job related and consistent with business necessity
Wards Cove: The company must PRODUCE evidence
(articulate -- not prove) that it had a legitimate reason for
their employment practice
Step 3: The challenger must prove that an equally valid, job-related
practice exists with less (or no) adverse impact
PURPOSES SEC. 3 [42 U.S.C. 1981 note]
The purposes of this Act are (1) to provide appropriate remedies for intentional discrimination
and unlawful harassment in the workplace; (2) to codify the concepts of "business necessity"
and "job related" enunciated by the Supreme Court in Griggs v. Duke Power Co., 401 U.S. 424
(1971), and in the other Supreme Court decisions prior to Wards Cove Packing Co. v. Atonio,
490 U.S. 642 (1989); (3) to confirm statutory authority and provide statutory guidelines for the
adjudication of disparate impact suits under title VII of the Civil Rights Act of 1964 (42 U.S.C.
2000e et seq.); and (4) to respond to recent decisions of the Supreme Court by expanding
the scope of relevant civil rights statutes in order to provide adequate protection to victims of
discrimination.
Earlier purpose statement:
S. 1745: "...to overrule the proof burdens and meaning of business necessity in Wards Cove
Packing Co. v. Atonio and to codify the meaning of business necessity used in Griggs v. Duke
Power Co...."
Previous job-relatedness definitions:
First Danforth-Kennedy proposal (Spring 1990): "substantial and demonstrable relationship to
effective job performance."
S. 2104 (original bill in 1990): "essential to effective job performance."
Bill passed by Senate in 1990: "significant relationship to successful performance of the job."
S. 1208: "manifest relationship to requirements for effective job performance”
DEFINITIONS SEC. 2000e. [Section 701]
(m) The term ``demonstrates'' means meets the burdens of production and
persuasion.
UNLAWFUL EMPLOYMENT PRACTICES SEC. 2000e-2. [Section 703]
In 1990 CRA, the
words “results in”
were used
Burden of Proof
(k) (1) (A) An unlawful employment practice based on disparate impact is established under this title
only if- (i) a complaining party demonstrates that a respondent uses a particular employment
practice that causes a disparate impact on the basis of race, color, religion, sex, or national origin
and the respondent fails to demonstrate that the challenged practice is job related for the position
in question and consistent with business necessity; or (ii) the complaining party makes the
demonstration described in subparagraph (C) with respect to an alternative employment practice
and the respondent refuses to adopt such alternative employment practice. (B) (i) With respect to
demonstrating that a particular employment practice causes a disparate impact as described in
subparagraph (A)(i), the complaining party shall demonstrate that each particular challenged
employment practice causes a disparate impact, except that if the complaining party can
demonstrate to the court that the elements of a respondent's decision making process are not
capable of separation for analysis, the decision making process may be analyzed as one
employment practice. (ii) If the respondent demonstrates that a specific employment practice does
not cause the disparate impact, the respondent shall not be required to demonstrate that such
practice is required by business necessity.
Adverse Impact Scenario Under CRA-91?
The term ``demonstrates'' means meets
the burdens of production and
persuasion.
Step 1: The challenger must demonstrate that a specific
employment practice that caused the discrimination in question
(disproportionately excludes protected group members)
… if the complaining party can demonstrate to
the court that the elements of a respondent's
decision making process are not capable of
separation for analysis, the decision making
process may be analyzed as one employment
practice.
Step 2: The company must demonstrate that the challenged
practice is job related for the position in question and consistent
with business necessity
Step 3: The challenger must prove that an equally valid, job-related
practice exists with less (or no) adverse impact
Uniform Guidelines --- Validity
Sec. 1607.5(A). Acceptable types of validity studies. For the purposes of
satisfying these guidelines, users may rely upon criterion-related
validity studies, content validity studies or construct validity studies,
in accordance with the standards set forth in the technical standards of
these guidelines, section 14 below. New strategies for showing the
validity of selection procedures will be evaluated as they become
accepted by the psychological profession. (TIP article on Guidelines
review)
Sec. 1607.C(1): A selection procedure based on inferences about mental
processes cannot be supported solely or primarily on the basis of
content validity. Thus, a content strategy is not appropriate for
demonstrating the validity of selection procedures which purport to
measure traits or constructs such as intelligence, aptitude, personality,
common sense, judgment, leadership and spatial ability [emphasis by
authors].
This view has been challenged by
many in the psychological
profession (e.g., Landy)
Most prefer a unitarian view of validity (any validity
evidence, however it is collected, can support the jobrelatedness of selection devices)
Content Validity and Decision-Making (Rank Ordering)
Use of content validity for rank ordering of candidates
from Guardians v. Civil Service (1980) --1. Suitable job analysis
2. Reasonable competence in test construction
3. Test content related to job content
4. Test content representative of job content
5. Scoring systems selecting applicants who are better job
performers
Gillespie v. Wisconsin (1985)
On the rejection of preferences for criterion-related validity
Q56: Why don't the Uniform Guidelines state a preference
for criterion-related validity over content or construct
validity?
A: Generally accepted principles of the psychological profession
support the use of criterion-related, content or construct validity
strategies as appropriate. American Psychological Association
Standards, E, pp. 25-26. This use was recognized by the Supreme
Court in Washington v. Davis, 426 U.S. 229, 247, fn. 13. Because
the Guidelines describe the conditions under which each
validity strategy is inappropriate, there is no reason to state a
general preference for any one validity strategy.
Gillespie v. Wisconsin (1985; cont.)
Q62: Under what circumstances may a selection procedure be used for
ranking?
A: … Use of a selection procedure on a ranking basis may be supported by
content validity if there is evidence from job analysis or other empirical
data that what is measured by the selection procedure is associated with
differences in levels of job performance.
Any conclusion that a content validated procedure is appropriate for
ranking must rest on an INFERENCE higher scores on the procedure are
related to better job performance.
The more closely and completely the selection procedure approximates
the important work behaviors, the easier it is to make such an
inference. ...
Cutoff Scores
Lanning v. SEPTA (1999)
• Key issue was the use of an aerobic test for selecting transit authority police officers (run
1.5 miles with full gear under 12 minutes)
• Led to adverse impact for females
 SEPTA did a job analysis demonstrating the importance of aerobic capacity to
safety of officers and the public
 SEPTA provided statistics showing a relationship between aerobic capacity and
number of successful criminal arrest and commendations for field work
Also, Lanning distinguished between:
• Job relatedness (validity evidence)
• Business Necessity (that cutoff score measures minimum qualifications needed)
Rejected by SIOP Principles and other courts (e.g., Bew v. Chicago; Int. Brotherhood v.
Mississippi Power (2006)
Isabel v. Memphis (2003)
[Endorsed view of business necessity from Lanning that a cutoff score must
measure minimum qualifications need for the job]*
Adverse impact on black applicants using job knowledge test
(promotion to Police Lieutenant)
• Test content did not represent total domain of KSAs for the job
• No data (e.g., via a job analysis) that the test differentiated between
qualified and unqualified candidates
* No real reason to bring up the Lanning decision – defense would likely have lost due to
failure to satisfy Step #4 from Guardians (that test content be representative of job
content)
4/5 rule is a guideline, a rule of thumb
Uniform Guidelines on Selection Procedures (1978)
“A selection rate for any race, sex, or ethnic group which is less than four-fifths
(4/5) (or eighty percent) of the rate for the group with the highest rate will
generally be regarded by the Federal enforcement agencies as evidence of
adverse impact, while a greater than four-fifths rate will generally not be
regarded by Federal enforcement agencies as evidence of adverse impact.
Smaller differences in selection rate may nevertheless constitute adverse
impact, where they are significant in both statistical and practical terms or
where a user's actions have discouraged applicants disproportionately on grounds
of race, sex, or ethnic group. Greater differences in selection rate may not
constitute adverse impact where the differences are based on small numbers and
are not statistically significant.”
~ Statistical Versus Practical Significance ~
What indicates a “meaningful” difference between 2 groups?
Statistical significance tests generally assess the level of confidence a that a
finding is not a chance event (e.g., Z test)
Practical significance measures generally assess the magnitude or consequences
of a finding (e.g., using 4/5th rule, effect sizes, odds ratio)
In Stagi v. Amtrack (2010) the 3rd Circuit noted that increased numbers make it more likely
to exclude chance as a cause of adverse impact. The court ruled that:
• Statistically significant results alone support causation
• The 4/5th rule was not persuasive
• There is no additional requirement of practical significance given the inference of
causation
This combination of ideas almost implies that statistical significance is practical
significance because a disparity is probably not due to chance.
Most recent OFCCP settlements have emphasized disparities using statistical
significance tests as standalone evidence.
From: Dunleavy. E. M., & Gutman, A. (2011). An Update on the Statistical Versus Practical Significance
Debate: A Review of Stagi v. Amtrak (2010). The Industrial-Organizational Psychologist, 48(4), 121-130.
Bew v. Chicago (2001)
N = 5,191 (Probationary Police Officers Certification Test)
Whites = 99.96
Blacks = 98.24 (No 4/5 Rule violation)
But --- 32/33 who failed were Black
Adverse Impact implied via test of independent proportions (5 SD difference in
failure rates)
Isabel v Memphis (2003)
4/5 Rule violation with cutoff score of 70; dropped cutoff score to “66”
Blacks = 74.6%
Whites = 89.5% (No 4/5 rule violation)
But, significant differences in mean scores (Blacks = 69.17 vs. Whites = 75.59
D = .9 (large, significant difference)
Apsley v. Boeing (2012): Statistical significance does not necessarily equal practical
significance
Statistically significant difference in hiring rates (older workers negatively impacted)
But, only 60 out of 9000 employees were affected (practically not a “big
deal”/meaningful)
Jones v. City of Boston (2014):
Drug Testing: White positive rate was 0.3% (30 people); for African Americans, it was
1.3% (55 people) – a 1% difference
The issue is “the extent to which we can be confident that the differences in the outcome,
whether large or small, were not random.”
Although only a 1% difference, the positive test rate for Blacks was several times that for
Whites
1)Showing statistical significance is not easy for plaintiffs
2)Job relatedness and business necessity defense for company
3)Difficult to prove an effective alternative with less adverse impact when the impact is
already tiny
Context Matters?
Why not a 2-stage process in determining adverse impact?
Step 1: A statistical significance test would as the 1st “hurdle supporting
that a disparity is probably not due to chance.
Step 2: Evidence that a non-chance disparity is substantial enough in size to
support a reasonable inference of discrimination
This two-hurdle process has NOT been observed in recent case law concerning
impact measurement.
Advice From a Technical Advisory Committee
(Created to help provide the assessment and equal employment opportunity
(EEO) communities with technical “best practice” guidance on how to
conduct adverse impact analyses and included 70 of the nation’s experts in
adverse impact analyses. Available at: http://cceq.org/
• Multiple methods of adverse impact detection should be used.
However, care should be taken to minimize redundancy and combine
only methods that each add unique information.
• Practical significance should be considered, and a variety of effect
sizes may be useful measures. Practical significance measures should be
paired with a statistical significance test.