Employment Regulation in the Workplace: Basic Compliance for Managers by Robinson, Franklin, and Wayland

1
Employment Regulation in the Workplace Basic
Compliance for Managersby Robinson, Franklin,
and Wayland

• Chapter 4
• Unintentional Discrimination Disparate Impact
• Spring 2009

2
Answers to Wonderlich
Question Answer Question Answer
1 3 15
0.31 2 3
16 e 3
2 17 4 4
Yes 18
4 5 4
19 3 6 2
20 F 7
4 21
1 8 1 22
3,5 9 5
23 31 10
40 24 2
11 3 25
1 12 .90
26 1500 13 4
27 2 14
3 28 1
3
Answers to Wonderlich
Question Answer Question Answer
29 3 41
1 30 1
42 1 31
1,5 43 6 32
31.85 44
2 33 1 45
15 34 3
46 8.40 35
20 47
1 36 .1 48
25 37 6
49 3,7 38
2 50
320 39 4 40 2,3,4
4
Griggs v. Duke Power Company

• Griggs v. Duke Power Company 401 U.S. 424 (1971)
• Five operating departments
• (1) Labor
• (2) Coal Handling
• (3) Operations
• (4) Maintenance, and
• (5) Laboratory and Testing

5
Griggs v. Duke Power Company

• Facially Neutral Selection Criteria
• High school diploma or a GED.
• Two aptitude tests.
• Wonderlich Personnel Test.
• Bennett Mechanical Comprehension Tests.

6
Griggs v. Duke Power Company

• High school diploma or a GED.
• 1960 North Carolina census, 34 of white males
  had completed high school, compared to only 12
  of black males
• Two aptitude tests.
• White candidates, 58 would pass.
• Black candidates, 6 would pass.

7
Griggs v. Duke Power Company

• Even though the same standards were administered
  to all applicants, the standards excluded a
  disproportionate number of blacks from favorable
  consideration.
• The Major Problem The requirements were not
  necessary to perform two of the job categories
  the coal handling and maintenance jobs.

8
Griggs v. Duke Power Company

• Employees who had not completed high school or
  had not taken the test (employees who had been
  hired before the new standards were implemented
  in 1965) had performed satisfactorily in the jobs
  which now required GED and passing test scores.

9
Disparate Impact

• Disparate Impact (unintentional discrimination)
• A facially neutral selection criterion has the
  effect of disqualifying a disproportionate number
  of protected class members.
• This is usually demonstrated by a manifest
  statistical imbalance.

10
Disparate Impact

• Prima Facie Case
• (1) Identify a specific employment practice.
• (2) Offer reliable statistical evidence the the
  practice causes the exclusion of sufficiently
  substantial number of applicants because of their
  membership in a protected group.
• (Wards Cove Packing Co. v. Atonio, 490 U.S. 642,
  657, (1989))

11
Employment Practices with Potential Disparate
Impact

• Height Requirements
• Weight Requirements
• Education Requirements
• Physical Agility Requirements
• Cognitive Ability Tests
• Language Requirements
• Arrest Records
• Conviction Record
• Marital Status
• Credit Reports

12
Methods for Establishing Statistical Imbalances

• Two or three standard deviations.
• Standard deviations are a measurement of the
  probability that a result is a random deviation
  from the predicted result.
• A 95 confidence interval.
• Four-Fifths Rule.

13
A 95 Confidence Interval

• A confidence interval gives an estimated range of
  values which is likely to include an unknown
  population parameter, the estimated range being
  calculated from a given set of sample data.

14
Four-Fifths Rule

• The EEOC OFCCP have adopted a rule of thumb
  under which they will generally consider a
  selection rate for any race, sex, or ethnic group
  which is less than four-fifths (4/5ths) or eighty
  percent (80) of the selection rate for the group
  with the highest selection rate as a
  substantially different rate of selection.
• (Source 44 Fed. Reg. 11,996).

15
Four-Fifths Rule
Selection rate Selection
rate of the protected lt .8 of the
group class
with the highest
rate
16
Four-Fifths Rule

• For example, if the hiring rate for Whites other
  than Hispanic is 60, for American Indians 45,
  for Hispanics 48, and for Blacks 51, and each
  of these groups constitutes more than 2 of the
  labor force in the relevant labor area, a
  comparison should be made of the selection rate
  for each group with that of the highest group
  (whites).
• These comparisons show the following impact
  ratios
• American Indians 45/60 or .75
• Hispanics 48/60 or .80
• Blacks 51/60 or .85
• Source 44 Fed. Reg. 11,996 (Mar. 2, 1979).

17
Four-Fifths Rule

• Should adverse impact determinations be made for
  all groups regardless of their size?
• Answer No. Section 15A(2) calls for annual
  adverse impact determinations to be made for each
  group which constitutes either 2 or more of the
  total labor force in the relevant labor area, or
  2 of more of the applicable workforce.
• For hiring, such determination should also be
  made for groups which constitute more than 2 of
  the applicants and for promotions,
  determinations should also be made for those
  groups which constitute at least 2 of the user's
  workforce.
• Note that there are record keeping obligations
  for all groups, even those which are less than
  2.
• Source 44 Fed. Reg. 11,996 (Mar. 2, 1979).

18
Four-Fifths Rule
Griggs v. Duke Power Co. Of the 181 White
applicants, taking the Wonderlich test, 105 pass.
Of the 102 Black applicants , 6 passed.
6 105 102 181
.06 lt .8 (.58) .06 lt .48
lt .8
or .06/.58 .103 which is less than .8
19
Test of Significance and Four-Fifths Rule

• Q Is it usually necessary to calculate the
  statistical significance of differences in
  selection rates when investigating the existence
  of adverse impact?
• A No. Adverse impact is normally indicated when
  one selection rate is less than 80 of the other.
  The federal enforcement agencies normally will
  use only the 80 (4/5ths) rule of thumb, except
  where large numbers of selections are made.
• Source Equal Employment Opportunity Commission,
  Uniform Employee Selection Guidelines
  Interpretation and Clarification (Questions and
  Answers).

20
Standard Deviation Rule
sx
(?) (1- ?) n
Where p proportion of one group N the number
selected
pn expected value (representation) µx 2sx
the confidence interval
21
Standard Deviation Rule
Standard Deviation Rule
Total Other Applicants
Total PG Applicants
Sx
Total Applicants Selected
X
X
Total Applicants
Total Applicants
22
Standard Deviation Rule
Of the 181 White applicants, taking the
Wonderlich test, 105 pass. Of the 102 Black
applicants , 6 passed.
Sx
102
181
X
111
X
283
283
23
Standard Deviation Rule
Of the 181 White applicants, taking the
Wonderlich test, 105 pass. Of the 102 Black
applicants , 6 passed.
Sx
102
181
Total applicants who passed
X
111
X
283
283

.36 X .64 X 111
of test takers who were black
25.57

5.057
24
Standard Deviation Rule
Standard Deviation Rule If the black applicants
(102) were selected at the same proportion that
they were represented in the test-taking (a total
of 283 applicants), we would expect 40 Blacks to
have been hired. Blacks were 36 of those taking
the test (102/283) from which 111 total
applicants passed. Hence 111 X .36 40
expected representation
25
Standard Deviation Rule
At 3 Sx we would expect the selection of blacks
to fall within a range of 25 to 55. 40 (5.057
X 3). Selecting only 6 falls more than six Sx
(6.72) from the expected representation.
40
25
55
6
-6sx
-3sx
3sx
26
Applicants with High School Diplomas - Race
Blacks Whites Total Applied
Passed Applied Passed Applied
Passed Male 110 37 146 81
256 118 Female 48 18
36 32 84 50 Total 158
55 182 113 340 168
(.348)
(.621)
.348 lt .8(.621) .348 lt .497 or .348/.621
.56 which is less than .8 Yes, Disparate Impact
27
Applicants with High School Diplomas - Sex
Blacks Whites Total Applied
Passed Applied Passed Applied
Passed Male 110 37 146 81
256 118 Female 48 18
36 32 84 50 Total 158
55 182 113 340 168
(.461)
(.594)
.594 lt .8(.461) Females have the highest
rate .594 lt .369 or .594/.461 1.29 which is
greater than than .8 No Disparate Impact
28
Applicants Passing Dexterity Test
Blacks Whites Total Applied
Passed Applied Passed Applied
Passed Male 37 37 81 76 118
113 Female 18 16 32
28 50 44 Total 55 53 113
104 168 157
(.96)
(.92)
. 96 lt .8 (. 92 ) note blacks had the highest
passing rate .96 lt .768 or .96/.92 1.04
which is greater than .8
No Disparate Impact
29
Hired-After 2nd Test
Blacks Whites Total Applied
Passed Applied Passed Applied
Passed Male 37 37 76
44 113 81 Female 16 16
28 28 44
44 Total 53 53 104 72
157 125
(1.0)
(.692)
Note 100 of Blacks were passing both tests
were hired This is compared to 69.2 of whites
passing both.
30
Bottom Line Statistics
Blacks Whites Total Applied
Hired Applied Hired Applied
Hired Male 110 37 146 44
256 81 Female 48 16
36 28 84 44 Total
158 53 182 72 340
125 (.335)
(.396)
.335 lt .8(.396) .335 lt .316 or .335/.396
.846 which is greater than .8
31
Practice Problem
Blacks Whites Total Applied
Passed Applied Passed Applied
Passed Male 110 47 146 71
256 118 Female 48 18
36 32 84 50 Total 158
65 182 103 340 168
Analyze using both the Four-Fifths Rule and Three
Standard Deviation Rule for Race and Gender
32
Disparate Impact

• Employers Rebuttal The criterion creating the
  imbalance is a business necessity (job-related).

33
Challenging Statistics

• Relevant labor market
• For statistical evidence to be probative, the
  statistical pool or sample used must logically be
  related to the employment decision at issue and
  the statistical method applied to the pool or
  sample must be meaningful and suitable under the
  facts and circumstances of the case.
• (Hazelwood School Dist. v. U.S., 433 U.S. 299,
  308, (1977))

34
Challenging Statistics

• Sample Size the sample size may be too small and
  thus the selection or rejection of a single
  individual would substantially affect
  proportional outcomes.
• (29 C.F.R. 1607.3D)

35
Challenging Statistics

• Sample Size
• The general convention suggests that
  correlational studies require at least 30
  subjects.
• In statistical analysis, the larger the sample
  size, the greater the degree of precision. The
  smaller the size, the greater the probability of
  random error.

36
Challenging Statistics

• Countervailing Statistics
• Stock Analysis (a measure of representativeness)
  
• Employers may choose to show that their internal
  workforce has more protected class members than
  one would expect in the relevant external labor
  market or
• Proportion of Protected Class in
  Proportion of Protected
• Employers Internal Workforce Class
  in the Relevant Labor Market
• The basis for utilization analysis in chapter 6

gt
37
Challenging Statistics

• Stock Analysis
• An employer has the following relevant labor
  market for laborers 116,000 individuals of whom
  13,950 are African-Americans, 10,000 are
  Hispanic, 2500 are of Asian ancestry and the
  remainder are nonHispanic white (89,580).
• Proportional estimates relevant labor market
  77.2 white, 12 Black, 8.6 Hispanic, and 2.2
  Asian.

38
Challenging Statistics

• Employees holding positions classed as laborers
  in the employers workforce are as follows
• 5400 total current employees in the
  classification.
• 950 are Black.
• 390 are Hispanic
• 65 are Asian
• 3995 are White

39
Challenging Statistics

• Employers workforce
• 5400 current employees.
• .176 or 17.6 are Black.
• .072 or 7.2 are Hispanic
• .012 or 1.2 are Asian
• .74 or 74.0 are White

40
Challenging Statistics

• Comparison of relevant labor market to employers
  workforce
• Black .176 lt .12 or .176/.12 1.467
• Hispanic .072 lt .086 or .072/.086 .837
• Asian .012 lt .022 or .012/.022 .545
• White .74 lt .772 or .74/.772 .959