Employment Law

1
Employment Law

• For many years I/O psychologists governed
  themselves.
• Civil Rights Movement
• Civil Rights Act - designed to reduce unfair
  discrimination against minorities.
• Title VII specifies unlawful employment
  practices.
• Employers may not hire or refuse to hire on the
  basis of the five protected groups.
• Organizations cannot indicate a preference for
  members of a particular group in employment or
  training advertisements.
• Protected groups race, sex, religion, color,
  national origin (later age).

2
Disparate Treatment

• Disparate Treatment - Refers to evidence that a
  member of a protected group is treated
  differently from other job applicants in the
  employment process.
• All job applicants should receive the same
  treatment with regard to selection methods and
  hiring standards.
• Singling out some applicants for different
  employment procedures is evidence of disparate
  treatment.
• Example, only female police officer applicants
  are given a physical abilities test.

3
Adverse Impact

• Adverse Impact - discrimination against members
  of a particular group.
• Example - an organization uses a selection
  measure that leads to the hiring of many more
  Whites than African-Americans.
• 80 or 4/5ths rule
• Adverse impact occurs if the selection ratio for
  any group of applicants is less than 80 of the
  selection ratio for another group.
• What happens if adverse impact is found to exist?
• The organization is obligated to use another
  selection method that does not produce adverse
  impact or to validate the test.
• Even if the test is valid, the organization is
  obligated to try and find a selection method(s)
  that dont produce adverse impact (or less) and
  have comparable validity.

4
Court Cases

• Griggs v Duke Power (1971)
• Determined that individuals who bring suit
  against a company do not have to prove that the
  companys employment test is unfair. The company
  has the burden of proof.
• Bakke v. University of California
• Reverse discrimination. Court ruled that Whites
  can be victims of discrimination as well as
  African-Americans.
• Watson v Fort Worth Bank
• Cost of alternative selection procedures must be
  considered in making decisions about selection
  methods.
• Wards Cove Packing Company v Antonio
• Shifted the burden of proof to the individual
  filing the suit.

5
Americans with Disabilities Act

• Extended protection to individuals with
  disabilities.
• Disability defined as, a physical or mental
  impairment that substantially limits one or more
  (of the) major life activities a record of such
  impairment or being regarded as having such an
  impairment.
• An employment test that screens out an individual
  because of factors related to his or her
  disability must be job-related and consistent
  with business necessity.
• Employers must provide disabled persons
  reasonable accommodation in being evaluated for
  employment and in the conduct of their jobs.
• Most accommodations are minor and cost under 50.

6
Recruitment

• Recruitment is the process of attracting people
  to apply for a job.
• Basic principle - the more applicants, the better
  chance there is of finding a strong applicant.
• Why is recruiting important?
• Influences the quality of applicants.
• The hiring process is expensive, want to do it
  right the first time.
• The process can influence whether an applicant
  decides to accept a job, their attitudes once in
  the company, and how they view the company (even
  if not hired).

7
Recruitment (cont.)

• To reduce turnover, a company should put forth
  both a good and an accurate impression.
• Realistic Job Preview (RJP)
• Attempts to give potential applicants a realistic
  view of the job and/or company.
• Might be a video tape, written information, or
  part of a visit.
• Mixed results
• Lower expectations
• Higher commitment (less turnover)

8
Affirmative Action

• Designed to correct previous and current
  inequities in workforce composition. It also
  provides role models and promotes diversity.
• Recruiting application - aiming recruiting
  efforts at audiences consisting of target groups.
• Preferential selection - organizations will
  select minority members if they are judged to
  have equal or better qualifications as
  nonminority members.
• Quotas - some organizations set aside a certain
  number of positions for minorities or base
  decisions on percentages of members in the
  organization.

9
Affirmative Action (cont.)

• Affirmative action is not required by law, but
  may be required in specific cases (e.g.,
  government contracts).
• Advantages
• Minority members are often underrepresented in
  the workforce, especially in certain jobs and
  positions. Affirmative action can balance the
  workforce.
• Disadvantages
• Research has found that affirmative action may
  have a negative effect on how others in the
  organization view an individual and the
  individuals view of him or herself (e.g.,
  self-esteem).
• Overall conclusion the effects of AA are complex.

10
Selection

• Regression
• A tool for predicting one variable (e.g.,
  criterion) from another (e.g., predictor).
• Typical regression equation
• Y a bX
• Y predicted criterion score (sales in ).
• X predictor score (extraversion score on the
  NEO).
• b slope of regression line (mathematical
  constant)
• a intercept of regression line (point where
  line crosses Y-axis)
• Example

11
Selection

• Multiple Regression
• Two predictors are better than one
• Regression using 2 or more predictors.
• Y a b1X1 b2X2
• Example
• When using multiple regression, it is best if
  your predictors relate strongly to the criterion
  but not to one another. You want as little
  overlap as possible between your predictors.
• It is difficult to find predictors that relate to
  the criterion but not one another (intelligence
  and personality have low intercorrelation which
  makes them good to use together.)

12
Selection Factors

• Selection Ratio
• Calculated by dividing the number of job openings
  by the number of applicants.
• A predictor is more valuable with smaller
  selection ratios. Smaller selection ratios allow
  you to be more picky.
• Base Rate
• Percentage of employees in the job deemed to be
  successful.
• Example - On an assembly line, workers must
  assemble a small portion of a toy at a rate of 2
  per minute. If they are slower, they will cause
  backups. A work sample indicates that 25 out of
  40 applicants are capable of this speed. Thus,
  the base rate is 25/40 which equals .625.
• A predictor is more valuable when the base rate
  is close to .50.

13
Selection Factors

• Criterion-related validity
• This is the validity of the predictor. In other
  words, the extent to which the predictor is
  related to (correlates with) the criterion.
• Example - a computer software company is
  evaluating a programming test and a personality
  test for selection. The company can only afford
  to administer one test. The programming test is
  correlated .60 with job performance, and the
  personality test is correlated .20 with job
  performance. Thus, the programming test has
  higher validity.
• The higher the validity, the better.

14
Selection Decisions

• If a predictor has less than perfect validity
  (correlation is less than 1.00), there will be
  errors.
• There are two types of errors
• People who are hired and wont perform well.
• People who are not hired and would have performed
  well.
• We can create a 2 x 2 matrix that compares what
  the predictor tells us and what the truth is in
  reality.

15
Use of Predictor Scores

• Top-down selection
• The number of people selected depends on the
  number of spaces that need to be filled. The
  individual with the highest score is selected
  first, and you work down until all the spaces are
  filled.
• Cutoff Scores
• Many organizations use cutoff scores to select
  applicants. Usually to ensure that all employees
  have a minimal level of competence.
• Most often, a combination of these two techniques
  is used.

16
Creating Cutoff Scores

• Predictor cutoff can be set by the criterion
  cutoff.
• Example, a small basket company determines that
  its workers should be able to make at least 20
  baskets in a day. Its selection test (e.g., a
  measure of motor skills) has the following
  relationship with basket-making Y 4 .5X.
  Therefore, the predictor cutoff is set at 32 (4
  .5(32) 20).
• Predictor cutoff can be determined for
  theoretical reasons.
• Example, a test given to salespeople is based on
  product knowledge. The company believes that in
  order to be successful, the salespeople should be
  able to answer all of the 30 most frequently
  asked customer questions. Thus, the company sets
  the cutoff at 30 out of 30.

17
Creating Cutoff Scores (cont.)

• A cutoff can be determined by the outcomes in
  will cause (e.g., the number of people who need
  to hired).
• Example - a company with very high turnover hires
  customer support personnel at regular 3 month
  intervals. They usually need about 10 new
  employees. A predictor cutoff of 70 means that
  (on average) 10 people are hired. So, the
  company sets the cutoff at 70.
• Banding
• A technique used to avoid adverse impact.
• Based on the premise that because tests arent
  completely reliable, people who score close to
  each other are equally skilled.
• Very controversial
• The government has prohibited the use of banding
  on some of its tests.

18
Multiple Predictors

• Selection Strategies for Multiple Predictors
• Multiple Regression
• 2 or more selection tests are weighted and added
  together.
• Remember Y a b1X1 b2X2
• This method assumes that the predictors can
  compensate for one another.
• Multiple cutoffs
• Applicants must score above a set level on each
  criterion.
• This method does not assume the predictors can
  compensate for one another.
• Rather, there is a minimum level on each
  predictor required for job performance.
• In this method, the predictors are all usually
  gathered and analyzed together (at the same time).

19
Multiple Predictors (cont.)

• Selection Methods for Multiple Predictors
• Multiple Hurdle
• Applicants must pass each test to continue the
  selection process.
• Advantages many applicants save time, because
  they are rejected without completing all the
  tests. The company can also save a lot of money
  by shortening the process for many applicants and
  also using least expensive tests up front.
• Civil service often uses this selection method.

20
Concurrent Predictive Validity

• Concurrent - predictor and criterion are assessed
  at roughly the same time.
• Less realistic - current employees are not
  motivated to pass.
• Range restriction - all current employees are
  performing the job and should score high on the
  predictor. You are probably sampling the top
  part of the distribution.
• Predictive - predictor is assessed at Time 1 and
  criterion is assessed at Time 2.
• Expensive.
• Time-intensive
• May not fix the range restriction problem.

21
Validity Generalization

• Does a good predictor in one situation also work
  well in another situation?
• Why do we care? Validation cost a great deal of
  money and requires large sample sizes. Sometimes
  it is not feasible to validate a test.
• So, we validate it based on previous findings.
• In general, the more similar the situations, the
  better a test will generalize.
• Some tests are more universal than others
  Intelligence tests vs. job knowledge tests.

22
Utility

• What (monetary) factors should a company take
  into account to determine whether or not to use a
  selection test or battery?
• Researchers have developed formulae for
  determining the economic utility of a test.
  Based mainly on the cost of turnover (e.g.,
  hiring new employees).
• Individuals in organizations actually are less
  likely to endorse a selection system if it is
  accompanied by economic utility information.
• My guess - this is due to the huge numbers given.

23
Summary

• Most important factors
• Cost of test development and administration.
• Test validity.
• Selection ratio.
• Base rate.
• Additional factors
• Changes in other outcomes (e.g., turnover) and
  their corresponding money values.
• Decreased training costs.
• Decreased accidents.
• Potential costs of using inappropriate tests
  (lawsuits, reduced number of people accepting the
  job, etc).