38. CASE: A SOLUTION FOR ADVERSE IMPACT
I. OVERVIEW
A federal government agency has recently had complaints regarding the selection procedures for
one of its entry-level law enforcement jobs. Presently, applicants must successfully complete a written
test and an unstructured interview. Although their current sta1ng practice had been adopted ten years
ago, the agency had not recently checked the selection rates for adverse impact. From the Uniform
Guidelines, it is suggested that the four-45hs rule be used to iden)fy instances of adverse impact. A5er
finding evidence of adverse impact for both women and minori)es, Bob Santa (personnel specialist for
the agency) conducts a job analysis to ensure that the new selection procedures will meet the valida)on
requirements of the Uniform Guidelines. The job analysis method chosen was the critical incident
technique and the result was an inventory of critical behaviors for successful and unsuccessful
performance for this job. This led to the generation of knowledge, skills, abili)es, and other
characteristic (KSAO’s) and a new test and a structured interview. Finally, the agency considers the
appropriateness of content validity versus criterion-related validity in this situation. The topics covered
in this case include selection methods, job analysis, valida)on methods, and adverse impact in employee
selection.
II. OBJECTIVES
1. To recognize and iden)fy instances of adverse impact.
2. To demonstrate the importance of doing an appropriate job analysis with
documenta)on.
3. To understand the difference between content and criterion-related validity.
4. To demonstrate the importance of structure and job relatedness in the development
of the selection interview.
III. DISCUSSION
The agency, acknowledging that its sta1ng practice must be updated to reBect the Uniform
Guidelines has to critically evaluate its present selection procedures for adverse impact. Once adverse
impact has been found, the agency must react appropriately by either eliminating the adverse impact,
providing suitable validity evidence to jus)fy the adverse impact, or finding alternative selection
procedures with less adverse impact. The organization, in this case, decides that its practice require
modi4ca)on. The decision involves a job analysis, documenta)on, new and modi4ed procedures, and
validity evidence.
In reality, the decision between content and criterion-related validity was an interesting one. At
first it was thought that the selection procedures represented a sample of the observable work
behaviors. However, this evidence is generally based on expert judgment rather than empirical
evidence. Due to the type of job and the process used, it was thought that a predic)ve validity study
should be carried out. Why? Content validity is somewhat controversial. In some jobs where the test
(e.g., typing test) assesses a representative sample of observable work behaviors, the inferen)al leap
between judging what the selection procedure measures and the content of the job is likely to be small.
When using a semi-structured interview where some of the KSAOs (i.e., judgment) are not directly
83
Instructor’s Manual – Part 2 84
observable but inferred from the selection device, content valida)on may not really be appropriate.
Very few court cases have dealt with the content validity of the interview.
IV. ANSWERS TO CASE QUESTIONS
1. Is there any evidence of adverse impact against any race, sex, or ethnic groups?
Yes. Using the four-45hs rule:
Test Interview Overall
Using the four-45hs rule or 80 percent rule, there is adverse impact against blacks, Hispanics,
2. If the total selection process for a job has no adverse impact, should the individual
components of the selection process are evaluated for adverse impact?
According to the Uniform Guidelines (Sec)on 15A2), when the total selection process does not
have adverse impact, the user should collect and have available information regarding the individual
components only if the user has just recently eliminated adverse impact. Otherwise, information
3. Which type of valida)on would you use? Why? What are the differences between
content and criterion-related validity studies?
Ideally, you would want your selection technique to be content valid and then assess them for
criterion-related validity. However, in many situations, you may not have adequate criterion measures or
If there is a challenge to the selection process, the content validity strategy is generally
4. Evaluate the job analysis procedures used in this case? Is it necessary to do such a
thorough analysis?
Instructor’s Manual – Part 2 85
The critical incident technique is a behavioral technique used to establish an inventory of critical
5. If you are doing a criterion-related validity study, should your criterion be success in
training or on-the-job performance?
Generally when we think of criterion measures we think of an assessment of on-the-job
performance. In Washington vs. Davis (1975) the Supreme Court ruled that performance in a training
This teaching note was contributed by Ronald J. Karren, Department of Management, University of
Massachuse s, Amherst.
39. CASE: The Use of Social Media for Employee Recruitment and selection
I. OVERVIEW
The recent use of social media (i.e. Facebook, LinkedIn, etc) has created issues for both
employers and applicants involved in the recruitment and selection process. More
specifically, applicants need to be certain that the use of social media does not reduce their
chances for being recruited or selected for desirable position based on information they
provide through social media. The employer has to be certain that all legal requirements
during these processes are maintained even though information available through social
media becomes an additional criterion for both recruitment and selection. This case
provides an example of how younger applicants and students may not be aware of how their
use of social media might impede their ability to qualify for desired position.
II. OBJECTIVES
1. To iden)fy potential challenges applicants face when they do not carefully monitor their
social media accounts.
2. To indicate how one organization handled social media and how all organizations
might utilize these processes in the recruitment and selection process
III. DISCUSSIONS
All young people, students, and adults need to be aware of how their pos)ngs on social
media might either enhance or detract from their job search. For example, sexual content is
not appropriate in social media if the individual expects to be applying for position,
gradua)ng, utilizing internships, etc. Why some cultures may view this in a posi)ve manor,
the business world tends to be less tolerant. In other words, most employers are fairly
conserva)ve, and those doing the hiring tend to view certain types of pos)ng nega)vely. At
Instructor’s Manual – Part 2 86
the same )me, employers need to be sensi)ve to not using social media in a way that has a
disparate impact on certain protected subgroups.
IV. ANSWERS TO CASE QUESTIONS
1. What lessons can you draw from this case regarding what students, employers, and faculty can
use social media such as Facebook during the recruitment and selection process.
The use of the web and social networks can cause some interesting advantages and disadvantages for
both employers and applicants. More employers are looking at online journals and social networking
sites such as Facebook to discover potentially damaging personal information about all applicants, but
Recruiters and hiring managers use social networks to learn more about candidates and as an additional
means to help 4nd potential job applicants as job pos)ngs get passed throughout these networks.
Career Builder conducted a survey of 3,169 hiring managers and found that 22% screened job seekers
The two legal issues are disparate treatment and disparate impact. Disparate treatment involves
inten)onally treating members of a protected class differently than others. Disparate impact involves
2. What policies, if any, should employers develop concerning the use of social media for various
purposes including employee recruitment and selection.
When employers utilize job seeker information on social networks, LinkedIn may be a slightly safer
choice than Facebook. LinkedIn is designed as a professional social networking site. Its purpose and
All methods used to make employment decisions are considered selection procedures and are subject to
Instructor’s Manual – Part 2 87
posi)on. To avoid discriminatory practice, try to standardize your process and treat everyone the same.
If you reject job seekers based on information found on their profile, make sure to record that
40. EXERCISE: EVALUATING THE RECRUITING FUNCTION
I. OBJECTIVES:
1. To make students aware of the necessity of evaluating the efficiency and
effectiveness of various recruitment sources.
2. To provide students with prac)ce in data analysis, drawing conclusions, and planning
a strategy to remedy identified problems or de4ciencies.
3. To make students aware of the linkages among sta@ turnover, recruitment sources,
recruitment methods, and adequate sta1ng.
II. OUT-OF-CLASS PREPARATION TIME: 2 hours
III. IN-CLASS TIME SUGGESTED: 45 Minutes
IV. PROCEDURES: See Text
V. POSSIBLE SOLUTIONS/DISCUSSION:
The completed calculation for Form 2.5 based on data presented in Exhibit 2.9 are shown on
the following page. For example, the yield ratio for potentially qualified write-ins is 1.26 (53 divided by
42) whereas it is 1.00 (24 divided by 24) for university programs. In terms of average cost per nurse
hired, the most efficient recruitment processes are direct applications (walk-ins and write-ins), employee
referrals, journal ads, and university nursing programs. The other recruitment sources yield few nurse
applicants who are hired, thus resulting in an extremely high cost per nurse hired.
However, when effectiveness in terms of yield ratios, one year survival rates, and above average
rating are considered, only employee referrals and university programs continue to show signiticantly
be0er than average yield ratios (i.e., low ratios).
For all recruitment sources, 5.25 potentially qualified nurses needed to be recruited in order to
generate one nurse who accepted a posi)on at St. Vincent’s. In addition, 8.27 nurses had to be recruited
to yield one who would survive for one year and 13.65 had to be recruited to yield one who received an
above average performance appraisal ra)ng.
Possible answers to the question are as follows:
1. How would you evaluate the nurse recruiting strategy currently being used by the
hospital? Is the hospital using too few or too many recruiting sources? Why?
Instructor’s Manual – Part 2 88
The present nurse recruiting strategy is using too many recruitment sources which are yielding
very li0le in the way of viable job candidates. The hospital would do be0er to concentrate its
recruitment resources on fewer, but more effective, recruitment sources. More effort should be directed
toward Internet recruiting, increasing referrals from present employees, and recruiting from university
based nursing programs. In addition, Exhibit 2.10 indicates the present recruiter (Sister Mary Louise) has
not created a posi)ve impression among those applicants she has interviewed and has failed to
follow-up on interviews in a timely manner. Use of a different recruiter and particular attention to timely
follow-up seem in order.
2. If you presently feel the hospital is using too many recruitment sources, which ones
would you emphasize and why?
The hospital should emphasize recruiting through Internet recruiting, employee referrals and
university based nursing programs because effectiveness and e1ciencies are be0er as discussed above.
Costs per recruited nurse are lower due to the relatively low numbers which must be recruited to yield a
qualified nurse ready and willing to accept a posi)on.
3. What stage or stages in the recruitment process seem to be most amendable to
improvements? What specific improvements would you suggest to decrease the
yield ratios? Why?
There seems to be a signiticant gap between the numbers who are offered position and the
numbers who accept the posi)on. Only 52 of the 98 offered a posi)on actually accepted. Exhibit 2.10
has identified some of the major problems which may have caused such rejec)on of job offer. As
mentioned above replacement of the recruiter and )melier follow-up a5er interviews may help. The
one year survival rates (33 of 52) could also be improved.
Instructor’s Manual – Part 2 89
FORM 2.5
YIELD RATIOS AT EACH STEP IN THE RECRUITMENT PROCESS AND
RECRUITMENT COST PER NURSE HIRED
ST. VINCENT’S HOSPITAL
Recruitment Sources
potentially
qualified
Accepted
Interview
offered
Job
Accepted
Job
One-Year
Survival
Above-Avg.
rating
Avg. Cost Per
Nurse
Hired
1. Internet application
2. Walk Ins
1.15
2.00
1.20
2.125
1.58
1.33
1.81
2.00
1.75
3.00
2.40
1.00
54.52
300.00
3. Employee Referrals 1.08 1.85 2.60 3.25 4.33 6.50 100.00
4. Newspaper Ads 1.50 3.00 6.00 12.00 24.00 —— 375.00
5. Journal Ads 1.06 1.90 2.38 4.75 9.50 9.50 112.50
6. educational institution
Junior Colleges
Hospital-based Schools
University Programs
1.23
1.00
1.00
1.45
2.67
1.50
2.67
4.00
1.71
8.00
8.00
2.40
8.00
——
3.00
16.00
——
3.43
600.00
800.00
130.00
7. Private Employment Agency 1.00 1.13 1.80 4.50 4.50 9.00 2,000.00
8. Public Employment Agency 2.00 4.00 8.00 8.00 —— —— 300.00
9. Direct Mail 1.07 3.75 5.00 15.00 —— —— 450.00
10. Job Fair 1.86 2.60 4.33 13.00 13.00 13.00 900.00
11. State Nursing Assoc. meeting 1.00 1.75 2.33 —— —— —— ——
Averages for All Sources 1.24 1.87 2.79 5.25 8.27 13.65 $283.65
solution to this problem is primarily associated with what happens on the job rather than during the recruitment process. However, providing
recruits with “realistic job previews” may help to reduce some of the turnover.
Instructor’s Manual – Part 2 90
4. On the basis of data in Exhibit 2.10, as well as the “Background” section, what changes
(if any) in the recruitment process would you suggest? Why?
The hospital should change its recruiter, emphasize follow-up of interviews, recruit through
employees and universities, and provide realistic job previews for reasons noted above.
41. EXERCISE: SELECTION DECISIONS
I. OBJECTIVES:
1. To make students aware of the complex criteria often used to select candidates for
administra)ve position.
2. To help students develop skills in planning and implementing semi-structured
interviews.
3. To give students prac)ce in preparing for, par)cipa)ng in, or evaluating the selection
interview.
II. OUT-OF-CLASS PREPARATION TIME: 2 hours
III. IN-CLASS TIME SUGGESTED: 45 Minutes
IV. PROCEDURES: See text
V. POSSIBLE SOLUTIONS/DISCUSSION:
Possible criteria for Form 2.6 include previous experience as a hospital CEO, other relevant
administra)ve experience, evidence of “success” in previous administra)ve position, ability to
communicate, future career goals, knowledge of the hospital, job knowledge, general health, ability to
work with others, motivation, education, initiative, management philosophy, and future plans for this
hospital. Form 2.6 requires up to ten criteria and a weigh)ng (1-5) for each criteria. committee
members will need to select what they consider to be the most important criteria and then weigh each.
Such criteria should be developed on the basis of the job description given in Exhibit 2.13.
Once these criteria are determined, committee members need to develop an agreed-upon set of
question directed to each of the criteria (Form 2.7). Examples of such interview question are as
follows:
Please tell the committee how your previous work experience has prepared you for the
present posi)on. (Experience.)
Do you feel your previous experience related to this posi)on has been successful? Why?
Give examples. (Experience.)
Now that you’ve had an opportunity to study our situation, what changes would you like to
implement if you were selected? (Future plans.)
Do you have any physical condition that have interfered with job performance? (Health.)
A5er seeing our job description, what areas do you feel are your strongest and weakest
based on your past history? (Knowledge.)
Why do you feel you are best qualified for this posi)on? (Mo)va)on.)
How would you establish good communication throughout this hospital? (Communica)on.)
Instructor’s Manual – Part 2 91
The observer should 4ll out Form 2.8 for evaluating the success of the interviews with the two
candidates. The criteria are listed on Form 2.6 itself. The instructor may also wish to complete Form 2.8
during and immediately a5er the interviews in order to add to or amplify the observer’s comments to
the class. The instructor should focus his or her comments on whether the most signiticant criteria were
identified and properly weighted, whether the specific question asked revealed information relevant to
these criteria, and how well both the committee members and the candidate played their roles during
the interviews.