CHAPTER 5: EMPLOYEE SELECTION: REASONS FOR USING REFERENCES AND

RECOMMENDATIONS
REFERENCES AND TESTING
Confirming Details on a Resume

Resume fraud – the intentional placement of

PREDICTING PERFORMANCE USING REFERENCES
AND LETTERS OF RECOMMENDATION untrue information on a resume.

Reference check – the process of confirming the Checking for Discipline Problems
accuracy of resume and job application
Negligent hiring – A situation in which an
information. A reference check is the process of
employee with a previous criminal record
confirming the accuracy of information provided
commits a crime as part of his/her employment.
by an applicant.
Negligent-hiring cases are typically filed
Reference – the expression of an opinion, either
in court as common-law cases, or torts.
orally or through a written checklist, regarding an
applicant’s ability, previous performance, work Discovering New Information About the

habits, character, or potential for future success. Applicant

A reference is the expression of an Former employers and professors can

opinion, either orally or through a written provide information about an applicant’s work

checklist, regarding an applicant’s ability, habits, character, personality, and skills.

previous performance, work habits, character, or Opinion provided by any particular

potential for future success. The content and reference may be inaccurate or purposefully
format of a reference are determined by the person untrue.
or organization asking for the reference.
Reference checkers should always obtain
Letter of recommendation – A letter expression specific behavioral examples and try to get
an opinion regarding an applicant’s ability, consensus from several references.
previous performance, work habits, character, or
Predicting Future Performance
potential for success.
In psychology, a common belief is that the
Expressing an opinion regarding an
best predictor of future performance is past
applicant’s ability, previous performance, work
performance.
habits, character, or potential for future success.
The content and format of a letter of Validity coefficient – the correlation between
recommendation are determined by the letter scores on a selection method (e.g. interview,
writer. cognitive ability test) and a measure of job
performance (e.g. supervisor rating, absenteeism).

Corrected validity – a term usually found with

meta-analysis, referring to a correlation coefficient
that has been corrected for predictor and criterion
reliability and for range restriction; sometimes to a prospective employer about a former
called “true validity”. employee’s potential for legal trouble.

Low validity is largely due to four main The use of professional reference-checking
problems with references and letters of companies can help alleviate this problem.
recommendation: leniency, knowledge of the
Knowledge of the Applicant. The person
applicant, low reliability, and extraneous factors
writing the letter often does not know the
involved in writing and reading such letters.
applicant well, has not observed all aspects of an
Leniency. Most letters of recommendation applicant’s behavior, or both.
are positive.
Reliability – the extent to which a score from a
References typically are so test or from an evaluation is consistent and free
positive. But keep in mind that from error.
applicants choose their own
The lack of agreement between two people
references.
who provide references for the same person.
Although coworkers are willing to
Research reveals that reference reliability is
say negative things about
only .22
unsatisfactory employees,
confidentiality concerns can hold Thus, letters of recommendation may say

them back. more about the person writing the letter than about

A third cause of leniency stems the person for whom it is being written. This low

from the fear of legal level of reliability probably results from the point

ramifications. A person providing cited earlier that a reference writer has not seen all

references can be charged with aspects of an applicant’s behavior.

defamation of character (slander if Extraneous Factors – The fourth problem with

the reference is oral, libel if letters of recommendation concerns extraneous
written) if the content of the factors that affect their writing and evaluation.
reference is both untrue and made The method used by the letter writer is often more
with malicious intent. important than the actual content. A promising
Conditional Privilege, which approach to increasing the validity of references is
means that they have the right to to increase the structure of the reference check.
express their opinion provided they
ETHICAL ISSUES
believe what they say is true and
have reasonable grounds for this Raynes (2005) lists three ethical guidelines

belief. that reference providers should follow.

Negligent reference – an organization’s failure to First, explicitly state your relationship with
meet its legal duty to supply relevant information the person you are recommending.
Second, be honest in providing details.
 Finally, let the applicant see your reference knowledge of mechanical principles. Standardized
before sending it, and give him the chance job knowledge tests are commonly used by state
to decline to use it. licensing boards for such occupations as lawyers
and psychologists.
________________________________________
They often result in adverse impact and
PREDICTING PERFORMANCE USING APPLICANT
TRAINING AND EDUCATION can be used only for jobs in which applicants are
expected to have job knowledge at the time of hire
For many jobs, it is common that
or promotion.
applicants must have a minimum level of
education or training to be considered. ________________________________________

Better educated employees had higher PREDICTING PERFORMANCE USING APPLICANT

ABILITY
performance, were more likely to engage in
organizational citizenship behaviors, less likely to Ability tests tap the extent to which an
be absent, and less likely to engage in on-the-job applicant can learn or perform a job-related skill.
substance abuse. Ability tests are used primarily for occupations in
which applicants are not expected to know how to
Meta-analyses indicate that a student’s
perform the job at the time of hire.
GPA can predict job performance, training
performance, promotions, salary, and graduate Cognitive ability – abilities involving the
school performance. knowledge and use of information such as math
and grammar - includes such dimensions as oral
________________________________________
and written comprehension, oral and written
PREDICTING PERFORMANCE USING APPLICANT
expression, numerical facility, originality,
KNOWLEDGE
memorization, reasoning (mathematical,
Job knowledge test – a test that measures the deductive, inductive), and general learning -
amount of job-related knowledge an applicant important for professional, clerical, and
possesses. supervisory jobs, including such occupations as

Used primarily in the public sector, supervisor, accountant, and secretary.

especially for promotions. Similar to the exams Cognitive ability test – tests designed to measure
given several times a semester in a college class. the level of intelligence or the amount of

Given in multiple-choice fashion for ease knowledge possessed by an applicant.

of scoring, but they also can be written in essay Commonly used because they are excellent
format or given orally in a job interview. predictors of employee performance in the United

Common examples of job knowledge tests States and in Europe. Easy to administer, and are

include tests of computer programming relatively inexpensive.

knowledge, knowledge of electronics, and

 Predict work performance in two ways: by would handle each one. These situations cover
allowing employees to quickly learn job-related such topics (constructs) as leadership skills,
knowledge and by processing information interpersonal skills, personality tendencies,
resulting in better decision making. teamwork, and job knowledge.

Drawbacks. Perhaps the most crucial of Situational judgment tests correlated

these is that they result: highly with cognitive ability tests (r = .46).

High levels of adverse impact and Perceptual Ability

often lack face validity.
Measure of facility with such processes as
Frequently challenged in court.
spatial relations and form perception - consists of
Difficulty of setting a passing
vision (near, far, night, peripheral), color
score.
discrimination, depth perception, glare sensitivity,
Wonderlic Personnel Test – The cognitive speech (clarity, recognition), and hearing
ability test that is most commonly used in (sensitivity, auditory attention, sound localization)
industry. - useful for such occupations as machinist, cabinet
maker, die setter, and tool and die maker.
The short amount of time (12 minutes)
necessary to take the test, as well as the fact that it Psychomotor Ability
can be administered in a group setting.
Measure of facility with such processes as
Siena Reasoning Test (SRT). The finger dexterity and motor coordination - includes
developers of this test theorized that the large race finger dexterity, manual dexterity, control
differences in scores on traditional cognitive precision, multilimb coordination, response
ability tests were due to the knowledge needed to control, reaction time, arm-hand steadiness, wrist-
understand the questions rather than the actual finger speed, and speed-of-limb movement -
ability to learn or process information useful for such jobs as carpenter, police officer,
(intelligence). sewing-machine operator, post office clerk, and
truck driver.
To reduce the reliance on previous
knowledge, the SRT items contain nonsense Physical Ability
words and words that are commonly known (e.g.,
Physical ability tests – tests that measure
lamp, couch).
an applicant’s level of physical ability required for
SRT predict college grades and work a job - Often used for jobs that require physical
performance at least as well as traditional strength and stamina, such as police officer,
cognitive ability tests, but almost eliminated the firefighter, and lifeguard.
racial differences in test scores.
Physical ability is measured in one of two ways:
Situational Judgment Test. Applicants job simulations and physical agility tests.
are given a series of situations and asked how they
With a job simulation, applicants actually passing scores, and the time at which they should
demonstrate job-related physical behaviors. be required.

Job analyses consistently indicate that the physical

requirements of police officers can be divided into
Job Relatedness
two categories: athletic and defensive.
Many argue whether it is necessary to be
Athletic requirements are easy to simulate
physically fit. Two reasons for questioning the
because they involve such behaviors as
necessity of physical agility: current out-of-
running, crawling, and pulling.
shape cops and technological alternatives.
Defensive requirements are difficult to
safely and accurately simulate because Passing Scores

they involve such behaviors as applying Relative standards indicate how well an
restraining holds, kicking, and fending off individual scores compared with others in a
attackers. group.

9 BASIC PHYSICAL ABILITIES Advantage to using relative standards is

1. Dynamic strength (strength requiring that adverse impact is eliminated.

repetitions). Female applicant might be strong

2. Trunk strength (stooping or bending compared with other women, yet not strong
over). enough to perform the job.
3. Explosive strength (jumping or throwing
Absolute passing scores are set at the
objects).
minimum level needed to perform a job.
4. Static strength (strength not requiring
repetitions). _____________________________________
5. Dynamic flexibility (speed of bending, PREDICTING PERFORMANCE USING
stretching, twisting). APPLICANT SKILL
6. Extent flexibility (degree of bending,
Work Samples – an applicant performs actual
stretching, twisting).
job-related tasks.
7. Gross body equilibrium (balance).
8. Gross body coordination (coordination excellent content validity.

when body is in motion). excellent criterion validity

9. Stamina (ability to exert effort over long excellent face validity

periods of time). lower racial differences

can be expensive to both construct and
Because physical ability tests have tremendous
administer.
adverse impact against women, they have been
criticized on three major points: job relatedness, Assessment Centers - a method of selecting
employees in which applicants participate in
several job-related activities, at least one of which Research on the reliability and validity of the
must be a simulation, and are rated by several in-basket technique provides modest support for
trained evaluators. its usefulness.

Development and Components Simulation – An exercise designed to

place an applicant in a situation that is
The first step in creating an assessment
similar to the one that will be encountered
center is, of course, to do a job analysis.
on the job.
From this analysis, exercises are developed
Real backbone of the assessment center
that measure different aspects of the job. Common
because they enable assessors to see an applicant
exercises include the in-basket technique,
“in action.”
simulations, work samples, leaderless group
discussions, structured interviews, personality and To reduce the high costs associated with actual
ability tests, and business games. simulations, have developed situational exercises
on videotape. The development of simulation
exercises can be expensive, but prepackaged
Common Exercises included in Assessment exercises can be purchased at a much lower price.
Centers:
Work sample – A method of selecting
In-basket technique – An assessment employees in which an applicant is asked
center exercise designed to simulate the to perform samples of actual job-related
types of information that daily come across tasks.
a manager’s or employee’s desk in order to
Usually, when a simulation does not involve a
observe the applicant’s responses to such
situational exercise, it is called a work sample.
information.
Leaderless group discussions – an
The technique takes its name from the wire
exercise where applicants meet in small
baskets typically seen on office desks. Usually
groups and are given a job-related problem
these baskets have two levels: the “in” level,
to solve or a job-related issue to discuss.
which holds paperwork that must be handled, and
the “out” level, which contains completed No leader is appointed, hence the term
paperwork. leaderless group discussion. As the applicants
discuss the problem or issue, they are individually
The applicant is observed by a group of
rated on such dimensions as cooperativeness,
assessors, who score him on several dimensions,
leadership, and analytical skills.
such as the quality of the decision, the manner in
which the decision was carried out, and the order Business game – An exercise, usually
in which the applicant handled the paperwork. found in assessment centers, that is
designed to simulate the business and
 marketing activities that take place in an Biodata – A method of selection involving
organization. application blanks that contain questions about an
applicant’s life, school, military, community, and
Allow the applicant to demonstrate such
work experience of which research has shown will
attributes as creativity, decision making, and
predict future performance.
ability to work with others.

Evaluation of Assessment Centers

Biodata instruments have several advantages:
Research indicates that assessment centers
have been successful in predicting a wide range of  Can predict work behavior in many jobs,
employee behavior. May be excellent in including sales, management, clerical,
predicting certain aspects of employee behavior, mental health counseling, hourly work in
other, less expensive methods may be as good if processing plants, grocery clerking, fast-
not better. Assessment centers were once thought food work, and supervising.
to result in low levels of adverse impact, meta-  They have been able to predict criteria as
analysis results. varied as supervisor ratings, absenteeism,
accidents, employee theft, loan defaults,
________________________________________
sales, and tenure.
PREDICTING PERFORMANCE USING PRIOR
 Biodata instruments result in higher
EXPERIENCE
organizational profit and growth.
Experience Ratings  Easy to use, quickly administered,

The basis for experience ratings is the idea inexpensive, and not as subject to

that past experience will predict future experience. individual bias as interviews, references,
and résumé evaluation.
In giving credit for experience, one must
consider the amount of experience, the level of Development of a Biodata instrument

performance demonstrated during the previous File approach – The gathering of biodata from
experience, and how related the experience is to employee files on employee’s previous
the current job. employment, education, interests, and

Biodata demographics rather than by questionnaire.

Biodata is a selection method that With the file approach, we obtain

considers an applicant’s life, school, military, information from personnel files on employees’

community, and work experience. Meta-analyses previous employment, education, interests, and

have shown that biodata is a good predictor of job demographics.

performance, as well as the best predictor of

future employee tenure.
Questionnaire approach – The method of Derivation sample, is used to form the
obtaining biodata from questionnaires rather than weights; the other sample, the hold-out sample, is
employee files. used to double-check the selected items and
weights.
The major drawback to the questionnaire
approach is that information cannot be obtained Hold-out sample – A group of employees who
from employees who have quit or been fired. are not used in creating the initial weights for a
After the necessary information has been biodata instrument but instead are used to double-
obtained, an appropriate criterion is chosen. check the accuracy of the initial weights.

It is essential that a chosen criterion be The use of incumbents to create and scale
relevant, reliable, and fairly objective. items may reduce validity.

Criterion group – Division of employees into Criticisms of Biodata

groups based on high and low scores on a
The first holds that the validity of biodata
particular criterion.
may not be stable—that is, its ability to predict
Once a criterion has been chosen, employee behavior decreases with time, suggests
employees are split into two criterion groups that declines in validity found in earlier studies
based on their criterion scores. may have resulted from small samples in the
initial development of the biodata instrument.
Vertical percentage method – For scoring
biodata in which the percentage of unsuccessful The second criticism is that some biodata
employees responding in a particular way is items may not meet the legal requirements stated
subtracted from the percentage of successful in the federal Uniform Guidelines, which establish
employees responding in the same way. fair hiring methods.

Once employee data have been obtained To make biodata instruments less disagreeable to
and the criterion and criterion groups chosen, each critics, Gandy and Dye (1989) developed four
piece of employee information is compared with standards to consider for each potential item:
criterion group membership.
1. The item must deal with events under a
A problem with creating a biodata person’s control.
instrument is sample size. Large sample sizes are 2. The item must be job-related.
difficult if not impossible to obtain. In creating a 3. The answer to the item must be verifiable.
biodata instrument with a small sample, the risk of 4. The item must not invade an applicant’s
using items that do not really predict the criterion privacy.
increases.
Biodata can be faked, a charge that has been
Derivation sample – A group of employees who made against every selection method except work
were used in creating the initial weights for a samples and ability tests.
biodata instrument.
To reduce faking, several steps can be taken, (2) statistically based - Test is determined
including: through a statistical process called factor
analysis.
 warning applicants of the presence of a lie
(3) empirically based - Test is determined by
scale.
grouping answers given by people known
 using objective, verifiable items.
to possess a certain characteristic.
 asking applicants to elaborate on their
answers or to provide examples. Minnesota Multiphasic Personality Inventory-2
 By including bogus items. (MMPI-2) The most widely used objective test of
psychopathology.

________________________________________
Popularly known as the “Big Five” or the five-
PREDICTING PERFORMANCE USING
PERSONALITY, INTEREST, AND CHARACTER factor model, these dimensions are:

Personality Inventory – Psychological  openness to experience (bright,

assessment designed to measure various aspects of inquisitive);
an applicant’s personality.  conscientiousness (reliable, dependable);
 extraversion (outgoing, friendly);
Personality inventories fall into one of two
 agreeableness (works well with others, a
categories based on their intended purpose:
team player); and
measurement of types of normal personality or
 emotional stability (not anxious, tense).
measurement of psychopathology (abnormal
personality). That objective personality inventories are useful
in predicting performance in the United States and
Tests of normal personality
Europe is indicated in meta-analyses.
Tests of normal personality. Measure the
Though there is some disagreement across the
traits exhibited by normal individuals in everyday
various meta-analyses, probably the best
life. Determination of the number and type of
interpretation is that:
personality dimensions measured by an inventory
can usually be:  personality can predict performance at low
but statistically significant levels;
(1) based on a theory – Test is identical to the
 personality inventories can add
number postulated by a well-known
incremental validity to the use of other
theorist.
selection tests;
 Myers-Briggs Type Indicator has
 conscientiousness is the best predictor in
four scales.
most occupations and for most criteria; and
 Edwards Personal Preference
 the validity of the other four personality
Schedule, with 15 dimensions.
dimensions is dependent on the type of job
 and criterion for which the test is being provide the respondent with unstructured tasks
validated. such as describing ink blots and drawing pictures

One of the concerns about using personality
inventories for employee selection is that, because Rorschach Inkblot test – a projective personality
they are self-reports, they are relatively easy to test in which subjects’ perceptions of inkblots are
fake. people often act differently at work than they recorded and analyzed.
do in other contexts.
Thematic Apperception Test (TAT) – a projective
personality test in which test-takers are shown
pictures and asked to tell stories; it is designed to
measure various need levels.

Objective tests – A type of personality test that is

structured to limit the respondent to a few answers

“Aberrant” personality types and employee that will be scored by standardized keys.

behavior. Aberrant personality types are Minnesota Multiphasic Personality Inventory-2 -

“peculiarities that do not necessarily lead to a the most widely used objective test of
clinically impaired function (like personality psychopathology
disorders), but that may affect daily functioning
INTEREST INVENTORIES
(at work) in such ways that they deserve further
attention”. Interest Inventory – A psychological test
designed to identify vocational areas in which an
Three such aberrant types, referred to as the
individual might be interested.
“Dark Triad”:
Strong Interest Inventory (SII) – a popular
Machiavellianism.
interest inventory used to help people choose
Narcissism
careers.
Psychopathy
The theory behind these tests is that an
Tests of psychopathology
individual with interests similar to those of people
Determine whether individuals have in a particular field will more likely be satisfied in
serious psychological problems such as that field than in a field composed of people
depression, bipolar disorder, and schizophrenia - whose interests are dissimilar.
Generally scored in one of two ways: objectively
Other popular interest inventories include
or protectively.
the:
Projective tests – A subjective test in which a
 Minnesota Vocational Interest
subject is asked to perform relatively unstructured
Inventory.
tasks, such as drawing pictures, and in which a
 Occupational Preference Inventory
psychologist analyzes his or her responses,
  Kuder Occupational Interest honesty. They measure attitudes by asking the test
Survey taker to estimate the frequency of theft in society,
 Kuder Preference Record how harsh penalties against thieves should be,
 California Occupational Preference how easy it is to steal, how often he has personally
System been tempted to steal, how often his friends have
stolen, and how often he personally has stolen.
The relationship between scores on interest
inventories can be complex.  Personality-based integrity test – A type of
honesty test that measures personality traits
Vocational counselling – The process of helping
thought to be related to antisocial behavior *
an individual choose and prepare for the most
Shrinkage – the amount of goods lost by an
suitable career (helping people find the careers for
organization as a result of theft, breakage, or other
which they are best suited).
loss

If polygraph results are used, the

researcher is essentially comparing integrity test
scores with the scores of a test—the polygraph—
that has been made illegal for most employment
INTEGRITY TESTS
purposes partly because of questions about its
Integrity tests – Also called an honesty test; a accuracy. If self-admissions are used, the
psychological test to predict an applicant’s researcher is relying on dishonest people to be
tendency to steal. honest about their criminal history. If shrinkage is
used, the researcher does not know which of the
Electronic integrity tests:
employees is responsible for the theft or, for that
 Polygraph – an electronic test intended to
matter, what percentage of the shrinkage can be
determine honesty by measuring an individual’s
attributed to employee theft as opposed to
physiological changes after being asked questions
customer theft or incidental breakage. Even if
 Voice stress analyzer – an electronic test to
actual employee theft is used, the test may predict
determine honesty by measuring an individual’s
only employees who get caught stealing, as
voice changes after being asked questions.
opposed to those who steal and do not get caught.
Paper-and-pencil integrity tests:
Shrinkage – The amount of goods lost by an
 Overt integrity test – A type of honesty test organization as a result of theft, breakage, or other
that asks questions about applicants’ attitudes loss.
toward theft and their previous theft history.
Another drawback to integrity tests is that,
Based on the premise that a person’s not surprisingly, applicants don’t think highly of
attitudes about theft as well as his previous theft them.
behavior will accurately predict his future
Conditional Reasoning Tests
Conditional Reasoning Tests – Test designed to graphology is not a useful technique in employee
reduce faking by asking test-takers to select the selection.
reason that best explains a statement and to get a
________________________________________
more accurate picture of a person’s tendency to
PREDICTING PERFORMANCE LIMITATIONS DUE TO
engage in aggressive or counterproductive
MEDICAL AND PSYCHOLOGICAL PROBLEMS
behavior.
Drug Testing – Tests that indicate whether an
Provide test takers with a series of applicant has recently used a drug.
statements and then ask the respondent to select
Drug testing usually is done in two stages:
the reason that best justifies or explains each of
the statements. First, an employee or applicant provides a urine,
saliva, or hair sample that is subjected to an initial
Credit History
screening test. For urine or saliva, employers can
Employers conducting credit checks for at least use an on-site “instant” screener that provides
some jobs and employees with good credit are results in seconds or they can send the employee
more responsible and conscientious and thus will to a drug testing lab. Hair samples must be sent to
be better employees. a lab for testing. Although hair samples are less
convenient and more expensive than urine or
saliva samples, they have the advantage of
Graphology
measuring drug use occurring further back than is
Graphology – Also called handwriting analysis, the case for urine or saliva samples. If the initial
a method of measuring personality by looking at test for drugs is positive, then the second-stage
the way in which a person writes. testing is done by a medical review officer (MRO)

To analyze a person’s handwriting, a graphologist at a testing lab to ensure that the initial results

looks at the size, slant, width, regularity, and pressure were accurate.

of a writing sample. From these writing Psychological exams – Commonly used by

characteristics, information about temperament and employers to applicants in jobs involving public
mental, social, work, and moral traits is obtained. safety after a conditional offer of hire has been
Research on graphology has revealed interesting made, if an applicant fails the exam, the offer is
findings. First, graphologists are consistent in rescinded - Consists of an interview by a clinical
their judgments about script features but not in psychologist, an examination of the applicant’s
their interpretation about what these features. life history, and the administration of one or more
Second, trained graphologists are no more of the psychological tests mentioned above.
accurate or reliable at interpreting handwriting Psychological exams are not designed to
samples than are untrained undergraduates or predict employee performance. Therefore, they
psychologists. Most importantly, the small body should only be used to determine if a potential
of scientific literature on the topic concludes that employee is a danger to himself or others.
Medical Exams – Used to determine if there are Rejection letter – A letter from an organization to
any medical conditions that will keep the an applicant informing the applicant that he or she
employee from safely performing the job. will not receive a job offer.

________________________________________ Following guidelines are probably a good place to

start:
COMPARISON OF TECHNIQUES
Send rejection letters or emails to
Validity
applicants.
It is clear that the unstructured interview, Don’t send the rejection letter
education, interest inventories, and some immediately.
personality traits are not good predictors of future Be as personable and as specific as
employee performance for most jobs. It is also possible in the letter.
clear that ability, work samples, biodata, and Including a statement about the individual
structured interviews do a fairly good job of who received the job.
predicting future employee performance. that the Do not include the name of a contact
most valid selection battery includes a cognitive person.
ability test and either a work sample, an integrity
test, or a structured interview.

Legal Issues

Cognitive ability and GPA will result in

the highest levels of adverse impact, whereas
integrity tests, references, and personality
inventories will result in the lowest levels.
Interviews, work samples/simulations, and
résumés as being the most job-related/fair, and
they view graphology, integrity tests, and
personality tests as being the least job-related/fair.

________________________________________

REJECTING APPLICANTS