Human Resource Management

BS in Psychology 4th Year Credit Hours 3

Semester : Spring 2020

Lecture # 8 (Online)

Institute of Professional Psychology

Bahria University – Karachi Campus
Employee Testing and Selection
 Learning Outcomes
• Why is it important to test and select employees
• What is meant by reliability and validity of tests
• Basic categories of selection tests
• How to use work simulations for tests
• How to use various tools to select the best
candidate for the job
 Why employee selection is important
• After reviewing the applicants’ résumés, the manager turns to
selecting the best candidate for the job.
• This usually means reducing the applicant pool by using the
screening tools: tests, assessment centers, interviews, and
background and reference checks.
• The aim of employee selection is to achieve person-job fit.
• This means matching the knowledge, skills, abilities, and other
competencies (KSACs) that are required for performing the job
(based on job analysis) with the applicant’s KSACs.
• Of course, a candidate might be “right” for a job, but wrong for the
organization.
 Why employee selection is important
• For example, an experienced airline pilot might excel at American
Airlines but perhaps not at Southwest, where the organizational
values require that all employees help out, even with baggage
handling.
• Therefore, while person-job fit is usually the main consideration,
person-organization fit is important too.
• Selecting the right employee is important for three main reasons.
• First, employees with the right skills will perform better for you and
the company. Employees without these skills or who are abrasive or
obstructionist won’t perform effectively, and your own performance
and the firm’s will suffer.
• The time to screen out undesirables is before they are in the door.
 Why employee selection is important
• Second, effective selection is important because it is costly to recruit
and hire employees.
• One survey found that the average cost of hiring an employee who
doesn’t work out is about $50,000.
• Third, it’s important because inept hiring has legal consequences.
• We saw earlier that equal employment laws require nondiscriminatory
selection procedures.
• Negligent hiring is another such problem. It means hiring employees
with criminal records or other problems who then use access to
customers to commit crimes.
• During litigations, the court holds employers responsible for not
checking the background properly
 The Basics of Testing and Selecting Employees

• In this chapter, we’ll discuss several popular selection tools,

starting with tests.
• A test is basically a sample of a person’s behavior. Any test or
screening tool has two important characteristics:
• Reliability and
• Validity.
 RELIABILITY
• Reliability is a selection tool’s first requirement and refers to its
consistency:
• “A reliable test is one that yields consistent scores when a person takes
two alternate forms of the test or when he or she takes the same test
on two or more different occasions.”
• If a person scores 90 on an intelligence test on a Monday and 130 when
retested on Tuesday, you probably wouldn’t have much faith in the test.
• You can measure reliability in several ways.
• One is to administer a test to a group one day, re-administer the same
test several days later to the same group, and then correlate the first
set of scores with the second (called test-retest reliability estimates).
 RELIABILITY
• Or you could administer a test and then administer what experts
believe to be an equivalent test later; this would be an equivalent or
alternate form estimate.
• (The Scholastic Assessment Test [SAT] is one example.)
• Or, compare the test taker’s answers to certain questions on the test
with his or her answers to a separate set of questions on the same test
aimed at measuring the same thing.
• This is an internal comparison estimate.
• For example, a psychologist includes 10 items on a test believing that
they all measure interest in working outdoors, and then determines
the degree to which responses to these 10 items vary together.
 RELIABILITY
• Many things cause a test to be unreliable.
• These include physical conditions (quiet one day, noisy the next), differences
in the test taker (healthy one day, sick the next), and differences in test
administration (courteous one day, curt the next).
• Or the questions may do a poor job of sampling the material;
• for example, test one focuses more on Chapters 1 and 3, while test two
focuses more on Chapters 2 and 4.
• Because measuring reliability generally involves comparing two measures
that assess the same thing, it is typical to judge a test’s reliability in terms of
a reliability coefficient.
• This basically shows the degree to which the two measures (say, test score
one day and test score the next day) are correlated.
Test Score Correlation examples
 RELIABILITY
• Figure 6-1 illustrates correlation.
• In both the left and the right scatter plots, the psychologist compared each
applicant’s time 1 test score (on the x-axis) with his or her subsequent (time
2) test score (on the y-axis).
• On the left, the scatter plot points (each point showing one applicant’s test
score and subsequent test performance) are dispersed.
• There seems to be no correlation between test scores obtained at time 1
and at time 2.
• On the right, the psychologist tried a new test. Here the resulting points fall
in a predictable pattern.
• This suggests that the applicants’ test scores correlate closely with their
previous scores.
 VALIDITY
• Reliability, while indispensable, tells you only that the test is
measuring something consistently.
• Validity tells you whether the test is measuring what it’s
supposed to measure.
• It refers to the correctness of the inferences that we can make
based on the test.
• For example, if Jane’s scores on mechanical comprehension tests
are higher than Jim’s, can we be sure that Jane possesses more
mechanical comprehension than Jim?
 VALIDITY
• With employee selection tests, validity often refers to evidence
that the test is job related—in other words, that performance on
the test accurately predicts job performance.
• A selection test must be valid since, without proof of validity,
there is no logical or (under EEO law) legally permissible reason
to use it to screen job applicants.
• A test is a sample of a person’s behavior.
• A swimming test clearly corresponds to a lifeguard’s on-the-job
behavior.
• On the other hand, there may be no apparent relationship
between the test and the behavior.
A Slide from the Rorschach Test
 VALIDITY
• Thus, in Figure 6-2, the psychologist asks the person to interpret
the picture, and then draws conclusions about the person’s
personality and behavior.
• Here it is more difficult to prove that the tests are measuring
what they are said to measure, in this case, some trait of the
person’s personality— in other words, prove that they’re valid.
 VALIDITY
• There are several ways to demonstrate a test’s validity.
• Criterion validity statistically demonstrates a relationship between
scores on a selection procedure and job performance of workers.
• It means demonstrating that those who do well on the test also do
well on the job, and that those who do poorly on the test do poorly
on the job.
• The test has validity to the extent that the people with higher test
scores perform better on the job.
• In psychological measurement, a predictor is the measurement (in
this case, the test score) that you are trying to relate to a criterion,
such as performance on the job.
• The term criterion validity reflects that terminology
 Content Validity
• Content valid is a test that contains a fair sample of the tasks and
skills actually needed for the job in question.
• It is a demonstration that the content of a selection procedure is
representative of important aspects of performance on the job.
• For example, employers may demonstrate the content validity of
a test by showing that the test constitutes a fair sample of the
job’s content.
• The basic procedure here is to identify job tasks that are critical
to performance, and then randomly select a sample of those
tasks to test.
 Content Validity
• In selecting students for dental school, one might give applicants
chunks of chalk, and ask them to carve something like a tooth.
• If the content you choose for the test is a representative sample
of the job, then the test is probably content valid.
• Clumsy dental students need not apply. Subject matter experts
(SMEs, such as practicing dentists) help choose the tasks.
 Construct validity
• Demonstrating that:
• (1) a selection procedure measures a construct (an abstract idea
such as morale or honesty) and
• (2) that the construct is important for successful job
performance.
• At best, invalid tests are a waste of time; at worst, they are
discriminatory.
• Tests you buy “off the shelf” should include information on their
validity.
 Construct validity
• But ideally, you should revalidate the tests for the job(s) at hand.
• Tests rarely predict performance with 100% accuracy (or
anywhere near it).
• Therefore, don’t make tests your only selection tool; also use
other tools like interviews and background checks.
 Evidence-Based HR: How to Validate a Test
• Employers demonstrate evidence of a test’s validity using
criterion validity.
• For a selection test to be useful, you need evidence that scores
on the test relate in a predictable way to performance on the
job.
• Students who score high on the graduate admissions tests also
do better in graduate school.
• Applicants who score high on mechanical comprehension tests
perform better as engineers.
• Validate the test before using it by ensuring that scores on the
test are a good predictor of some criterion like job performance
— thus demonstrating the test’s criterion validity.
 Evidence-Based HR: How to Validate a Test
• An industrial psychologist usually conducts the validation study.
• HR coordinates the effort.
• Supervisor’s role is just to make sure that the job’s human
requirements and performance standards are clear to the
psychologist.
• But in practice, anyone using tests (or test results) should know
something about validation.
• Then you can better understand how to use tests and interpret
their results.
 Evidence-Based HR: How to Validate a Test
• The validation process consists of five steps:
• Step 1: Analyze the Job
• Includes; write job descriptions and job specifications.
• Aim here is to specify the human traits and skills required for job
performance.
• For example, must an applicant be verbal, a good talker?
• These requirements become the predictors, the human traits
and skills you believe predict success on the job.
 Evidence-Based HR: How to Validate a Test
• The validation process consists of five steps:
• Step 1: Analyze the Job
• For an assembler’s job, predictors might include manual
dexterity and patience.
• Manual dexterity is the ability to make coordinated hand and
finger movements to grasp and manipulate objects.
• Manual dexterity includes muscular, skeletal, and neurological
functions to produce small, precise movements.
• Development of these skills occurs over time, primarily during
childhood.
 Evidence-Based HR: How to Validate a Test
• The validation process consists of five steps:
• Step 1: Analyze the Job
• In this first step, also define “success on the job,” since it’s this
success for which you want predictors.
• The standards of success are criteria.
• You could use production related criteria (quantity, quality, and
so on), personnel data (absenteeism, length of service, and so
on), or worker performance (reported by supervisors).
 Evidence-Based HR: How to Validate a Test
• The validation process consists of five steps:
• Step 2: Choose the Tests
• Once you know the predictors (such as manual dexterity), next
step is to decide how to test for them.
• Employers usually base this choice on experience, previous
research, and “best guesses.”
• They usually don’t start with just one test. Instead, they choose
several tests and combine them into a test battery.
• The test battery aims to measure an array of possible predictors,
such as aggressiveness, extroversion, and numerical ability.
 Evidence-Based HR: How to Validate a Test
• Step 2: Choose the Tests
• What tests are available and where do you get them?
• Ideally, use a professional, such as an industrial psychologist.
• However, many firms publish tests.
• Some tests are available to virtually any purchaser, others only to
qualified buyers (such as with degrees in psychology).
• Wonderlic, Inc., publishes a well-known intellectual capacity test
and other tests, including aptitude test batteries and interest
inventories.
 Evidence-Based HR: How to Validate a Test
• Step 3: Administer the Test:
• One option is to administer the tests to employees currently on
the job.
• Compare their test scores with their current performance; this is
concurrent validation.
• Advantage is that data on performance are readily available.
• The disadvantage is that current employees may not be
representative of new applicants (who, of course, are really the
ones for whom you are interested in developing a screening test).
• Current employees already had on-the-job training and screening
by existing selection techniques.
 Evidence-Based HR: How to Validate a Test
• Step 3: Administer the Test:
• Predictive validation is the second and more dependable way to
validate a test.
• Here you administer the test to applicants before you hire them,
then hire these applicants using only existing selection
techniques, not the results of the new tests.
• After they’ve been on the job for some time, measure their
performance and compare it to their earlier test scores.
• Determine whether you could have used their performance on
the new test to predict their subsequent job performance.
 Evidence-Based HR: How to Validate a
Test4: Relate your Test Scores and Criteria:
• Step
• Ascertain if there is a significant relationship between test scores
(the predictor) and performance (the criterion).
• The usual method is to determine the statistical relationship
between
• (1) scores on the test and
• (2) job performance using correlation analysis, which shows the
degree of statistical relationship.
• If there is a correlation between test and job performance, you
can develop an expectancy chart.
 Evidence-Based HR: How to Validate a
Test4: Relate your Test Scores and Criteria:
• Step
• This presents the relationship between test scores and job
performance graphically.
• Split the employees into five groups according to test scores,
with those scoring the highest fifth on the test, the second
highest fifth, and so on.
 Evidence-Based HR: How to Validate a
Test4: Relate your Test Scores and Criteria:
• Step
• Compute the percentage of high job performers in each of these
five test score groups and present the data in an expectancy
chart like that in Figure 6-4.
• In this case, someone scoring in the top fifth of the test has a
97% chance of being a high performer, while one scoring in the
lowest fifth has only a 29% chance of being a high performer.
• Expectancy chart: A graph on the next slide showing the
relationship between test scores and job performance for a
group of people
Figure 6-4 expectancy Chart
 Evidence-Based HR: How to Validate a
Test
• Step 5: Cross-Validate and revalidate
• Before using the test, you may want to check it by “cross-
validating”—in other words, by again performing steps 3 and 4
on a new sample of employees.
• At a minimum, revalidate the test periodically.
• Some tests (such as the 16PF® Personality Profile) are
professionally scored and interpreted.
 Evidence-Based HR: How to Validate a
Test
• Step 5: Cross-Validate and revalidate
• Thus Wonderlic, Inc., lets an employer administer the 16PF. The
employer then faxes (or scans) the answer sheet to Wonderlic,
which scores the candidate’s profile.
• And faxes (or scans) back the interpretive report.
• Psychologists easily score many psychological tests online or
using interpretive Windows-based software.
• However, managers can easily score many tests, like the
Wonderlic Personnel Test, themselves.
 Evidence-Based HR: How to Validate a
Test
• Bias:
• Most employers know they shouldn’t use biased tests in the
selection process.
• A particular IQ test may provide a valid measure of cognitive
ability for middle-class whites but not for some minorities, if the
score depends on familiarity with certain aspects of middle-class
culture.
• Until recently, many industrial psychologists believed they were
adequately controlling test bias, but that issue is under review.
• Employers should therefore redouble their efforts to ensure that
the tests they’re using aren’t producing biased decisions
 Evidence-Based HR: How to Validate a
Test
• Utility Analysis
• Test predicted performance isn’t always of practical use.
• For example, if it is going to cost the employer $1,000 per
applicant for the test, and hundreds of applicants must be
tested, the cost of the test may exceed the benefits derived from
hiring a few more capable employees.
 Evidence-Based HR: How to Validate a
Test
• Utility Analysis
• Answering the question, “Does it pay to use the test?” requires
utility analysis.
• Two selection experts say, “Using dollar and cents terms, [utility
analysis] shows the degree to which use of a selection measure
improves the quality of individuals selected over what would
have happened if the measure had not been used.”
 Evidence-Based HR: How to Validate a
TestAnalysis
• Utility
• The information required for utility analysis generally includes:
• Validity of the selection measure,
• Measure of job performance in dollars,
• Applicants’ average test scores,
• Cost of testing an applicant,
• and the number of applicants tested and selected.
• Prudent employers endeavor to streamline their selection
processes, to minimize how long it takes to fill a position.
 Evidence-Based HR: How to Validate a
TestAnalysis
• Utility
• For example, with almost 60,000 job applicants per day, the U.S.
federal government was taking about 122 days to fill a position.
• By reviewing each step in its hiring process, it reduced time to
hire to about 105 days by, for instance, eliminating the applicant
essay.
• The accompanying HR as a Profit Center feature shows how
employers use tests to improve performance.
 Using Tests to Cut Costs and Boost
Profits
• Financial services firm Key Bank knew it needed a better way to
screen and select tellers and call-center employees.
• The company calculated it cost about $10,000 to select and train an
employee, but it was losing 13% of new tellers and call-center
employees within the first 90 days.
• That turnover number dropped to 4% after Key Bank implemented a
computerized virtual job tryout candidate assessment screening tool.
• “We calculated a $1.7 million cost savings in teller turnover in one
year, simply by making better hiring decisions, reducing training costs
and increasing quality of hires,” said the firm’s human resources
director.
 Using Tests to Cut Costs and Boost
Profits
• Outback Steakhouse has used pre employment tests almost from
when the company started. The testing seems successful.
• While annual turnover rates for hourly employees may reach
200% in the restaurant industry, Outback’s turnover ranges from
40% to 60%.
• Outback wants employees who are highly social, meticulous,
sympathetic, and adaptable.
• They use a personality assessment test to screen out applicants
who don’t fit the Outback culture.
 Using Tests to Cut Costs and Boost
Profits
• This test is part of a three-step pre employment screening
process.
• Applicants take the test, and managers then compare the
candidates’ results to the profile for Outback Steakhouse
employees.
• Those who score low on certain traits (like compassion) don’t
move to the next step.
• Those who score high are interviewed by two managers, who ask
behavioral questions such as “What would you do if a customer
asked for a dish we don’t have?