Work Samples and Simulations

With work samples, you present examinees with situations representative of the job for which
they’re applying, and evaluate their responses.80 Experts consider these (and simulations,
like the assessment centers we also discuss in this section) to be tests. But they differ from
most tests because they directly measure job performance.
Using Work Sampling for Employee Selection
The work sampling technique tries to predict job performance by requiring job candidates to
perform one or more samples of the job’s tasks. For example, work samples for a cashier may
include counting money.81
Work sampling has advantages. It measures actual job tasks, so it’s harder to fake answers.
The work sample’s content—the actual tasks the person must perform—is not as likely to be
unfair to minorities (as might a personnel test that possibly emphasizes middle-class concepts
and values).82 Work sampling doesn’t delve into the applicant’s personality, so there’s little
chance of applicants viewing it as an invasion of privacy. Designed properly, work samples
also exhibit better validity than do other tests designed to predict performance.
The basic procedure is to select a sample of several tasks crucial to performing the job, and
then to test applicants on them.83 An observer monitors performance on each task, and
indicates on a checklist how well the applicant performs. For example, in creating a work
sampling test for maintenance mechanics, experts first listed all crucial job tasks (like “install
pulleys and belts”). Four crucial tasks were installing pulleys and belts, disassembling and
installing a gearbox, installing and aligning a motor, and pressing a bushing into a sprocket.
Since mechanics could perform each task in a slightly different way, the experts gave
different weights to different approaches.
Figure 6-7 shows one of the steps required for the task installing pulleys and belts— “checks
key before installing . . .” Here the examinee might choose to check the key against (1) the
shaft, (2) the pulley, or (3) neither. The applicant performs the task, and the observer checks
off the score for the approach used.
Situational Judgment Tests
Situational judgment tests are personnel tests “designed to assess an applicant’s judgment
regarding a situation encountered in the workplace.”84 For example:
You are a sales associate at Best Buy in Miami, Florida. The store sells electronics, including
smart phones. Competition comes from other neighborhood retailers, and from online firms.
Many customers who come to your store check the product with you, and then buy it on
Amazon. As a sales associate, you are responsible for providing exceptional customer
service, demonstrating product knowledge, and maximizing sales. You get a weekly salary,
with no sales incentive. How would you respond to this situation?
Situation:
A customer comes to you with a printout for a Samsung Galaxy phone from Amazon. com,
and proceeds to ask detailed questions about battery life and how to work the phone, while
mentioning that “Amazon’s price is $50 less than yours.” You have been with this customer
for almost 30 minutes, and there are other customers waiting. You would:
1. Tell the customer to go buy the phone on Amazon.
2. Tell the customer to wait 20 minutes while you take care of another customer.
3. Tell the customer that the local Sprint Mobility dealer has the phone for even less than
Amazon.
4. Explain the advantages of similar phones you have that may better fulfill the buyer’s
requirements.
5. Ask your supervisor to come over and try to sell the customer on buying the Galaxy from
you.
Management Assessment Centers
A management assessment center is a 2- to 3-day simulation in which 10 to 12 candidates
perform realistic management tasks (like making presentations) under the observation of
experts who appraise each candidate’s leadership potential. For example, The Cheesecake
Factory created its Professional Assessment and Development Center to help select
promotable managers. Candidates undergo 2 days of exercises, simulations, and classroom
learning to see if they have the skills for key management positions.86
Typical simulated tasks include:
The in-basket. The candidate gets reports, memos, notes of incoming phone calls, e-mails,
and other materials collected in the actual or computerized in- basket of the simulated job he
or she is about to start. The candidate must take appropriate action on each item. Trained
evaluators review the candidate’s efforts.
Leaderless group discussion. Trainers give a leaderless group a discussion question and tell
members to arrive at a group decision. They then evaluate each group member’s interpersonal
skills, acceptance by the group, leadership ability, and individual influence.
Management games. Participants solve realistic problems as members of simulated
companies competing in a marketplace.
Individual oral presentations. Here trainers evaluate each participant’s communication skills
and persuasiveness.
Testing. These may include tests of personality, mental ability, interests, and achievements.
The interview. Most require an interview with a trainer to assess interests, past performance,
and motivation. Supervisor recommendations usually play a big role in choosing center
participants. Line managers usually act as assessors and arrive at their ratings through
consensus. 87 Assessment centers are expensive to develop, take longer than conventional
tests, require managers acting as assessors, and often require psychologists. However, studies
suggest they are worth it.88 For many years, studies suggested that cognitive ability tests did
a better job of predicting job performance then did assessment centers. A recent study found
assessment centers were superior.89
Situational Testing and Video-Based Situational Testing
Situational tests require examinees to respond to situations representative of the job. Work
sampling (discussed earlier) and some assessment center tasks (such as in-baskets) are
“situational,” as are miniature job training (described next) and the situational interviews we
address in Chapter 7.90
The video-based simulation presents the candidate with several online or computer video
situations, each followed by one or more multiple-choice questions. For example, the
scenario might depict an employee handling a situation on the job. At a critical moment, the
scenario ends, and the video asks the candidate to choose from several courses of action. For
example:
(A manager is upset about the condition of the department and takes it out on one of the
department’s employees.)
Manager: Well, I’m glad you’re here.
Associate: Why?
Manager: I take a day off and come back to find the department in a mess.
You should know better.
Associate: But I didn’t work late last night.
Manager: But there have been plenty of times before when you’ve left this department
in a mess.
(The scenario stops here.)
If you were this associate, what would you do?
a. Let the other associates responsible for the mess know that you took the heat.
b. Straighten up the department, and try to reason with the manager later.
c. Suggest to the manager that he talk to the other associates who made the mess.
d. Take it up with the manager’s boss.91
The Miniature Job Training and Evaluation Approach
Miniature job training and evaluation involves training candidates to perform several of the
job’s tasks, and then evaluating their performance prior to hire. Like work sampling,
miniature job training and evaluation tests applicants with actual samples of the job, so it is
inherently content relevant and valid.
For example, when Honda built an auto plant in Lincoln, Alabama, it had to hire thousands of
new employees. Working with an Alabama industrial development training agency, Honda
began running help wanted ads.
Honda and the agency first eliminated those applicants who lacked the education or
experience, and then gave preference to applicants near the plant. About 340 applicants per 6-
week session received special training at a new facility about 15 miles south of the plant. It
included classroom instruction, watching videos of current Honda employees in action, and
actually practicing particular jobs. Some candidates who watched the videos simply dropped
out when they saw the work’s pace and repetitiveness.
The training sessions served two purposes. First, job candidates learned the actual skills
they’ll need to do the Honda jobs. Second, the training sessions enabled special assessors
from the Alabama state agency to scrutinize the trainees’ work and to rate them. They then
invited those who graduated to apply for jobs at the plants. Honda teams, consisting of
employees from HR and departmental representatives, did the final screening.92
Realistic Job Previews
Sometimes, a dose of realism makes the best screening tool. For example, when Walmart
began explicitly explaining and asking about work schedules and work preferences, turnover
improved.93 In general, applicants who receive realistic job previews are more likely to turn
down job offers, but their employers are more likely to have less turnover and be more
resilient.94 The Strategic Context feature illustrates this principle.
Choosing a Selection Method
The employer should consider several things before choosing a particular selection tool (or
tools). These include the tool’s reliability and validity, its practicality (in terms of utility
analysis), applicant reactions, adverse impact, cost, and the tool’s selection ratio (does it
screen out, as it should, a high percentage of applicants or admit virtually all?).97 Table 6-1
summarizes the validity, potential adverse impact, and cost of several assessment methods.
The HR Tools discussion shows how line managers may devise their own tests.