Understanding Behavior in the

Workplace

Blaine Landis, PhD

Course Overview
 Welcome!
Plan for Today:
- Benefits of taking this class
- A bit about me
- How to succeed in this class
- Communication and organizational
behavior! 
 Overview of the Assessments
Student presentations in seminars (20%)

Midterm exam after 5th class, before

reading week (30%)

Final exam held during the central exam

period between April 24-May 26 (50%)
 Student Presentations
• Your presentation must cover: (a) what the paper is
about (what important puzzle, problem, or question does
the paper seek to address?), (b) a brief summary of
relevant past research, (c) how the paper tells us
something new that we did not know before, and (d) how
the research could be applied in the workplace.
• The rating scale is as follows. You can earn 5 points for
each section, for a total score worth 20%.
1 2 3 4 5
Failed to cover Did so in a
this aspect of clear and
the article effective
manner
 Student Presentations
• Let’s use this article as an
example of how to
structure your
presentation…
• Adam Grant, an
organizational
psychologist at Wharton,
wanted to study whether
extraverts do in fact make
productive salespeople
(Grant, 2013)
 Student Presentations
• The article starts with a
statement of why the
question is important…
 Student Presentations
• Then continues with a
brief summary of past
research (i.e., what do we
already know?)…
 Student Presentations
• Then states the major
point of the paper (i.e., to
introduce a new
perspective on
extraversion and sales
performance—the
relationship is not linear,
but curvilinear)
 Student Presentations
• Finally, you should use
your own creativity in
thinking about how this
could be applied.
• Hiring: Are people too
extraverted for roles?
• Training and development:
Do we need programs to
address people who may be
overly assertive/talkative with
customers?
• Leadership: Should leaders
be aware of extraverts on
their sales teams?
 Student Presentations
• Public speaking is an incredibly important skill to hone,
but can also make people nervous
• It is normal to feel nervous! 
• You will not be penalized for displaying nervousness during your presentation
• The purpose of this assessment is for you to be able to practice your public
speaking skills in a psychologically safe environment
• Peer feedback will help you gain insight into your public speaking
skills, whereas your mark will be based on how well you cover the
points mentioned in the previous slides and on the syllabus
 Student Presentations
• Practice, practice, practice
• Do the presentation in front of friends/family members
and see if they can repeat the four major points back to
you
• You can use your phone or notes during the
presentation, but must not read from them
• The seminar leaders are there to support and encourage
you—we all want you to succeed!
• Feel free to ask them for tips and advice
 Midterm and Final Exams
• 10 true-or-false questions:
• True or false: Communication and behavior in
organizations is one of the most useful classes I will take
as an undergraduate at UCL.
• Then…
• Midterm exam: Two short essay questions
(approximately 500 words each)
• Final exam: Three short essay questions (approximately
500 words each)
• The essay questions will ask students to describe,
discuss, critique, and apply topics from organizational
behavior
 Midterm and Final Exams
• The essay questions will ask students to describe,
discuss, critique, and apply topics from organizational
behavior

Example question
• Select one trait from the Big Five model of personality.
Describe and discuss the major findings associated with
this trait. Next, offer a critique of the available evidence
and provide insight into how knowledge of this particular
trait can be applied in the workplace.
 Midterm and Final Exams
• Describe means telling us what something means.

• “Conscientiousness captures the extent to which people

vary in their tendency to be industrious, orderly, diligent,
and detailed oriented (McCrae & Costa, 1992). As a
broad trait, it subsumes narrow facets such as
orderliness and industriousness. The trait is highly
similar to grit (Ponnock et al., 2020)....”
 Midterm and Final Exams
• Discuss means telling us what you know about this
theory, concept, area, etc.

• “Conscientiousness is the best personality predictor of

job performance across jobs that have been studied
(Barrick & Mount, 1991; Tett et al., 1991). One of the
reasons why those with higher conscientiousness tend to
receive favorable performance ratings at work is due to
their desire to accomplish tasks and get ahead in
organizations (Barrick et al., 2002)…”
 Midterm and Final Exams
• Critique means offering an assessment of strengths and
weaknesses.

• “Because personality is a fixed characteristic that is not

amenable to experimental manipulation, the bulk of
evidence linking personality to job performance ratings is
observational. This raises the possibility that these
effects are due to unmeasured variables. Although
conscientiousness is strongly tied to job performance
across nearly all jobs, there are always exceptions. For
example, research shows that, in roles requiring
flexibility and spontaneity, such as jazz musicians,
having low conscientiousness can help (rather than hurt)
performance (Hogan, 1976)….”
 Midterm and Final Exams
• Apply means illustrating how it can be used in the
workplace.

• “The use of personality measures in organizational

settings presents key opportunities as well as
challenges. First, knowing one’s degree of
conscientiousness can be helpful in leadership
development, because leaders can use their knowledge
to their advantage. For example, leaders who have low
conscientiousness may benefit from delegating tasks
that require attention to detail and persistence to others
instead… Second, in hiring scenarios, employers can
select employees who are likely to be hardworking….”
 General Writing Tips
When writing essays, we highly recommend:
- Using a clear structure that separates
each of your main ideas/points (e.g., “First,
… Second, …. Finally, ….).
- Write using clear, concise language
- Providing in-text citations to reference your
work
- Avoiding obvious or overly general
statements
 Expectations
Excellent participation means:
• Contribute actively in all class discussions, exercises, and activities.
• Share relevant insights from readings and personal experience.
• Make succinct comments to move the discussion forward rather
than offer unrelated remarks.
• Clarify points that others may not understand.
• Demonstrate an excellent ability to apply, analyse, and synthesise
course material.
• Demonstrate willingness to take risks by answering challenging
questions or offering important insights from personal experience.
• Never miss class (except for an unavoidable emergency).
 Organizational Behavior
We are interested in the science of how people think, feel, and behave in organisations.
• How can we move from…
• Opinions from (even respected) authorities
• Guesses (even educated ones)
• Doing things because “that’s just the way they’ve been done in the past”
• To…
• Using data to inform our course of action and rule out alternative courses of action
• Critiquing dubious management practices and offering demonstrably useful alternatives
• To achieve these aims we will draw from:
• Psychology
• Sociology
• Political science
• Economics
• Anthropology
END
An Illustrative Case
Forecasting and Reducing Medical Errors

Blaine Landis, PhD

An Illustrative Case
Forecasting and Reducing Medical Errors
 Some of Medicine’s
Most Famous Findings

Relationship Correlation
Nicotine patch and quitting smoking .18
Ibuprofen and pain reduction .14
Antihistamine use and nose stops running .11
Chemotherapy and breast cancer survival .03
Smoking and lung cancer .08
Aspirin and reduced death by heart attack .02
Antibiotics and cured ear infections .08
 How Does OB Compare?
A Few Examples

Relationship Correlation
Conscientiousness and job performance .24
Grades and job performance .16
Pay incentives and individual productivity .32
Structured interview and predicting success .20
Top management commitment to “management .67
by objectives” (goal setting, participative
decisions, feedback) and organisational
productivity

Physical height and income .26

 But Can Organisational
Research Save Lives?
• Predicting medical errors:
• Amputating the wrong leg.
• Delivering medicine to the
wrong patient who has a
deadly allergy
• Finding
• Hospital teams with higher
trust make more medical
errors
• Your challenge
• Why?
 Edmondson’s Diagnosis
• High trust is costly because
it prevents doctors and
nurses from:
• Questioning each other
• Double-checking each
other’s work
• Need to intervene to
temper trust:
• Rotate teams
• Minimise socialising and
encourage only professional
interaction
 Questions
• What do you think of the diagnosis and
plan for intervention?
• What additional information do you need
to evaluate?
• What would you do if you were in the
researcher’s or consultant’s role?
• What did Edmondson do?

END
An Illustrative Case
Productivity at the Hawthorne
Works Plant
Blaine Landis, PhD
An Illustrative Case
Productivity at the Hawthorne
Works Plant
Early Days: The Hawthorne Studies
• From its founding in 1876, AT&T had a virtual monopoly
of the telephone industry.
• Western Electric, the manufacturing unit of the company,
produced telephones, cables, and other equipment.
• By 1929, over 40,000 worked at the Western Electric
Hawthorne Works Plant.
 The Hawthorne Studies
• The first set of behavioural science experiments were
carried out to examine the factors that influenced
productivity and satisfaction.
• One of the earliest studies examined the relationship
between lighting levels and productivity (1924–1927).
 The Hawthorne Studies
• Beginning in 1927, the next
experiments examined
productivity in the assembly
department.
• Workers were assembling
electromagnetic switches that
made telephone connections
possible.
• There were over seven million
relays produced annually.
• Overall production levels were
constrained by the speed of
individual workers.
 The Hawthorne Studies
But…
The research yielded little
insight into the factors that
actually affected productivity.
• Was it due to:
• Rest breaks?
• Working hours?
• Financial incentives?
• Working in a small group?
• Special attention the women
received?
 The Hawthorne Studies
• The Hawthorne Works Plant
invited Elton Mayo from Harvard
Business School to help them
understand the problem.
• Fritz Roethlisberger, his research
assistant, accompanied him.
• Together, the Hawthorne
experiments began to feature
extensive interviewing.
• Roethlisberger found that what
employees found most rewarding
were close relationships with one
another.
 The Hawthorne Studies
• The Hawthorne effect:
• People change their performance in response to
being observed.
• The Hawthorne studies:
• Provided an intimate portrait of the day-to-day working
conditions of people in the 1920s and 1930s.
• Illuminated ideas about:
• Motivation
• Job satisfaction
• Resistance to change
• Group norms
• Group participation
END
• Effective leadership
Metrics and Testing Our Ideas

Blaine Landis, PhD

Metrics and Testing Our Ideas
 Measurement
Measurement:
The bedrock of organisational behaviour.

If a thing exists, it exists in some amount;

and if it exists in some amount, it can be
measured.
—E. L. Thorndike (1914)
 Metrics
Data are everywhere…
• Electronic devices
• Emails
• Sociometric badges
• Job performance ratings
• Annual salary
• Number of promotions
• Perceptions of helping behaviour
• Self-reported job satisfaction
• Leadership evaluations
• Number of people who quit the company
• Job interview evaluations
 Measurement: Theory Testing
There are two basic models in organisational behaviour research.
1. There is the theoretical model, which concerns the logical
reasons why one variable would influence another variable.
(“Intelligent people have higher performance because they can
learn quicker and solve problems faster.”)
• Predictor variable: Intelligence → Outcome variable:
Job performance.
• Theory: Intelligence helps people learn and solve problems faster.
• The theory is the “why” – the logic linking the predictor to the outcome.
2. There is also the measurement model, which concerns the link
between the constructs and the measures.
• Construct: Intelligence → Measure: Wonderlic IQ test (score from
answering test questions).

Source: Edwards, J. R. (2003) Construct validation in organizational behavior research. In: Greenberg, J.
(Ed.) Organizational behavior: The state of the science. Mahwah, NJ: Erlbaum, 2nd ed., pp. 327–371.
 Measurement: Theory Testing
• The measurement model is important—theory testing is incomplete unless
the relationships between constructs and measures are scrutinised.
• A construct is a conceptual term used to describe a phenomenon of
theoretical interest.
• Constructs are terms researchers invent to describe, organise, and assign
meaning to phenomena relevant to a domain of research.
• Although constructs are literally constructed, or put together, by
researchers, the phenomena constructs describe are real and exist
independently of the researcher.
• Job satisfaction is very real to the people who love (or hate!) their jobs.
• A measure is an observed record or trace that serves as imperfect
empirical evidence of a construct.
• A measure is an observed score gathered through self-report, interview,
observation, or some other method.
END

Source: Edwards, J. R. (2003) Construct validation in organizational behavior research. In: Greenberg, J.
(Ed.) Organizational behavior: The state of the science. Mahwah, NJ: Erlbaum, 2nd ed., pp. 327–371.
Two Critical Concepts

Blaine Landis, PhD

Two Critical Concepts
 Measurement: Two Critical Concepts

We assess measures in two ways:

• Reliability refers to the consistency of a
measure.
• Examples: Internal consistency, inter-rater reliability,
test-retest reliability.
• Validity refers to the accuracy of a measure.
• Examples: Construct validity, content validity, criterion
validity, divergent validity, convergent validity.

Source: Edwards, J. R. (2003) Construct validation in organizational behavior research. In: Greenberg, J.
(Ed.) Organizational behavior: The state of the science. Mahwah, NJ: Erlbaum, 2nd ed., pp. 327–371.
 Reliability
Example:
• Richard and Hannah own several pubs
around London. To hire new staff, they
require that at least three managers
interview new applicants and make a
decision about whether a person should
be hired. However, the managers often
disagree about whom to hire.
What could be happening?
 Validity
Example:
• Consultants from BCG have been asked
to assess employee satisfaction—but the
only time to deliver the survey is before
the workday begins at 8:30 am. Results
reveal that satisfaction scores are
significantly lower than the industry
average.
What could be happening? END
How to Study People at Work
Key Concepts

Blaine Landis, PhD

How to Study People at Work
Key Concepts
 Key Concepts
• An independent variable is the variable
that is changed or controlled in a scientific
experiment to test its effects on the
dependent variable.
• A dependent variable is the variable
being tested and measured in a scientific
experiment.
 Key Concepts
• In observational studies, we usually refer
to the independent variable as the
predictor variable, because it’s not being
changed or controlled, only measured.
• Likewise, we refer to the dependent
variable in observational studies as the
outcome variable or criterion variable.
 Key Concepts
• Variable: A quantity that can take on different
values.
• Continuous variable: One that can theoretically
assume an infinite number of values between
any two points on the scale (e.g. salary).
• Binary variable: One that can only assume two
values (e.g. present or absent).
• Categorical variable: One that can assume
only a relatively small number of possible values
(e.g. department, ethnicity).
 Key Concepts
• Internal validity refers to the validity
(accuracy) of assertions regarding the
effects of the independent variable(s) on
the dependent variable(s).
• Put simply:
• Is what has taken place (i.e. the phenomenon
observed) due to the variables the researcher
claims to be operating (i.e. the measured or
manipulated variables), or can it be attributed
to other variables?
 Key Concepts
• External validity refers to the
generalisability of findings to or across
target populations, times, and the like.
• Put simply:
• Does what happened in my study apply in
other settings or contexts too?

END
How to Study People at Work
Common Approaches

Blaine Landis, PhD

How to Study People at Work
Common Approaches
 Research Methods: Experiments

• Experiments are studies in which at least one

variable is manipulated and people are
randomly assigned to the different conditions.
Do goals affect productivity?
Control condition: Experimental Experimental
no instructions condition: tell condition: tell
given to employees to do employees to set a
employees their best specific,
challenging goal
3.00 (1.56) 4.85 (1.44) 8.43 (1.47)

These are the average

productivity scores for
each group.
 Research Methods: Experiments
Here’s the logic behind an experiment:
• When we manipulate one variable and hold other
variables constant, we can be confident that the change
in the dependent variable is due to our manipulation,
because it was the only thing that changed between
conditions.
• When we randomly assign people to conditions, we
assume that individual differences (any differences
between people that may affect their scores on the
dependent variable) will wash out (i.e. they will be
roughly evenly distributed).
 Research Methods: Experiments
• Considerations in Experimental Research #1:
Setting.
• Laboratory (e.g. controlled environment, such as lab
space in the university) versus field (e.g. uncontrolled
office environment).
• For example, will there be noise or influence from
uncontrolled variables in a field experiment (e.g.
coworkers, customer interruptions) that may interfere
with my ability to examine whether there is a
significant effect?
 Research Methods: Experiments
• Considerations in Experimental Research #2:
Realism
• Artificiality versus Realism.
• Will my manipulation of the independent variable and
my measurement of the dependent variable
generalise to other settings?
• For example, do creativity tasks reflect real-world
creativity?
 Research Methods: Experiments
• Considerations in Experimental Research #3:
Manipulation Strength
• Too strong versus too weak.
• How strong does my manipulation have to be to test
whether there is a significant effect?
• For example, how much money would I have to pay
employees to see if money motivates them?
 Research Methods: Experiments
• Considerations in Experimental Research #4:
Manipulation Checks
• How can I check to see whether my manipulation
worked as intended?
• But be careful: checking the manipulation itself can
have an effect on the outcome!
 Research Methods: Experiments
• Considerations in Experimental Research #5:
Pre-Registration
• Telling people what you’re going to do before you do it
means that you have to stick to the protocol.
• This increases the credibility of your work and
reduces the likelihood that researchers will be
tempted to “play” with the data (e.g. develop
hypotheses after results are known).
 Research Methods:
Benefits and Drawbacks of Experiments
Benefits Drawbacks
• Provides causal evidence • It is sometimes difficult,
concerning the effect of the unethical, or impossible to
independent variable on the effectively manipulate certain
dependent variable. variables (e.g. intelligence).
• Gold standard in • Doesn’t demonstrate what
organisational research, “does” happen in a naturally
although not as frequently occurring environment.
used as one may think. • Individual differences between
• Demonstrates what “can” groups are not always evenly
happen (vs. what “does” distributed.
happen).
• Strong internal validity, and
possibly high external validity
(especially for field
experiments).
Research Methods: Quasi-Experiments

• In a true experiment, participants are

randomly assigned to either the treatment or
the control group, whereas they are not
assigned randomly in a quasi-experiment.
• Example: Daylight saving time and worker
accident rates.
 Research Methods:
Benefits and Drawbacks of Quasi-Experiments

Benefits Drawbacks
• Provides evidence suggesting • It raises questions about
possible causality concerning whether the lack of random
the effect of the independent assignment means that the
variable on the dependent conditions differ in other
variable. meaningful ways.
• Still does not provide causal
• Useful when it is difficult to
evidence, because the
randomly assign people to
experiment lacks random
conditions (e.g., workers in
assignment.
different parts of the company,
or in different offices).
• Typically high external validity,
but lower internal validity than
experiments.
 Research Methods:
Observational/Non-Experimental Research
• Observational studies are those in which we lack both
control over the manipulation and random assignment
to condition.
• Predictor and outcome variables are measured
(or “observed”).
• Such designs are common when we can’t manipulate or
randomly assign for ethical or practical reasons
(e.g. height).
• Susceptible to various threats to internal validity, but may
have high external validity due to the fact that variables
are measured or observed in natural settings.
 Research Methods:
Observational/Non-Experimental Research
• Example:
• Height and career success: Can we manipulate a
person’s height?
• Tall people are more likely to receive high supervisor
ratings, attain leadership roles, and have high
salaries.
• Over a 30-year career, an individual who is six feet tall
earns, on average, $166,000 more than an individual
who is 5'5".
• Tall people tend to have higher:
• Self-esteem: greater personal confidence. END

• Social esteem: greater respect from others.

Analytics
Basic Logic

Blaine Landis, PhD

Analytics
Basic Logic
 Data and Statistics: The Basic Logic

• In an ideal world, we could run the experiment or study

on everyone, at all times, under all conditions, but this
just isn’t possible or feasible.
• Instead, what we do is take a sample from the
population, which we hope generalises to the broader
population of interest.
• In statistics, the population is the complete and well-
defined set of people we’d ideally like to study, and the
sample is the subset of the population we can actually
study.
• We then compute statistics to describe and make
inferences about the larger group (population) on the
basis of observations made on a smaller group (sample).
 Sampling
When we collect data from a sample, we primarily
think about three things:
• Randomness: Every person in the population
has an equal chance of being selected.
• Size: Bigger is usually better! Sample size
affects statistical power—the ability to detect a
relationship if one actually exists.
• Representativeness: With large, randomly
collected samples, we hope our sample is
representative (i.e. a good match) to the larger
population of interest.
 The Role of Statistics
Two types of statistics in organizational behaviour:
1. Descriptive statistics: Procedures for organising, summarising, and
displaying data.
 The Role of Statistics
Two types of statistics in organisational behaviour:
2. Inferential statistics: Procedures by which inferences are made to a larger
group (population) on the basis of observations made on a smaller group
(sample).

END
Analytics
Introducing Analysis of Variance
and Regression
Blaine Landis, PhD
Analytics
Introducing Analysis of Variance
and Regression
 Commonly Used Inferential Statistics

T-test or analysis of variance Correlation or multiple

(ANOVA) regression
• I want to know whether one • I want to know whether there is
group has a significantly higher a statistically significant
or lower score than the other relationship between one (or
groups. more) variable(s) and another
• Statistically significant variable.
essentially means “beyond • Statistically significant
what we would expect by essentially means “beyond
chance” (usually less than a what we would expect by
5% chance). chance” (usually less than a
5% chance).
 Commonly Used Inferential Statistics

T-test or analysis of variance Correlation or multiple

(ANOVA) regression
• Example: I ran an experiment • Example: I conducted a study
where I manipulated whether of executives where I
employees set goals (asked measured intelligence, self-
employees to set challenging confidence, and motivation,
goals vs. a control condition) and I want to know if these
and measured their variables predict (i.e. are
performance at the end of the significantly related to) their
month. Did people in the goal- performance. Do any of these
setting condition have higher variables predict job
performance than the people performance?
in the control condition?
 Two Important Terms:
Moderation and Mediation
• Moderation refers to a situation in which one variable (a moderator) affects
the strength or direction of a relationship between two other variables.
• Put simply, it answers the question: When/under what conditions does a
particular effect occur?
• Example: The relationship between intelligence and job performance is
moderated by job complexity—intelligence is more strongly related to job
performance when job complexity is high.
 Two Important Terms:
Moderation and Mediation
• Mediation refers to the process by which one variable affects another
variable.
• Put simply, it answers the question: How does X explain Y? (Answer:
Because of Z, the mediator!)
 Spot the Issue
Researchers at McKinsey and Co. find that individuals with
upper class backgrounds tend to perform better in jobs
requiring analytical reasoning. In a sample of managers
working in a variety of industries, they examined whether
people whose parental income exceeded the top 1% of
salaries in the United Kingdom tended to perform better at
work. In their analysis, upper class people were 34% more
likely to have higher performance ratings in their latest
review. The researchers concluded that societal class
causes higher performance.
 Spot the Issue
A new study finds that students (n = 15) who slept
fewer than six hours per night while at university
tend to have higher salaries five years after
graduation. Researchers concluded that sleep,
despite its benefits, is clearly overrated when it
comes to job success.
 Spot the Issue
Do people who check their email first thing in the
morning tend to lead happier lives at work? To find
out, researchers assessed job satisfaction in two
departments at Dell Computers: accounting and
human resources. The human resources
department was instructed to check their email in
the morning, whereas the accounting department
was instructed to check their email after 12 noon.
Results revealed that the human resources
department was significantly happier than the
accounting department.
 Spot the Issue
When it comes to personality, what are the
characteristics of success? Researchers
hypothesised that people who are highly
cooperative tend to have higher performance at
work. They asked people, “How cooperative are
you with others in your workplace?” Respondents’
answers to this question were compared with their
most recent performance evaluation. Results
showed that agreeable people were in fact
better performers.
END
Understanding
Performance at Work

Blaine Landis, PhD

Understanding
Performance at Work
What Does Performance Really Mean?

• Performance: The • Task Performance:

“aggregated value to The “proficiency with
the organization of which job incumbents
the discrete perform activities that
behavioral episodes are formally
that an individual recognized as part of
performs over a their jobs…”
standard interval of
time.”
Source: Motowildo, Borman & Schmit, 1997, p. 72 Source: Motowildo, Borman, & Schmit, 1993, p. 73
 Performance Management Issue #1

Performance reviews can have the potential do

more harm than good
• The leader has not been giving regular
feedback, so the praise or correction has
already either been resolved or is in the distant
past
• Many industries are filled with “accidental
managers”—their actual job involves doing
something else, but they put on their supervisor
hat once a year to do a performance review
 Performance Management Issue #2

Strong evidence about alternatives to

performance reviews is lacking
• Because performance reviews are often bad or
unpleasant, many companies have
experimented with using them sparingly or
eliminating them altogether
• The jury is still out
• Simplifying performance reviews and making
them flexible and relevant to the work being
done does work
 Performance Management Issue #3

Effective coaching requires understanding an

employee beyond their performance numbers
and the limited observations made by
managers – and that means having real
conversations
• If only an annual or semi-annual performance
chat was all it took to fix things or motivate
someone…
• Real improvement comes from having regular,
constructive conversations with employees
about how to improve performance
 How to Manage Performance

Two strategies are key:

1. Have regular, constructive conversations about

performance

2. Strategic alignment—what constitutes

“performance” has to be aligned with what the
organization hopes to accomplish
 Key Insight About Performance #1:
Mean, Maximal, and Variable Performance

• Performance is not static.

• Changes in performance can affect
compensation.
• Being a typically high performer boosts
compensation among NBA players.
• Being an inconsistent performer hurts
compensation.
• And maximum performance has no effect.

Source: Barnes & Morgeson, 2007.

 Key Insight About Performance #2: Supervisor Ratings of
Performance Often Do Not Reflect Actual Performance

• Supervisors have a hard job to do when judging

employee performance.
• Performance of employees can be affected by factors
outside of their control.
• Their perceptions may be biased by political factors
(such as favouritism).
• Being seen as having a high-status friend can boost
performance ratings, even if no such friendship exists.
• Supervisors may struggle to remember occasions when
the employee performed well, or may place too much
weight on occasions when the employee performed
poorly, because poor performance was particularly
memorable.
 Key Insight About Performance #3: Performance Distributions are
Asymmetric—Some Employees Do Most of the Work

Performance is not evenly distributed in an organisation—

some people are more proficient and do more of the work.

Source: Aguinis & O’Boyle, 2014.

 Key Insight About Performance #4:
Performance Trends Matter
Even if two employees have the same performance rating, the
employee who has been “trending upward” will be rated as the higher
status employee than the one who has been “trending downward”.
Key Insight About Performance #5: The Same Performance Behaviour
Can Have Different Consequences for Males and Females

• Being helpful is thought to be less optional for

women than for men.
• Women receive less favourable evaluations
and recommendations when they refrain
from helping others, whereas men do not
receive the same penalty.
• Instead, men receive a boost for helping
others, but face no penalty when they choose to
refrain from helping out.
END

Source: Heilman & Chen, 2005.