Lecture 1 – Introduction to HR Analytics:
1. Edwards & Edwards (2019): Chapter 1
2. Khan & Millner (2020): Chapter 1 & 2
Chapter 1 - Understanding HR analytics (Edwards & Edwards)
“Analytics present a tremendous opportunity to help organizations understand what they
don’t yet know… By identifying trends and patterns, HR professionals and management
teams can make better strategic decisions about the workforce challenges that they may soon
face.” (Huselid, 2014).
Even though many HR professionals may have a conceptual understanding of what HR
analytics might involve, very few people have the relevant competencies to be able to
actually carry out sophisticated predictive HR analytics.
Why is HR-analytics important?
1. Using data to form decisions for the management.
2. Statistical analysis and data presentation are both in the top 10 most important skills
that are globally needed.
Predictive HR analytics: the systematic application of predictive modelling using
inferential statistics to examine HR people-related data in order to inform judgements about
possible causal factors driving key HR-related performance indicators. The results make it
possible to predict causes, results and outcomes.
• We take the sophisticated statistics and quantitative analyses techniques that scientists
use to predict things (such as what may cause heart disease or what might help to cure
cancer) and apply them to the information we hold about people in organizations.
• This enables us to predict things such as what might drive high performance or what
might cause an employee to leave the organization.
• We can also apply these predictive models to make tangible predictions about particular
results or outcomes that we might expect to find, given certain conditions.
• Being able to apply predictive statistical models to HR-related data requires some
knowledge of statistics and the capability to understand and interpret meaning behind
results that analyses are telling us. Understanding these statistical tools and how they
apply to HR data provides a deep foundational basis for any budding HR analytics
professional.
Understanding the need (and business case) for mastering and utilizing predictive HR
analytic techniques
Many HR teams produce descriptive reports again and again. However,these reports only
present a picture or snapshot of what is occurring at that particular time. Unfortunately, they
say little about what is really going on, they do not give us insights into why things are
happening in the organization. Besides,they lack to check whether the data is robust and/or
valid. Also, descriptive reports do not help to predict what we might find in the future.
1
,Importance of predictive analytics: offers the opportunity to help organizations predict and
understand historical data (the performance of a person or a group) and identify, and analyze
patterns in order to help understand causal factors. By identifying trends and patterns, HR
professionals and management teams can make better strategic decisions about the
workforce challenges that they may soon face.
• Knowing what has happened in our organization and having evidence for why things
have happened; what the drivers are of certain behaviors within our organization, will
undoubtedly help us to make better decisions.
Human capital data storage and ‘big (HR) data’ manipulation
The success of HR analytics is completely reliant on the availability of good people-related
information; we cannot look for patterns in data when the available data is limited and
sketchy. However, HR managers will probably face the problem that there is too much data
available to know what to do with it. The biggest challenge is to get the data into the right
format for analysis. Useful HR-related data is made up of many different types of
information and might include the following:
• skills and qualifications
• measures of particular competencies
• training attended
• levels of employee engagement
• customer satisfaction data
• performance appraisal records
• pay, bonus and remuneration data.
Ultimately, the data available (and the data that is missing) is the key determining factor on
what kind of analysis can be carried out and what business questions can be answered.
Predictors, prediction and predictive modelling
Central to the idea of “predictive” HR analytics is that something can be “predicted”. There are
three uses of the term “prediction” that we can use in the context of “predictive HR analytics”.
1. Predictors: the idea of identifying “predictors” or potential “causal” factors that help
explain why a particular feature or measure shows variation → some of the analytic
techniques aim to explore relationships between many different types of data (variables)
in order to identify “predictors” of some important HR outcomes (employee
performance, staff turnover). This type of analysis is used to identify trends and
relationships between multiple factors with the hope of obtaining information that
suggests the possible causes of variation in the phenomenon that we are hoping to
predict. In this context one can also refer to these predictors as potential “drivers” of our
outcome. Importantly, the use of the word “predictors” here implies that we seek out
and have found potential “causes” of variation on the feature we are trying to predict.
2. Predictive modeling: we take features and findings of our analysis, then we apply our
model to help demonstrate or “predict” what would happen to our key outcome variable
if we could do something to change or adjust the key drivers that we have identified.
3. The third use is that we can translate the findings from our “predictive models” where
we identified “predictors” of variation in our particular outcome variable, and we use
the resulting model to “predict” how current or future employees may behave in the
2
, future. Through identifying patterns and trends in existing data, you can apply a
particular algorithm to newly collected information to provide evidence-based
predictions of future behavior that can help managers to make a decision.
Current state of HR analytics professional and academic training
At the time, the majority of people who enter the HR profession do not have the required skills
to carry out predictive HR analytics. However, recent research is beginning to show how
important analytic competencies are in the HR profession. For example, reports on a Deloitte
survey showed that “human capital analytics” was important for business performance, but
only 8% had strong capabilities. Often data scientists/quantitative analysts who do not originate
from the field of HR are an important resource to draw upon to include in the HR analytics
team. This is mainly because HR specialists rarely have the requisite skills, even though there
is a perceived need for the HR profession to have these capabilities. Whether the HR
professional is a generalist, a specialist in one particular area, or the head of HR for a large
multinational organization, the ability to identify and understand trends and patterns, to take
bias out of decision making and to predict organizational challenges is something that will set
them apart in becoming a credible, high-performing HR professional (helping the organization
to be more successful). Consequently, it can be stated that this gap needs to be filled.
Business applications of modeling
One of the things that the HR analytics team will need to be able to do, as a matter of course,
is to be able to translate analysis findings to potential business applications. Importantly, any
HR analytics team should introduce a mentality of always looking to answer the “so what?”
question → one that they will be asked when presenting their analytical results. The analytics
team needs to be always on the lookout for how their findings could be translated into useful
practice knowledge, and whether any particular knowledge gained can help to strengthen and
steer the organization’s people strategy.
HR analytics & HR people strategy
It is possible to use analytical models to help steer, adjust and even drive business strategy.
Ultimately the analytics approaches recommended can provide evidence-based pointers for the
practice and can help take some emotion and gut instinct out of “people” decision making.
Becoming a persuasive HR function
“The development of HR’s strategic role has been an evolution… The next step in the evolution
is for HR professionals, and particularly senior HR professionals, to develop what we call
analytic literacy.” - Huselid & Becker (2005)
We believe that this “analytic literacy” will help transform the HR function. An HR function
that fully utilizes predictive HR analytics capabilities will be more credible because the
function will be able to present robust “hard” evidence to show that is has a good understanding
of what makes its people tick, along with knowledge of who is likely to perform well, who is
likely to leave, which parts of the organization are showing race or gender bias, which
candidates are likely to be successful in the organization, and which interventions had a
significant impact on the organization (and which not). The function will be able to carry out
substantial “what if” scenario modeling to help build solid business cases that help the
organization to make decisions around whether particular investments are likely to be
worthwhile, and what the return of these investments are likely to be.
3
, Chapter 1 – Redefining HR (Khan & Millner)
HR and the new world of work
The Three Ds:
1. Data: Data is crucial to understand. The success of practices depends on the value you
can prove that it positively influences the business. This is difficult to prove. Also, data
can clarify the alignment between practices and the organization strategy (e.g. low-cost
strategy and HR-practices).
2. Design: First, the overall design of an organization is changing (structure). But also, the
reshaping of the jobs itself (switching to robots).
3. Digital: Making HR more efficient through the fast pace of changes, business and
employees are expected to embrace innovations like never before.
The following figure highlights the most important themes that have to be considered to be
‘future proof’.
The digital world of work
The key elements of the digital work are driven by the “changing disruptors”:
1. The overwhelmed workforce: Expectations of employees increasing all the time. At the
same time, managers have to manage across more complex relationships. Through
increasing automation and ongoing changes, the key challenges revolve around:
• Execution - achieve more with less driven by the challenging cost agenda.
• Employee expectations - there is an increasing need to learn new processes, skills,
and practices as automation increases.
• Manager priorities - challenge to find the right balance between operational
task/processes completion and the increasing desire for the ‘human touch’ which is
vital when creating an environment that employees want to work in.
• Leader ‘bandwidth’ - beyond their responsibility to execute the demands of
stakeholders, issues are becoming increasingly complex (e.g. different cultures).
• Wellbeing - operational challenges versus wellbeing – stress and ill-health becoming
operational challenges that need to be facilitated.
→ Data can be collected to understand how the workforce is really feeling.
4
