Module 5: HR Analytics and Predictive Modelling

Different phases of HR analytics and predictive modelling – data and information for HR
predictive analysis – software solutions – predictive analytics tools and techniques –
understanding future human resources.
 Different Phases of HR Analytics / Predictive Modelling
HR analytics evolves through multiple phases, from basic operational tracking to advanced predictive insights. The four
key phases are:

Advanced Predictive
Operationa Advanced
Analytics Analytics
l Phase Reporting
Phase Phase
(Descriptiv Phase
(Prescriptiv (Strategic
e (Diagnostic
e HR
Analytics) Analytics)
Analytics) Analytics)
 This is the basic level of HR analytics, where organizations track and report
HR data without performing deep analysis. The focus is on data
collection and summarization.

Operationa 🔹 Key Features:

l Phase ✅ Basic HR reports (headcount, attendance, turnover rates)
(Descriptiv ✅ Payroll processing and compliance tracking
e ✅ Simple tracking of recruitment, training, and employee engagement
Analytics)
🔹 Example Metrics:
📌 Employee Turnover Rate = (Employees Left / Total Employees) × 100
📌 Average Time to Hire = (Total Hiring Days / No. of Hires)
 🔹 Business Value:
💡 Helps HR teams manage daily operations efficiently but does not
provide deep insights for decision-making.
 This phase goes beyond tracking and focuses on understanding the why behind HR
trends. It provides comparisons, benchmarks, and correlations to identify key HR
issues.

🔹 Key Features:
Operationa
Advanced ✅ Dashboards and visual analytics
Reporting
l Phase ✅ Identifying reasons for turnover, low engagement, or hiring delays
(Descriptiv
Phase ✅ Benchmarking HR performance against industry standards
(Diagnostic
e
Analytics) 🔹 Example Metrics:
📌 Recruitment Efficiency Ratio = (No. of Quality Hires / Total Hires)
📌 Absenteeism Rate = (Total Unscheduled Absences / Workdays) × 100
 🔹 Business Value:
💡 Helps HR leaders understand the root causes of workforce trends and take
corrective actions.
 At this stage, organizations use statistical models, machine learning, and
automation to make data-driven HR decisions.

🔹 Key Features:
Advanced ✅ Use of HR scorecards and dashboards for decision-making
Operationa
Advanced
Analytics ✅ Predictive modeling to understand employee performance trends
Reporting
l Phase
Phase ✅ Workforce planning models to forecast hiring needs
(Descriptiv
Phase
(Prescriptiv
(Diagnostic
e
e 🔹 Example Metrics:
Analytics)
Analytics) 📌 Training Effectiveness Score = (Performance Improvement Post-
Training / Pre-Training Performance) × 100
📌 Flight Risk Score = Predicting which employees are likely to leave based
on engagement trends
 🔹 Business Value:
💡 Enables organizations to align HR strategy with business goals, leading
to more effective workforce management.
 This is the most advanced phase, where HR analytics predicts future workforce
trends using AI, big data, and machine learning models.

🔹 Key Features:
✅ Predicting employee turnover based on engagement, salary, and market
Advanced
Predictive trends
Operationa
Advanced ✅ Forecasting future skill needs to drive workforce planning
Analytics
Reporting
l Phase ✅ AI-driven insights for talent acquisition and retention
Phase
(Descriptiv
Phase
(Prescriptiv
(Strategic 🔹 Example Use Cases:
(Diagnostic
e
HR
e 📌 Predicting Employee Attrition: Identifies high-risk employees and
Analytics)
Analytics) suggests retention strategies
📌 Optimizing Workforce Productivity: Predicts performance outcomes
based on training, engagement, and work environment
 🔹 Business Value:
💡 Provides actionable insights to make proactive HR decisions, reducing risks
and optimizing workforce efficiency.
 Data and Information for HR Predictive Analysis
HR Predictive Analysis involves using historical HR data, statistical models, and
machine learning techniques to forecast workforce trends and improve decision-making.
To perform predictive analysis effectively, high-quality HR data is required, which can
be categorized into different types.
Type of HR Data Examples Source of Data
Age, Gender, Educational Qualification,
Demographic Data HR Records, Employee Profiles
Location
Time to Hire, Cost per Hire, Offer
Recruitment Data ATS (Applicant Tracking System)
Acceptance Rate
KPI Scores, 360-Degree Feedback,
Performance Data Performance Management System
Productivity Metrics
Employee Surveys, Sentiment
Engagement Data Pulse Surveys, Employee Feedback
Analysis, Feedback Scores
Salary, Bonuses, Incentives, Pay
Compensation Data Payroll System
Discrepancies
Training Completion, Skill
Training & Development Data LMS (Learning Management System)
Assessments, Learning Hours
Sick Leaves, Unplanned Absences, PTO
Absenteeism Data Attendance Tracking System
Usage
Exit Interviews, Turnover Rate,
Attrition Data HR Exit Surveys, Employee Records
Resignation Trends
 HR Information Used for Predictive Analysis
Once HR data is structured, it is used to generate insights that help in forecasting and
decision-making.
HR Information Use in Predictive Analytics

Predicts who is likely to leave based on past

Employee Turnover Trends
attrition patterns.

Identifies best hiring channels and improves hiring

Recruitment Success Patterns
efficiency.

Predicts future performance based on past trends

Performance Forecasting
and training.

Determines factors affecting employee satisfaction

Engagement Insights
and retention.

Evaluates the impact of learning programs on

Training Effectiveness
job performance.
Techniques Used in HR Predictive Analytics

Technique Application in HR Analytics

Regression Analysis Predicts salary trends, turnover probability.

Identifies key factors causing employee

Decision Trees
attrition.

Segments employees based on behaviour

Cluster Analysis
and engagement.

Forecasts workforce productivity using big

Neural Networks (AI/ML)
data.
Example: Predicting Employee Turnover
 A company wants to predict employee attrition and
takes the following steps:
 Collects historical data (employee demographics, salary,
engagement scores, performance ratings).
 Uses machine learning models to analyze trends (e.g.,
logistic regression, decision trees).
 Identifies key risk factors (low engagement, low salary
growth, poor training opportunities).
 Develops retention strategies (better benefits, career
growth opportunities).
📌 Result: The company reduces turnover by 20% using
predictive HR insights
 Software Solutions for HR Analytics
 Various statistical and analytical software solutions help HR professionals analyse
workforce data, forecast trends, and make data-driven decisions. Below is a detailed
explanation of some key software solutions used in HR analytics.

SPSS
(Statistic SAS
JASP (Just
al (Statistic
Another
Package Minitab Stata al R
Statistical
for the Analysis
Package)
Social System)
Sciences)
Software Solutions for HR Analytics
🔹 Developed by: IBM
🔹 Best For: Statistical analysis and predictive modeling
🔸 Features:
SPSS ✔ User-friendly interface for statistical analysis
(Statistical ✔ Supports regression, factor analysis, and correlation tests
Package for ✔ Helps in HR forecasting and workforce planning
the Social ✔ Provides data visualization tools
Sciences)
🔸 HR Applications:
📌 Analyzing employee satisfaction surveys
📌 Predicting employee attrition based on historical data
📌 Identifying training program effectiveness

SAS
JASP (Just
(Statistic
Another
Minitab Stata al R
Statistical
Analysis
Package)
Software Solutions for HR Analytics
🔹 Developed by: Minitab, LLC
🔹 Best For: Six Sigma and Quality Improvement Analysis
🔸 Features:
SPSS ✔ Strong statistical analysis and data visualization tools
(Statistical ✔ Predictive analytics capabilities
Minitabfor ✔ Advanced regression and hypothesis testing
Package
the Social
Sciences) 🔸 HR Applications:
📌 Measuring training effectiveness using hypothesis tests
📌 Identifying key recruitment factors affecting hiring success
📌 Conducting workforce optimization studies

SAS
JASP (Just
(Statistic
Another
Stata al R
Statistical
Analysis
Package)
Software Solutions for HR Analytics
Developed by: StataCorp
🔹 Best For: Econometrics and social science research
🔸 Features:
SPSS ✔ Advanced statistical modeling (regression, survival
(Statistical analysis, panel data)
Stata for ✔ Handles large datasets efficiently
Package
Minitab
the Social ✔ Strong data management capabilities
Sciences)
🔸 HR Applications:
📌 Salary trend analysis using regression models
📌 Employee retention and turnover analysis
📌 Workforce productivity measurement

SAS
JASP (Just
(Statistic
Another
al R
Statistical
Analysis
Package)
Software Solutions for HR Analytics
🔹 Developed by: SAS Institute
🔹 Best For: Business intelligence and advanced analytics
🔸 Features:
SPSS ✔ Machine learning and AI-driven analytics
SAS ✔ Predictive modeling and forecasting
(Statistical
(Statistical
Stata for ✔ Data integration from multiple sources
Package
Minitab
Analysis
the Social
System)
Sciences) 🔸 HR Applications:
📌 Predicting workforce attrition using AI models
📌 Optimizing recruitment processes through data-driven
insights
📌 Analyzing compensation structures for pay equity

JASP (Just
Another
R
Statistical
Package)
Software Solutions for HR Analytics
🔹 Developed by: R Foundation
🔹 Best For: Open-source statistical computing
🔸 Features:
SPSS ✔ Powerful for data visualization and statistical modeling
SAS ✔ Extensive libraries for HR analytics
(Statistical
(Statistical
R for ✔ Supports machine learning for predictive workforce
Package
Stata
Minitab
Analysis
the Social
System) analytics
Sciences)
🔸 HR Applications:
📌 Conducting employee sentiment analysis using text mining
📌 Predicting high-performing employees based on HR data
📌 Creating HR dashboards for workforce analytics

JASP (Just
Another
Statistical
Package)
Software Solutions for HR Analytics
🔹 Developed by: University of Amsterdam
🔹 Best For: Open-source statistical software with a user-
friendly interface

SPSS 🔸 Features:
JASPSAS
(Just ✔ Easy-to-use with point-and-click functionality
(Statistical
(Statistical
Another
Package
R for
Stata
Minitab
Analysis ✔ Bayesian and frequentist statistical methods
Statistical
the Social
System)
Package) ✔ Free alternative to SPSS
Sciences)
🔸 HR Applications:
📌 Performing employee satisfaction analysis
📌 Analyzing recruitment success rates
📌 Conducting workforce diversity studies
Comparison Table: HR Analytics Software Solutions

Software Best For Ease of Use Key HR Applications

Statistical analysis & Employee satisfaction,

SPSS Easy
predictive modeling attrition prediction

Quality control & Six Training effectiveness,

Minitab Moderate
Sigma recruitment factors

Econometrics & Salary trends,

Stata Moderate
research analysis workforce productivity

Attrition prediction,
Advanced business
SAS Complex recruitment
analytics & AI
optimization
Open-source data
HR dashboards,
R science & Complex
sentiment analysis
visualization
HR reporting,
Free alternative to
JASP Easy workforce diversity
SPSS
analysis
 Predictive Analytics Tools and Techniques for HR
Analytics
Predictive analytics helps HR professionals forecast workforce trends, anticipate
employee behaviour, and improve decision-making using data-driven techniques.
Below is a detailed explanation of key predictive analytics tools and
techniques used in HR analytics.

Neural
Networ
Decisio
Multino ks
n Tree
Discret Logisti mial (Artifici
Linear Probit and
e c Logisti al
Regres Regres Scenari
Choice Regres c Intellig
sion sion o
Models sion Regres ence -
Analysi
sion AI
s
Models
)
 Predictive Analytics Tools and Techniques
Multinom for HR N
Analytics Discrete
Linear Logistic ial Probit Net
Regressio Choice Regressio Logistic Regressio (Ar
n Models n Regressio n Inte
n - AI

Linear regression is used to predict a continuous

outcome based on one or more independent
variables. It assumes a linear relationship
between variables.
📌 Formula:
Y=a+bX+eY = a + bX + eY=a+bX+e
Where:
Y = Dependent variable (e.g., employee salary)
X = Independent variable (e.g., years of
experience)
a = Intercept
b = Slope (impact of X on Y)
 Predictive Analytics Tools and Techniques
Multinom for HR N
Analytics Discrete
Linear Logistic ial Probit Net
Regressio Choice Regressio Logistic Regressio (Ar
n Models n Regressio n Inte
n - AI

These models are used when an individual chooses between two or more discrete options
(e.g., Yes/No, Male/Female, Hybrid/Remote/On-site).
📌 HR Example:
Scenario: An HR team wants to analyze employee preferences for work location
(Remote, Hybrid, or On-Site).
 Predictive Analytics Tools and Techniques
Multinom for HR N
Analytics Discrete
Linear Logistic ial Probit Net
Regressio Choice Regressio Logistic Regressio (Ar
n Models n Regressio n Inte
n - AI

Logistic regression predicts the probability of

an event occurring when the outcome is
binary (Yes/No, Stay/Leave, Promote/Not
Promote).
📌 Formula:
P(Y)=11+e−(a+bX)P(Y) = \frac{1}{1 + e^{-(a
+ bX)}}P(Y)=1+e−(a+bX)1​
Where P(Y) is the probability of an event
happening.
📌 HR Example:
Scenario: HR wants to predict if an employee
will stay or leave based on job satisfaction
scores (scale of 1 to 10).
 Predictive Analytics Tools and Techniques
Multinom for HR N
Analytics Discrete
Linear Logistic ial Probit Net
Regressio Choice Regressio Logistic Regressio (Ar
n Models n Regressio n Inte
n - AI

This is an extension of logistic regression used when

there are more than two outcome categories.
📌 HR Example:
Scenario: HR wants to classify employees into three
performance levels based on performance scores:
Low Performer (0-40)
Average Performer (41-70)
High Performer (71-100)
 Predictive Analytics Tools and Techniques
Multinom for HR N
Analytics Discrete
Linear Logistic ial Probit Net
Regressio Choice Regressio Logistic Regressio (Ar
n Models n Regressio n Inte
n - AI

Similar to logistic regression, but

it assumes a normal distribution
of probability. It is used when HR
decisions are based on uncertain
outcomes.
📌 HR Example:
Scenario: HR wants to predict the
probability of an employee
accepting a job offer.
 Discr Analytics
Predictive nomiTools and Techniques for HR
Line Logis
ete al Neural Networks Decision
Analytics
ar
Choi
tic
Logis
Probit
(Artificial Tree and
Reg Regr Regressio
ce tic Intelligence - AI Scenario
ress essio n Models) Analysis
Mod Regr
ion n
els essio
n
Neural networks mimic the human
brain to analyze complex,
nonlinear patterns and improve HR
decision-making.
📌 HR Example:
Scenario: A company uses AI to
predict employee attrition based
on multiple factors like salary, work
environment, and job satisfaction.
 Discr Analytics
Predictive nomiTools and Techniques for HR
Line Logis
ete al Neural Networks Decision
Analytics
ar
Choi
tic
Logis
Probit
(Artificial Tree and
Reg Regr Regressio
ce tic Intelligence - AI Scenario
ress essio n Models) Analysis
Mod Regr
ion n
els essio
n
Decision Tree: Breaks down
HR decisions into different
possible paths.
Scenario Analysis:
Examines "what-if"
scenarios for HR strategies.
📌 HR Example:
Scenario: HR wants to
analyze promotion criteria
based on performance and
experience.