HUMAN RESOURCE MANAGEMENT AND ANALYTICS

THE CAPSTONE PROJECT

Group Name – GROUP 13 IIM K DEC’22 - JUN'23

Total Members in the Group - 5

Group Member Name Email ID Phone Number

Member

1 Akanksha Nag [email protected] 7791862569

2 Kavita Kaushal [email protected] 9914909110

3 Priyanka Patel [email protected] 7219106071

4 Ahmed Siyal [email protected] +966540039965

5 Kamika Tomar [email protected] 8527797999

PROJECT TITLE
"Analyzing Factors Contributing to Employee Attrition in IBM: A Secondary Research Exploration"

PROJECT ABSTRACT
This project is focused to analyze the factors affecting attrition in IBM. As employee’s attrition very
negatively affects the organizational growth; organization come up with different lucrative schemes
from time to time to ensure retention of employees. Losing the trained manpower affects the business
growth negatively and in some cases, the historical memory of the organization also becomes
irremeable. To go ahead, the descriptive and diagnostic analysis have been done and suggestions as a
part of prescriptive analysis have also been inscribed. HR Attrition and Retention Metrics have been
calculated which enables to evaluate the periodical attrition and Live Dashboard shows the live
absolute data for attrition.

1 Group 13 IIM K Dec’22 – Jun’23

 TABLE OF CONTENTS
Description Page Number

Project Title 1

Project Abstract 1

Area of Research and Research Objective 3

Data Collection Method and Link to the Data Collected 3

Profile of the Organization 4

Theoretical Understanding of Key Concepts Related to the Problem 5

List of Metrics Identified 5

Analysis 6

Analysis By Applying Linear Regression 12

Link of Live HR Dashboard 15

Solutions/Recommendations 16

Drive Link for Group Video 16

Conclusion and Summary 16

2 Group 13 IIM K Dec’22 – Jun’23

 AREA OF RESEARCH AND RESEARCH OBJECTIVE
There are 5Ms in management i.e., Money, Material, Machinery, Methods, and Manpower (Human
Resource). Manpower (Human Resource) is that resource which manages all other resources. Human
Resource is the most dynamic resource in any organization as this is the one resource which is capable
of thinking, learning, and adapting to changing situations. Unlike other resources such as technology,
equipment or buildings, human resources can be trained, motivated, and developed to improve their
skills, knowledge, and performance. Human Resources can be a source of innovation, creativity, and
strategic thinking. Human Resource is also capable of driving the organization’s culture, environment,
and values, which can help to create a positive, productive, and pleasing work environment. Therefore,
Human Resource is the superior of all other resources, and it needs to be managed in a very effective
manner.
We are living in an era, where even the day to day lives are influenced by the rapidly changing
technology; thereby the today’s human resource is comparatively much aware and takes informed
decisions. In this fast-changing era, managing the attrition of human resource has become a major
challenge. The decision of employees to stay back in any organization or change the organization gets
affected by many factors. Getting a new and better job offer is a great ego boost; it shows that the
education and skills possessed by employees are desired and is valuable. Analyzing the factors
affecting the decision of the employees to change the organization is the need of the hour so that those
factors can be well screened and corrective actions may be taken to ensure less attrition and more
retention of employees.
Hence, the group has selected “Analyzing the factors of Attrition” as the research area and objective
is to present those factors in an analytical and a graphical way by using some of the statistical
techniques and live dashboard.

DATA COLLECTION METHOD

https://1drv.ms/x/s!AmdwMH9a4wdYgQc8raV18rIEE2_P?e=Na5I4h

Dataset Description – The dataset used in this project has been dragged from www.kaggle.com and
has been distributed by IBM Analytics. The group has not collected the data through any kind of survey
or other method, so this is a secondary data. This data set has 1470 rows and 35 data fields/ variables.
There are some numeric fields, and some are categorical fields like the variable/ data field with head
as “Attrition” has values as “Yes” or “No”, therefore the type of this data field is Categorical whereas
the data field with name “Job Satisfaction” has numeric values in it, therefore it has data type as
Numeric. The dataset has target variable/ data field as “Attrition” which has values as “Yes” and “No”’;
the values “Yes” represent the employees who left the organization and “No” represents the employees
who did not leave the company. The variable/ data field is our target variable/ data field and a
dependent variable as well.

3 Group 13 IIM K Dec’22 – Jun’23

 VARIABLES/ DATA FIELDS IN DATA SET
Age Marital Status
Attrition Monthly Income
Business Travel Monthly Rate
Daily Rate Num Companies Worked
Department Over18
Distance From Home Over Time
Education Percent Salary Hike
Education Field Performance Rating
Employee Count Relationship Satisfaction
Employee Number Standard Hours
Environment Satisfaction Stock Option Level
Gender Total Working Years
Hourly Rate Training Times Last Year
Job Involvement Work Life Balance
Job Level Years At Company
Job Role Years In Current Role
Job Satisfaction Years Since Last Promotion
Years With Curr Manager

PROFILE OF THE ORGANIZATION

About IBM – IBM are the Creators, Partners and clients putting technology to work in the real world.
It brings together all the necessary technology and services, regardless of where those solutions come
from, to help clients solve the most pressing business problems.

Detailed profile of IBM can be accessed from the link: www.ibm.com/about?lnk=flathl

Theoretical Understanding of Key Concepts Related to the Problem

Attrition – is the departure of an employee from an organization due to any reason (Voluntary or
Involuntary). Attrition rate refers to the rate at which employees leave an organization over a specific
period. It is a measure of employee turnover and can help organization understand how well they are
retaining their HR. High attrition rate is a cause of concern as it can lead to increased costs associated
with recruiting and training new employees.

4 Group 13 IIM K Dec’22 – Jun’23

 METRICS FOR “ATTRITION”
The following metrics have been calculated for the target dataset: -

Employee Turnover Rate/ Attrition Rate: Employee turnover rate refers to the number of people
who have left the organization in a specific period by the total number of employees in payroll during
that specific period. Formula for calculating Employee Turnover Rate is as follows: -

Number of attritions in a specified period * 100

Average number of employees during the period

In the target dataset: -

No. of employees at start = 1470

No. of attritions (employees with attrition "Yes") = 237
No. of employees at end = 1233
Average no. of employees = (1470+1233)/2 = 1351.5
Employee Turnover rate/ Attrition Rate/ Churn Rate = 17.54%

Turnover of High Performers: High-potential employees are the performance drivers of an

organization. Calculating the turnover rate of high performers helps assess the work environment and
processes. Formula for calculating Turnover of High Performers is as follows: -

Turnover of high performers *100

Total number of high performers in the organization

In the target dataset: -

No. of High performers (employees with attrition “Yes”) = 37

Total no. of high performers (“Yes” + “No” with rating 4) = 189+37 = 226
Turnover of High Performers = 16.37%

Stability Index: The employee stability index measures the stability of the workforce. This is
calculated by measuring the percentage of employees with more than 1 year of service out of the total
number of employees. Formula for calculating Stability Index is as follows: -

Number with service of 1 year or more * 100

Total number of employees

5 Group 13 IIM K Dec’22 – Jun’23

 In the target dataset: -

No. of employees with more than one year of service = 1255

Total no. of employees = 1470
Stability Index = 85.37

Workforce retention rate: Workforce retention rate is the metric that calculates the total percentage
of employees who have stayed with the organization over a specific period. Formula for calculating
Workforce retention rate is as follows: -

Number of new joiners who remain during the period * 100.

Total Number of new joiners during the period

In the target dataset: -

Total No. of new joiners (with less than one year at company) = 44
Total no. of new joiners who continued with Attrition “No” = 28
Workforce Retention Rate = 63.64%

DESCRIPTIVE ANALYSIS
In simple words, Descriptive Analysis is to analyze “What happened”. The first step for descriptive
analysis is to observe and draw out analysis with respect to various variables with Attrition. In the
target dataset, there are 1470 employees out of which 237 (16%) have left the service and 1233 (84%)
are still in place. Some of the major variables taken up to depict the relation with target variable/ data
field i.e., “Attrition” for descriptive analysis is as below: -
Monthly Income and Attrition - The exhibit – 1(a) depicts the Attrition with respect to Monthly
Income of the employees. The Income Group as 1000-2999 as their monthly income has the highest
attrition i.e., 28.61%. Out of total 237 employees with Attrition as “Yes”, 113 employees (47.68%) are
of the Income group 1000-2999. So, the employees with lower income are more likely to leave the job,
whereas the income group with mediocre income as 15000-16999 shows Attrition as “0” inferencing
that at this level employees are more stable in the organization.

6 Group 13 IIM K Dec’22 – Jun’23

 Monthly Income and Attrition %age
19000-20999 8 92

17000-18999 2.13 97.87

15000-16999 0 100

13000-14999 7.69 92.31 Yes

11000-12999 10 90 No

9000-10999 18.64 81.36

7000-8999 12.75 87.25

5000-6999 9.44 90.56

3000-4999 14.12 85.88

1000-2999 28.61 71.39

0 20 40 60 80 100 120

%age of
%age of %age of
Count of Count of employees
employees Total
employees employees Total with
Monthly with Attrition
with with count of attrition
Income level attrition (within the
attrition attrition employees "No" in
"Yes" in group
"Yes" "No" the
dataset "Yes")
dataset
1000-2999 113 282 395 28.61 71.39 47.68
3000-4999 50 304 354 14.12 85.88 21.1
5000-6999 27 259 286 9.44 90.56 11.39
7000-8999 13 89 102 12.75 87.25 5.49
9000-10999 22 96 118 18.64 81.36 9.28
11000-12999 3 27 30 10 90 1.27
13000-14999 4 48 52 7.69 92.31 1.69
15000-16999 0 36 36 0 100 0
17000-18999 1 46 47 2.13 97.87 0.42

7 Group 13 IIM K Dec’22 – Jun’23

 19000-20999 4 46 50 8 92 1.69
Grand Total 237 1233 1470
Exhibit – 1(a)
Age and Attrition - The exhibit – 1(b) depicts the Attrition age group wise. The age group 18 -24
years has the highest attrition i.e., 39.18% which means this age group is more likely to leave the
organization. So, more hiring of people of this age group may lead to more attrition. While observing
the absolute number, out of total 237 employees with Attrition as “Yes”, 80 employees are of the age
group 25-31 (33.76%) which is almost 1/3rd of the cluster of total attrition. So, the young employees
are more likely to leave the job, some may be at the start of their career. As the age group is increasing,
the %age of attrition is decreasing gradually after the age group of 25-31 years.

Age Group and Distribution of Attrition

87.58 90.03 87.43 87.74

77.65

60.82

39.18

22.35
12.42 9.97 12.57 12.26

18-24 25-31 32-38 39-45 46-52 53-60

Age Group (in years)

%age of employees with attrition "Yes" %age of employees with attrition "No"

%age of
Count of Count of %age of %age of
Total
employees employees Total count employees employees
Attrition
Age Group with with of with with
(within
attrition attrition employees attrition attrition
the group
"Yes" "No" "Yes" "No"
"Yes")
18-24 38 59 97 39.18 60.82 16.03
25-31 80 278 358 22.35 77.65 33.76
32-38 56 395 451 12.42 87.58 23.63
39-45 29 262 291 9.97 90.03 12.24

8 Group 13 IIM K Dec’22 – Jun’23

 46-52 21 146 167 12.57 87.43 8.86
53-60 13 93 106 12.26 87.74 5.49
Grand Total 237 1233 1470
Exhibit – 1(b)
Distance from Home and Attrition - The exhibit – 1(c) observes the attritions with the variable
Distance from home which depicts that as the distance from home increases the percentage of attrition
increases. The percentage of attrition is highest where the distance from home is more than 20 KMs,
while on the contrary, the absolute numbers of employees who left the company is highest where the
distance from home is less than 5 KMs i.e., 87 employees (36.71%) of total 237 employees who left
the organization.

DISTRANCE FRO M HOME & DI STRIBUTION O F

ATRRITION
Attrition"Yes" Attrition "No"

86.23 85.53 80 77.94

13.77 14.47 20 22.06

0 TO 5 KM 5 TO 10 KM 10 TO 20 KM MORE THAN 20
KM
DISTRANCE FROM HOME

Distance from Count of Count of Total %age of %age of %age of

Home employees employees count of employees employees Total
with with employees with with Attrition
attrition attrition attrition attrition (within
"Yes" "No" "Yes" "No" the group
"Yes")
0 to 5 KM 87 545 632 13.77 86.23 36.71
5 to 10 KM 57 337 394 14.47 85.53 24.05
10 to 20 KM 48 192 240 20 80 20.25

9 Group 13 IIM K Dec’22 – Jun’23

 More than 20 KM 45 159 204 22.06 77.94 18.99
Grand Total 237 1233 1470
Exhibit – 1(c)
Years at Company and Attrition - The exhibit – 1(d) depicts the attritions with the variable No. of
years since an employee is/ was working. For the group of less than 03 years, the percentage of attrition
is highest and the absolute numbers of employees who left the company by working for less than 03
years is also the highest. In the group 30 years of working, the attrition is again higher than the other
groups; while the absolute numbers are very low like 1 or 2 only; %age is higher because the total
employees working in this group is also lower like 6 or 9. So, it may be interpreted that the employees
leave early like before completion of 03 years in the organization; whereas after completion of 03 years
of service in the organization, employees are comparatively stable.

Years at Company and Attrition %age

93.42 91.11 93.1 93.75

87.29 88.22
83.33
77.78 75
74.04

25.96 25
22.22
16.67
12.71 11.78
6.58 8.89 6.9 6.25

0-3 4-7 8-11 12-15 16-19 20-23 24-27 28-31 32-35 36-40
Attrition - Yes Attrition - No

%age of
Count of Count of %age of %age of
Total
employees employees Total employees employees
Years at Attrition
with with count of with with
company (within the
attrition attrition employees attrition attrition
group
"Yes" "No" "Yes" "No"
"Yes")
0-3 122 348 470 25.96 74.04 51.48
4-7 60 412 472 12.71 87.29 25.32
8-11 37 277 314 11.78 88.22 15.61

10 Group 13 IIM K Dec’22 – Jun’23

 12-15 5 71 76 6.58 93.42 2.11
16-19 4 41 45 8.89 91.11 1.69
20-23 4 54 58 6.9 93.1 1.69
24-27 1 15 16 6.25 93.75 0.42
28-31 1 5 6 16.67 83.33 0.42
32-35 2 7 9 22.22 77.78 0.84
36-40 1 3 4 25 75 0.42
Grand Total 237 1233 1470

YEARS WITH
CURRENT MANAGER
Attrition - Yes Attrition - No
94.59

100
91.15
87.45

87.22
78.62
21.38

12.78
12.55

8.85

5.41

0-2 3-5 6-8 9-11 12-14 15-17

Higher the years with Current Manager, Lesser the Attrition Exhibit – 1(e)

Job Satisfaction
83.57 83.48 88.67
77.16

22.84
16.43 16.52 11.33

1 2 3 4

Attrition - Yes Attrition - No

Higher the Job Satisfaction, Lesser the Attrition Exhibit – 1(f)

11 Group 13 IIM K Dec’22 – Jun’23

 DEPARTMENT WISE Department performance to
CONTRIBUTION TO Retention & Attrition

%age of employees with

ATTRITION (IN %AGE)

Attrition "Yes"
Human 20.63%
19.05%
Resources
Sales 5.06% 13.84%
38.82%

Research
&
Develop
ment Human Research & Sales
56.12% Resources Development

Exhibit – 1(g)

Exhibit – 1(g)

DIAGNOSTIC ANALYSIS BY APPLYING LINEAR REGRESSION

As the name says, Diagnostic Analysis is to check “Why did it happen”? Steps for applying Regression
as a statistical technique: -

Data Cleaning – To start with, Data Cleaning is the first step for applying Regression. The categorical
values of relevant variables/ data fields were transformed into Numeric Values with codes like 1, 2, 3
etc. For Example – The variable “Attrition” consisting of two values, was transformed as follows: -
 The value “Yes” was mapped as 1.
 The values “No” was mapped as 2.
The values of other relevant variables/ data fields were also mapped in the same manner.

Creating Hypothesis – In simple words, Hypothesis is basically an assumption which may be true or
false. The opposite of Hypothesis is a Null Hypothesis. For the target dataset, the following Hypothesis
has been created taking “Attrition” as dependent variable and others as independent variables.

12 Group 13 IIM K Dec’22 – Jun’23

List of Hypothesis taken for Regression Analysis: -
Output after
applying
Sr. No. Hypothesis
regression as
below
H1: Gender and attrition are dependent P-value > 0.05
Hypothesis - 1
H0: Gender and attrition are NOT dependent H1 is rejected
H1: Marital Status and attrition are dependent P-value < 0.05
Hypothesis – 2
H0: Marital Status and attrition are NOT dependent H1 is accepted
H1: Distance from home and attrition are dependent P-value < 0.05
Hypothesis – 3
H0: Distance from home and attrition are NOT dependent H1 is accepted
H1: Age and attrition are dependent P-value < 0.05
Hypothesis – 4
H0: Age and attrition are NOT dependent H1 is accepted
H1: Business Travel and attrition are dependent P-value > 0.05
Hypothesis – 5
H0: Business Travel and attrition are NOT dependent H1 is rejected
H1: %age Salary Hike and attrition are dependent P-value > 0.05
Hypothesis – 6
H0: %age Salary Hike and attrition are NOT dependent H1 is rejected
H1: Performance Rating and attrition are dependent P-value > 0.05
Hypothesis – 7
H0: Performance Rating and attrition are NOT dependent H1 is rejected
H1: Years at comp and attrition are dependent P-value > 0.05
Hypothesis - 8
H0: Years at comp and attrition are NOT dependent H1 is rejected
H1: Years with Cur Manager and attrition are dependent P-value < 0.05
Hypothesis - 9
H0: Years with Cur Manager and attrition are NOT dependent H1 is accepted
H1: Monthly Income and attrition are dependent P-value < 0.05
Hypothesis – 10
H0: Monthly Income and attrition are NOT dependent H1 is accepted
Summary Output
Regression Statistics
Multiple R 0.336546742
R Square 0.11326371
Adjusted R Square 0.105346422
Standard Error 0.347947402
Observations 1470
Here, Significance of F is less than 0.05
ANOVA
df SS MS F Significance F
Regression 13 22.51566977 1.731974598 14.30587159 1.29168E-30
Residual 1456 176.2741261 0.121067394
Total 1469 198.7897959

13 Group 13 IIM K Dec’22 – Jun’23

 Coeffic Standar t Stat P-value Lower Upper Lower Upper
ients d Error 95% 95% 95.0% 95.0%
1.5258 0.12281 12.4238 1.28495 1.7667 1.28495 1.76679
Intercept 9.32966E-34
76363 837 4473 6508 96218 6508 6218
- - - -
0.01859 0.0104 0.01049
Gender Code 0.0259 1.39715 0.162580067 0.06245 0.06245
5433 96013 6013
80689 4259 739 739
- - - - - -
Marital 0.01253
0.0735 5.86875 5.42798E-09 0.09816 0.0489 0.09816 0.04898
Status 6131
71532 9159 2338 80725 2338 0725
- - - - - -
Distance 0.00112
0.0036 3.24609 0.0011966 0.00584 0.0014 0.00584 0.00144
From Home 1822
41542 5539 2103 40982 2103 0982
0.0033 0.00115 2.93026 0.00111 0.0056 0.00111 0.00564
Age 0.003439496
81267 3911 6528 7761 44772 7761 4772
- - - -
Business 0.01370 0.0252 0.02527
0.0016 0.11680 0.907028649 0.02847 0.02847
Travel 1619 7657 657
00453 7576 7476 7476
- -
Percent 0.0051 0.00392 1.30226 0.0128 0.01281
0.193030498 0.00258 0.00258
Salary Hike 15077 782 873 19866 9866
9713 9713
- - - -
Performance 0.03980 0.0420 0.04203
0.0360 0.90540 0.365401152 0.11411 0.11411
Rating 2408 38927 8927
37262 4076 3452 3452
- - - -
Years At 0.00256 0.0018 0.00186
0.0031 1.23534 0.21690115 0.00820 0.00820
Company 565 633 33
69465 5739 223 223
Years With
0.0151 0.00401 3.76764 0.00725 0.0230 0.00725 0.02300
Current 0.00017137
31189 6089 3119 3251 09128 3251 9128
Manager
Monthly 6.6754 2.48227 2.68924 1.80623 1.1544 1.80623 1.15446
0.007243084
Income 3E-06 E-06 7766 E-06 6E-05 E-06 E-05
Job 0.0635 0.01283 4.94973 0.03835 0.0887 0.03835 0.08870
8.29676E-07
Involvement 26172 426 3862 0557 01788 0557 1788
Job 0.0383 0.00825 4.65043 0.02219 0.0545 0.02219 0.05458
3.61264E-06
Satisfaction 8839 48 2594 5819 80961 5819 0961
Environment 0.0351 0.00831 4.22829 0.01884 0.0514 0.01884 0.05145
2.50132E-05
Satisfaction 49306 2874 7573 2817 55796 2817 5796
The best way to understand P-value is as the probability of error. P-value should be as small as
possible.

14 Group 13 IIM K Dec’22 – Jun’23

 INTERACTIVE HR DASHBOARD
https://1drv.ms/x/s!AmdwMH9a4wdYgQc8raV18rIEE2_P?e=Na5I4h

Please click the above link for interactive live Dashboard: (Snapshot of the same is presented)

15 Group 13 IIM K Dec’22 – Jun’23

 CONCLUSION SUMMARY (SOLUTIONS / SUGGESTIONS)

&

Group Video Link

https://1drv.ms/v/s!AmdwMH9a4wdYgQszTpLTmU1WVaGY?e=kIDqLo

Conclusion & Summary: This study tried to answer some simple questions related to Human
Resource Management –

1. Which are the prime factors contributing to attrition?

2. Who is leaving the organization?
3. What measures are to be taken to ensure the retention of employees in the organization?

Major variables affecting Attrition are summed up as Monthly Income, Distance from home, Age,
Years with Current Manager, Job Involvement and Environment Satisfaction. The metrics of stability
index of the target dataset is higher, which may be concluded as the employees stay for longer duration
after one year of completion of their service. The workforce retention rate is moderate, which may be
concluded as the organization needs to work more upon the orientation and resolution of issues being
faced in the first year of service by new employees, so that they can be retained in the service.

Observing the exhibit 1(g), 20.63% employees of Sales department have left the service, closer to that
is Human Resources department in which 19.05% employees have left the service; while out of the
total employees who left the service, major share is of Research & Development department (56.2%).
Attrition Rate of Sales department i.e., 23% which is the highest of all the three departments
(HR – 21.05%, R & D is 14.87%).

Who is leaving the service? – Employees are hired for a particular job role considering their
qualifications (Education). In the job market, candidates with a particular education field/ profile are
looked for as per the vacant position in an organization. Therefore, it is a very important and crucial
aspect to check and review who is leaving the organization. As per exhibit 1(h), the attrition rate of
employees w.r.t. their Education Field is presented.

16 Group 13 IIM K Dec’22 – Jun’23

 Count of employees
Total count of
Education Field with attrition Attrition Rate
employees
"Yes"
Human Resources 7 27 29.79
Life Sciences 89 606 15.85
Marketing 35 159 24.73
Medical 63 464 14.57
Other 11 82 14.38
Technical Degree 32 132 27.59
Total 237 1470
Exhibit – 1(h)

The attrition rate of employees with education field Human Resource is the highest i.e., 29.79%;
however, these are handful of employees – only 1.8% of total 1470 employees in this organization.
The second highest attrition rate is of the employees with education field as Technical Degree 27.59%;
this may be considerable as these may be the employees with technical knowledge and skills as well.
To proceed further, the attrition rate of employees with education field as Marketing is higher 24.73%,
this reflexes that the employees with Marketing as Education field are leaving at a considerable speed.
The overall attrition rate is calculated as 17.64%. The attrition rate of the rest of the education fields is
lesser than the overall attrition rate.

The attrition rate of the Sales department is the highest and employees with education field as
Marketing are also leaving the service at comparatively higher speed. The reasons for their leaving
may include a higher pay package in other organizations, more lucrative incentives, a better
environment etc. By doing the competitive study about the pay packages, organization can offer better
pay packages comparatively, offer handsome incentives depending upon their performance, add some
allowances in the pay package and should ensure a congenial working environment in the organization.
To overcome the factor of distance from home, some pick & drop kinds of facilities can be offered
starting from an approachable central location in the local area. So, these measures may be adopted as
per feasibility for retaining the employees.

17 Group 13 IIM K Dec’22 – Jun’23