Factors Influencing Attrition In It Sectors

Published:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The important aspect of organizations globally is the management of human capital which involves tangible and intangible costs. Besides tangible costs like replacement costs, intangible costs involve loss of intellectual capital and low employee morale. Attrition and recruitment are some of the challenges being faced by the HR managers. Attrition is the problem with almost every sector in India especially with IT/ITES sectors. In the era of globalization it is the major challenge for these sectors. The current study helps in identifying the major reasons for attrition in these sectors using Business Intelligence and come up with a strategic decision which would help the companies reduce employee attrition. In the current study, the authors could identify that rewards & recognition, career development and working conditions are the major factors influencing the attrition in IT/ITES sectors.

KEY WORDS

Attrition, Business Intelligence

INTRODUCTION

Human resource management is a modern term for the classical term personnel management. But just the nomenclature didn't change. Along side, the function too became more complex (Singh, Shelley, 2003). A HR manager is incharge of all activities related to man power in his/her organisation (Stahl, G. K. & Bjorkman, I. 2006). Though every sub-function is crucial, managing attrition appears to be seemingly more important as attrition is prevalent in the emerging sectors like the IT industry. Today, one of the biggest challenges of the HR manager is control attrition in the organisation. This problem is posing many problems to the smooth functioning of the organisations. If one approaches a HR manager of an IT company, asking on what is the biggest challenge faced, the answer may not be technology or schedules or cost but attrition (Ferrat, et. al., 2005). Attrition not only causes huge loss to the company, in terms of manpower but also in monetary terms. Hence, there is a need to look at this problem from the point of view of effective HR management.

As Attrition is the main problem of HR department, they realized the requirement of HR information system ( SAS, 2008) . The main goal of the HR department is to predict how many employees are about to leave the organization. Human capital assets are very important as it involves not only the tangible costs such as replacement, training and loss in the productivity and also the intangible costs such as loss of intellectual capita; and low employee morale. It is important for the organizations to work to retain their employees as it is expensive to hire and train new people(Vipul Mehta ,2008) . The companies are in need of retention model for prediction to anticipate, forecast and analyze information and trends and plan for the future and reduce cost.

If the organization can predict which employees are likely to leave, it is possible to implement programs to retain the employees and help the organization in reducing the cost associated with replacing valuable employees. Predictive modeling combines with information technology to address these challenges. The organizations can use this modeling to predict the likely events based on the past and current data. Predictive modeling helps in applying strategic human capital initiatives to meet the objectives of the organizations which include identifying work force trends and forecasting changes before they occur identifying the unusual patterns in advance before they adversely impact the organization and modeling the voluntary turn over and performance abilities to identify key talent for retention, predicting changes in human capital resources within the organization. (SAS, 2008)

The HR department needs critical information which is provided by the technology and analytical tools so as to access and analyze data from all HR functional areas and employ proper methodology to interpret the data, draw meaningful conclusions and make proper decisions. The advanced technologies help in assimilating the essential data and transform that data into business insight which supports the broad enterprise business plan (SAS, Cary NC, 2003).

The first step of the HR department in this regard is to extract and combine data from the various departments. This integrated information can be examined using relevant metrics and analytics to produce business intelligence which is helpful for the HR professionals in taking strategic decisions(Roselyn, Feinsod, 2006),

Business Intelligence (BI) helps organizations to address the cost and competitive issues associated with employee turnover. The predictive modeling capabilities of Business intelligence to the human capital management solution which identifies critical employees who are likely to resign voluntarily and helps the organization to ease the challenges from turnover as well as aging workforce and economic rebound. With this they can retain the specific employees with specific skills and knowledge which is essential for the success of the organization (SAS Survey, 2007).

These Human Capital Management solutions lead to data mining techniques to find group of employees who are likely to leave the organization. It helps in determining which employee characteristics such as salary, level of education and training, length of service contribute to turnover. The employees are also ranked and assigned probabilities of voluntary leaving within a specific time. The organizations with this insight can proactively align their work force in support of the objectives of the business.

Business Intelligence (BI) refers to the use of technology which collects information and uses it effectively to improve the business (Shmueli, et. al., 2008). As the volume of the information is constantly increasing both internally and externally it is the challenge for BI to use the information effectively for the success of an organization (Smith, W., 2006). Further, the organizations are increasingly becoming knowledge centric; it is required that large number of employees have access to the huge amount of information effectively. It also helps the organizations to do their jobs effectively and also allows analyzing and sharing the information with others. In the era of globalization the usage of Information Technology and IT enabled services is playing a vital role in providing the latest tools to sustain the competition globally (SAS Survey, March 2007). In this situation the HR professionals are required to show the affect of their workforce policies on the overall business plan. This information is not readily available with the HR managers and was not quantified. It is also required by the HR managers to answer how the ROI on the expenditure is measured and whether the turn over issue is resolved and helps in retaining the people and skills required meeting the growth plan of the company (Ang, et. al., 2002). In order to answer these questions strategically the HR managers needs critical information provided by the right technology and analytical tools. BI is helping the HR managers to analyze employee data at an individual level and on the whole (Mehta, 2008). The confidentiality of the employee records and details is ensured with the help of multiple levels of security within the business intelligence. The dependence of HR departments on BI is increasing so as to report on regulatory compliance, understanding the patterns of attrition, recruitment, compensation, retention, job performance which will help HR managers to identify the areas of training investment and revisions in compensation and recruitment (Feinsod, 2006).

Companies face the problem of assimilating the relevant HR data and further utilizing that data in strategic decision making (Verma, 2008). BI can help managers in this regard. The trends can also be observed over a period of time and conduct 'what-if' projections for the future. The human capital assets are important as it involves tangible costs such as replacement and learning and intangible costs such as loss of intellectual capital and low employee morale (Ferrat, et. al., 2005). Attrition and Recruitment are some of the challenges of HR department. HR policies are generally developed with the motto to attract, motivate, engage and retain the right talent by introducing some innovative practices to develop lasting relation ships. Attrition is a problem which almost every sector in India face with high rates. Though there is a boom in the IT/ITES sector the attrition rate is highest. The major challenge in this sector is employee satisfaction (Horwitz, et. al., 2003). The manpower has increased multiple times over the last five years and it is growing contributor to the GDP of India. Attrition is the single largest challenge for BPOs due to lack of growth opportunities in the organization and migration to more stable work environments with higher pay scales (Moore, 2000). In order to encounter these problems organizations are focusing on designing and implementing the best retention strategies like career paths and tie ups with educational institutions for post graduation programs (Shaw, et. al., 1998).

Table 1

PROBLEMS OF ATTRITION

CRITICAL ASPECT

AFFECT

Cost of recruitment & selection

Increase

Cost of training & development

Increase

Motivation & self esteem for existing employees

Decrease

Acquaintance with work

Delayed

Skilled employees

Loosing

Work load to the employees

Overload

Administrative issues like payroll management, providing employee benefits & welfare, etc.

Become more critical

Finding of suitable manpower

Hard

Work culture

Non-conducive

Source: Compiled by the authors

OBJECTIVE

To find out the factors for attrition in an IT companies and to quantify the most important factors to decide upon the strategic decision for reducing attrition

Through this we will study how business intelligence is used in the various human resources practices and processes. In order to see how exactly relevant human resources data can be used to decide upon some strategic decisions aligning to the goal of the company. For this we selected IT sector. Through the data available we try to find out reasons for attrition in the Indian IT sector. We will find out which of the factors are most predetermining for attrition in the sector. For this we use factor analysis in order to come up to the most important factors. Hence forth the integrated information would be used to come up with the most important factor that needs to be addressed immediately through decision tree. Through the entire data analysis we can come up with the most important strategic decision that would eventually be helpful for the company to reduce employee attrition.

METHODOLOGY

For the data collection we adopted the survey method using a convenience sample of 70 respondents. The samples are of 20-30 age group which belong to the major chunk of the population that generally attribute to attrition in IT/ITeS companies. The major challenge would be the collection of data for the study since HR data is generally confidential and companies are hesitant revealing it (Reichheld, 1996).

The major limitations of the study is that it is conducted in selected IT/ITES companies in Hyderabad, Andhra Pradesh, India, which indicates that for generalization of the results, the study can be extended further.

Procedure

The study involved the use of primary as well as secondary sources of data. For primary data, questionnaire was the main tool. The questionnaire contains the variables which were to be ranked in terms of importance ranging from least important to highly important. These variables were decided on the basis of external secondary data. We mainly relied on data obtained from research articles, books and online articles and surveys for external secondary data to decide the variables for the questionnaire. The sources are given at the end in the reference. We have mainly relied on online articles especially articles by various industry experts wherein they have suggested factors due to which employee generally walk out. Surveys done by ―Data-questâ€- and ―NASSCOMâ€- have also been part of the study. Not only this, on the other hand we have also studied the employees' point of view through various blogs by present and ex-employees in IT industry. We used factor analysis technique to analyze the important influential factors and the Bartlett's test of sphericity and KMO measure to examine the appropriateness of factor analysis. The decision tree approach was used to determine the most important factors from amongst all the factors.

Measures

The questionnaire consisted of a list of variables given below. All the variables were measured with a 5-point Likert scale (1=least important to 5=most important)

Dependent variable: Attrition

Independent variables

1) Job Clarity

11) Feedback about performance

2) Job expectation

12) Career growth in org.

3) Career Goals

13) Individual development

4) Performance factors

14) culture of respect for outstanding work

5) Appraisal process

15) Work stress

6) Rewarding system

16) Work-life balance

7) Grievance cell

17) Recognition

8) Communication from organization

18) Relationship with management

9) Quality of supervision

19) Work timings

10) Environment to speak freely

20) Health problems

Analyses

Factor analysis was used as a data reduction technique to zero upon a reduced set of composite variables from a large number of measured variables. This is done using Eigen values. Only factors with Eigen values greater than 1 are retained. In essence this is like saying that, unless a factor extracts at least as much as the equivalent of one original variable, we drop it. This criterion was proposed by Kaiser (1960), and is probably the one most widely used. For the factor analysis to be appropriate, the variables must be correlated. In practice, this is usually the case. If the correlation between all the variables is small, factor analysis may not be appropriate. I formed the hypothesis as follows:

H0: Variables are uncorrelated in the population

H1: Variables are correlated.

Bartlett's test of sphericity (Bartlett 1950) and the Kaiser-Meyer-Olkin measure of sampling adequacy (Kaiser 1970) helps in assessing the adequacy of their correlation matrices for factor analysis. For a large sample Bartlett's test approximates a chi-square distribution. Consequently it is usually assumed that the sample correlation came from a multivariate normal population with the variables being analyzed being independent.

The Bartlett test therefore forms something of a bottom line test for large samples, but is less reliable for small samples. Very small values of significance (below 0.05) indicate a high probability that there are significant relationships between the variables, whereas higher values (0.1 or above) indicate the data is inappropriate for factor analysis.

The Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy provides an index (between 0 and 1) of the proportion of variance among the variables that might be common variance. The SPSS software package suggests that a KMO near 1.0 supports a factor analysis and that anything less than 0.5 is probably not amenable to useful factor analysis

The variables are grouped under each factor based on the loading values. A loading represents the correlation between the variable and factor. The variable with highest loading is grouped under one factor. But in some cases the factor loadings of one variable may be high in two factors making interpretation difficult. So we go for rotation and get rotated component matrix. Through rotation the factor matrix is transformed into a simpler one that is easier to interpret. Rotation does not affect the communality and the percentage of total variance explained. We used orthogonal rotation with the most commonly used method of rotation called Varimax procedure to identify the variables which are clubbed into different factors. The Principal Component Analysis is used to identify the factors which lead to the attrition in IT industry. Communalities indicate the amount of variance in each variable that is accounted for. A high value of communality indicates that a variable has much in common with the other variables taken as a group. Final communalities show estimates of the variance in each variable accounted for by the factors (or components) in the factor solution.

The decision tree approach helps in classification, through a set of rules, that help identify the most significant factor which when catered to gives the maximum result i.e. help in reducing attrition in the organization. For constructing the decision tree, the data was taken as the sum of the variables under each factor.

RESULTS

Preliminary Analyses

Feasibility of Factor Analysis

In case of Bartlett's test, The P value should be less than .05 to reject the null hypothesis. Here the p value is .000 which indicates that we reject the null hypothesis. So the variables are correlated. The KMO value stands at 0.577 which is higher than 0.5, so factor analysis is appropriate. The communality values obtained using SAS enterprise guide 3.0 for factor analysis clearly shows that the communality for each variable is greater than 0.6 hence all the variables can be included in the factors which explain their influence rate in IT industry. In order to come up with a small number of factors out of the 20 variables that are taken initially data reduction is done and the number of factors is basically decided on Eigen values. In 8 cases the Eigen value is greater than one so it can be said that there are 8 factors [1] .

The different variables can be categorized under the 8 factors based on the highest loadings. For instance the first variable job clarity is categorized under factor 4 since it has the highest loading under factor 4 of 0.7704. Similarly the other variables are also categorized. The identified factors clearly present that in order to increase employee retention the companies need to look after employee satisfaction through employee motivation and employee engagement. It's certainly not practical for a company to cater to all the listed factors. It would be more appropriate if we could find the significance of each factor so that the most significant factor can be dealt with immediate effect. The methodology is not far to seek since this can be done by using decision tree. Decision tree is another analytical tool in SAS that would further help in business intelligence.

FEASIBILITY OF FACTOR ANALYSIS

Table 2

KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of SamplingAdequacy

Bartlett's Test Approx. Chi-Square

Of Sphericity df

Sig.

.577

827.122

153

.000

Table 3

Final Communality Estimates: Total = 12.46129

Job clarity

0.63923311

Health problems

0.73997218

Rewarding System

0.68745580

Environment to speak freely

0.65085387

Feedback about performance

0.68914735

Appraisal process

0.75744318

Quality of supervision

0.70725274

Work timings

0.75923380

Work-life balance

0.76964977

Career growth in org#

0.65564119

Grievance cell

0.67111251

Individual development

0.71967997

Work stress

0.63884034

Recognition

0.71341863

Job expectation

0.61885918

Relationship with management

0.67651331

Culture of respect for outstanding work

0.68407884

Communication from org#

0.68290475

EIGEN VALUES

Table 4

Prior Communality Estimates = ONE

Eigen values of the Correlation Matrix: Total = 18 Average =

Eigen value

Difference

Proportion

Cumulative

1

2.16346334

0.19732548

0.1202

0.1202

2

1.96613785

0.26614631

0.1092

0.2294

3

1.69999154

0.09602463

0.0944

0.3239

4

1.60396691

0.07975690

0.0891

0.4130

5

1.52421001

0.25556575

0.0847

0.4977

6

1.26864426

0.10645405

0.0705

0.5681

7

1.16219022

0.08950386

0.0646

0.6327

8

1.07268636

0.15941429

0.0596

0.6923

9

0.91327207

0.14085939

0.0507

0.7430

10

0.77241267

0.00174694

0.0429

0.785

11

0.77066573

0.14722076

0.0428

0.8288

12

0.62344498

0.05778851

0.0346

0.8634

13

0.56565647

0.06646056

0.0314

0.8948

14

0.49919591

0.06587433

0.0277

0.9226

15

0.43332158

0.05978052

0.0241

0.9466

16

0.37354106

0.04742391

0.0208

0.9674

17

0.32611715

0.06503525

0.0181

0.9855

18

0.26108190

0.0145

1.0000

Table 5: Factor Pattern

Factor1

Factor2

Factor3

Factor4

Factor5

Factor6

Factor7

Factor8

Job clarity

0.2664

-0.546

0.0191

0.1850

-0.081

-0.179

0.2924

0.3335

Health problems

-0.289

-0.409

-0.083

0.3339

0.3400

0.4779

0.0915

0.1338

Rewarding System

0.0616

0.1259

0.2649

-0.449

-0.015

0.6106

-0.148

-0.028

Environment to speak freely

-0.087

-0.358

-0.203

0.3594

-0.377

0.1109

-0.409

0.211

Feedback about performance

-0.035

0.6500

0.3147

0.2441

-0.143

-0.005

0.0359

0.2915

Appraisal process

-0.241

0.3551

-0.583

0.0648

0.3755

0.0409

0.2899

0.0543

Quality of supervision

0.5344

-0.072

0.0665

0.1854

-0.436

0.2722

-0.102

-0.320

Work timings

-0.492

-0.254

0.3503

-0.293

-0.087

-0.042

0.4315

0.2201

Work-life balance

-0.329

0.2650

-0.285

0.4923

-0.105

0.3233

0.3484

-0.172

Career growth in org#

0.4359

0.0563

0.2486

0.4349

0.4251

-0.037

0.1580

-0.066

Grievance cell

0.4221

-0.102

-0.175

0.0294

0.5693

0.2226

-0.264

0.085

Individual development

0.5375

0.4201

0.3072

0.2202

-0.010

-0.18045

0.25754

-0.11128

Work stress

-0.117

-0.236

0.4364

0.0654

0.0560

0.38240

0.24003

-0.40912

Recognition

0.2355

0.199

0.0383

-0.085

-0.390

0.40409

0.23432

0.48846

Job expectation

0.46038

-0.36

-0.326

0.2167

-0.273

-0.00611

0.21837

0.06199

Relationship with management

0.51087

-0.197

0.1034

-0.382

0.3672

0.07367

0.17622

0.22347

Culture of respect …

0.28807

0.39352

-0.54368

-0.31983

-0.090

0.18782

0.04356

0.05557

Communication from org#

-0.14124

0.24897

0.37792

0.39312

0.182

0.08452

-0.34755

0.37758

Analysis

A loading represents the correlation between the variable and factor. For instance 0.2664 represents the correlation between the first variable and first factor. The variable with highest loading is grouped under one factor. But in some cases the factor loadings of one variable may be high in two factors making interpretation difficult. So we go for rotation and get rotated component matrix.

Table 6: Rotated Factor Pattern

Rotated Factor Pattern

Factor1

Factor2

Factor3

Factor4

Factor5

Factor6

Factor7

Factor8

Job clarity

-0.019

0.07462

-0.16080

0.77004

-0.06536

0.07864

-0.05020

0.03886

Health problems

-0.419

0.34683

0.38057

0.20128

-0.14298

0.46406

0.14936

-0.01960

Rewarding System

-0.091

0.18326

-0.20848

-0.49599

0.04822

0.28069

-0.05097

0.52200

Environment to speak freely

-0.609

-0.08706

-0.01300

0.29600

0.34751

-0.06945

0.23876

-0.03536

Feedback about performance

0.3641

-0.25567

0.12602

-0.15147

-0.01769

-0.15399

0.59317

0.27659

Appraisal process

0.0439

0.18153

0.71253

-0.11734

-0.29977

-0.30659

-0.12349

-0.04457

Quality of supervision

0.1077

-0.03961

-0.12678

0.09712

0.76620

0.17102

-0.11545

0.19733

Work timings

-0.092

-0.35760

-0.13112

0.12603

-0.65227

0.36340

-0.06851

0.16600

Work-life balance

-0.011

-0.23270

0.81411

-0.00929

0.12593

0.16883

0.07564

0.04937

Career growth in org#

0.5232

0.40767

0.07914

0.22536

0.15049

0.19469

0.25604

-0.18041

Grievance cell

-0.022

0.80291

-0.01658

-0.01277

0.12801

-0.07525

0.03506

-0.04748

Individual development

0.7843

-0.00837

-0.03336

0.08650

0.25514

-0.03936

0.15395

0.07401

Work stress

0.0806

-0.03176

0.01953

-0.09892

0.05278

0.78181

-0.08456

-0.00056

Recognition

0.0387

-0.07267

0.03936

0.16375

0.06114

-0.07340

0.11918

0.80929

Job expectation

-0.019

0.07026

0.06220

0.61778

0.36159

-0.08522

-0.27444

0.12129

Relationship with management

0.2437

0.56480

-0.32961

0.16648

-0.19771

0.02264

-0.22912

0.26396

Culture of respect for outstanding work

0.0596

0.15980

0.23377

-0.20769

0.14855

-0.50135

-0.34227

0.40822

Communication from org#

-0.027

0.10391

-0.02069

-0.09713

-0.04323

0.02580

0.81176

-0.00414

NEED FOR DECISION TREE

Now that the different factors for attrition are determined, it's up to the organization to cater to them. These factors will help them work upon the grey areas so as to increase employee satisfaction, helping them form a bond with the organization hence increasing their commitment. It's very easy for any consultant to suggest "n" no. of factors for the organization to work upon but is it feasible for the organization to look to all the "n" factors? The answer is certainly "No". Here comes the role of business intelligence. Through BI a consultant can actually determine the most important of all factors by classifying them, identify the 2 or 3 three most significant domains working upon which will give the maximum result. Before investing into any initiative a company should always analyze the return on investment since then only the problem can be solved. And for calculating the ROI first one needs to know which would be the best areas or programs to invest. In this paper eight factors have been identified that actually explains the different domains that contribute to attrition. Now, through decision tree that helps in classification, through a set of rules, we tried to find out the most significant factor which when catered to give the maximum result i.e. help in reducing attrition in the organization. For constructing the decision tree, the data was taken as the sum of the variables under each factor. For example: the data for factor 1 includes the sum of the responses under the variables communication from management and feedback about performance.

TARGET VARIABLE: GENDER

Gender was taken as the target variable since when we talk about the definition of the word satisfaction and motivation for the two genders it's significantly different. Motivators for a female employee and that of a male employee can be different. Similarly the reasons for quitting an organization is different for the two genders, hence gender has been selected as the target variable. The target variable has been categorized as males or 1 and females or 0.

The Model assessment measure has been kept automatic

MISCLASSIFICATION RATE

Figure 1 shows the plot of misclassification rate on the training sets for the different number of leaves. The no. of leaves is 8 and as the graph shows there is almost same rate of misclassification for 3rd 4th 5th 6th 7th and 8th leaf and that too the rate is quite small. The variation between the leaves is large for the first three leaves. The rest of the leaves have zero variation among the error rates on training sets.

Figure 1

Table 8

Level

Training

1

0.3143

2

0.2429

3

0.1429

4

0.1429

5

0.1429

6

0.1429

7

0.1429

8

0.1423

RESULT: INTERPRETATION

Figure 2: Decision Tree

The figure 2 shows the decision tree for the required no. of factors and the target variable i.e. gender. For interpreting the results first it's important to bring into light is that out of the eight factors four have been eliminated and thus the rest four have been considered significant. They are factor 2, 5, 6 and 8. The tree starts with factor 8; factor 8 consists of values 6, 7, 8, 9, 10, 11, 12; so the trunk has split into two branches one with the values 6, 7, 8 & 9 and the other with the values 10, 11 & 12. This explains that the left branch consists of employees who think factor 8 is comparatively less important than other while the right side consists of employee who have rated factor 8 most important. For convenience I have named each node as A, B, C, D, E, F, G, H, I, J, K, L, M, N.

Figure 3

Figure 3 shows that A has 12.22% of males (1) only this shows that males have rated factor 8 has very important for them. Out of the 87.8% females who have rated factor 8 as less important, majority of them have rated factor two less than 8. This shows that for majority females, factor 8 and factor 2 are less important. Consequently if we see node D we find that 19 females who have given higher importance to factor 2, most of them (13) have given high importance to factor 6. Henceforth the majority about 11 females as we can see at node M have rated factor 5 as highly important.

Coming to the node B, it consists of employees who have voted factor 8 as highly important that is 10, 11 & 12. Out of the entire population we have majority males voting highly for factor 8, out of the 17 males 15 have given lower importance to factor six which consists of values from 6 to 14. But out of the 15 males 9 have given high importance to factor 2. As far as females are considered again here we see that out of the 3 females who have ranked factor 6 as less important all of them have given low importance to factor 2.

Now talking about the node F consists of employees who have given higher importance to factor 6, out of the 2 males, one has given higher importance to factor 8 and the other less importance so we can't conclude as such anything concrete for males. But for females out of the 9 who have ranked majority of them (5) have given lower importance to factor 8. Describing the factors the above mentioned result can be interpreted as follows, factor 8 is rewards & recognition, factor 6 is working conditions, factor 5 is job clarity and expectation and factor 2 is career development. As the decision tree shows the following rules can be derived out of the decision tree:

IF FACTOR_2 IS ONE OF: 5, 7, 8 (low importance)

AND FACTOR_8 IS ONE OF: 6, 7, 8, 9 (lower importance)

THEN

NODE : 4

N : 17

1 : 0.0%

0 : 100.0%

IF FACTOR_6 IS ONE OF: 6, 7, 8, 9 (lower importance)

AND FACTOR_2 IS ONE OF: 9, 10 (higher importance)

AND FACTOR_8 IS ONE OF: 6, 7, 8, 9 (lower importance)

THEN

NODE : 8

N : 6

1 : 50.0%

0 : 50.0%

IF FACTOR_2 IS ONE OF: 5 7 8 (lower importance)

AND FACTOR_6 IS ONE OF: 6, 7, 8, 9 (lower importance)

AND FACTOR_8 IS ONE OF: 10, 11, 12 (higher importance)

THEN

NODE : 10

N : 9

1 : 66.7%

0 : 33.3%

IF FACTOR_2 IS ONE OF: 9, 10 (higher importance)

AND FACTOR_6 IS ONE OF: 6, 7, 8, 9 (lower importance)

AND FACTOR_8 IS ONE OF: 10, 11, 12 (higher importance)

THEN

NODE : 11

N : 9

1 : 100.0%

0 : 0.0%

IF FACTOR_8 IS ONE OF: 10, 11, 12 (higher importance)

AND FACTOR_6 IS ONE OF: 10, 11, 12, 13, 14 (higher importance)

THEN

NODE : 12

N : 6

1 : 16.7%

0 : 83.3%

IF FACTOR_8 IS ONE OF: 10, 11, 12 (higher importance)

AND FACTOR_6 IS ONE OF: 10, 11, 12, 13, 14 (higher importance)

THEN

NODE : 13

N : 5

1 : 20.0%

0 : 80.0%

IF FACTOR_5 IS ONE OF: 4, 5, 6 (lower importance)

AND FACTOR_6 IS ONE OF: 9, 10, 11, 12, 13, 14 (higher importance)

AND FACTOR_2 IS ONE OF: 9, 10 (higher importance)

AND FACTOR_8 IS ONE OF: 6, 7, 8, 9 (lower importance)

THEN

NODE : 14

N : 7

1 : 28.6%

0 : 71.4%

IF FACTOR_5 IS ONE OF: 7, 8 (higher importance)

AND FACTOR_6 IS ONE OF: 9, 10, 11, 12, 13, 14 (higher importance)

AND FACTOR_2 IS ONE OF: 9, 10 (higher importance)

AND FACTOR_8 IS ONE OF: 6, 7, 8, 9 (lower importance)

THEN

NODE : 15

N : 11

1 : 0.0%

0 : 100.0%

Interpreting the rules tell us that

If employee is a male and ranks "rewards & recognition" high, then he also ranks "career development" high.

If employee is a female she ranks "rewards & recognition" low, then she also ranks "career development "low but she ranks working conditions high.

If employee is a male and ranks "rewards & recognition" low then also he ranks career development high and job clarity and expectations as low.

If employee is a female and ranks "rewards & recognition" high, "career development as low" but ranks "working conditions as high".

So, the crux of the entire study can be detailed as follows:

The study shows that the factors for attrition are different for males and females.

For males the most important factor is rewards & recognition followed by career development.

For females the most important factor is working condition.

CONCLUSION

The paper basically deals with the use of business intelligence in human resources management. In order to do so, we tried to use business intelligence with SAS to solve one of the most prominent problems in IT/ITeS companies in India i.e. attrition. For this purpose a questionnaire was designed consisting of twenty variables decided on the basis of secondary data research. Employees of the some IT/ITES organizations were asked to rank them in order of importance. Based on the responses of the 70 employees, through factor analysis those 20 variables were classified into 8 factors namely communication, career development, performance appraisal, work culture, job clarity & expectations, working conditions, grievance addressal and rewards & recognition.

It's not cost-effective for any organization to invest blindly in "n" number of areas, so through the use of decision tree I tried to find the most important factor for the organization to work upon. The results obtained from the decision tree implied that:

For males the most important factor that results into attrition is lack of proper incentives i.e. "rewards & recognition". Also a proper career development for the individuals boasts the confidence of the employee and motivates him to stay in the company.

For females the most important factor that attributes to attrition is adverse or lack of proper "working conditions". This includes proper timings, work-life balance, work related stress, environment free from physical or mental abuse, sexual harassment and health related issues. Even for females who are keen on rewards & recognition look for good working conditions. So, in order to prohibit a female employee from leaving the company it's very important to provide her with good and adequate working conditions.

The difference in the attitude of both the genders can be attributed to the behavioral difference. Males have dominant traits of aggressiveness, achievement and are growth-oriented whereas for females the important behavioral indicators include vulnerability of both physical & mental space and affiliation. Females tend to at times over look factors like career development since most of the cases in India females work for a limited period of time because of marriage or family related issues hence for them security and safety are of greater importance.

RECOMMENDATIONS

Human Resources Department in most of the organization is the most neglected function. The primary reason for this is that most of the jobs performed by the HRD is qualitative and hence is difficult to quantify due to which the top management thinks twice before making any investment in any activity related to the HR function on a whole. In order to convince them the HR manger needs to have concrete facts and figure along with reasonable cost-benefit analysis. BI helps this task makes this once thought impossible task possible. As already explained through BI, it was concluded that in order to reduce attrition in the organization the company has to understand the different needs of both genders. On the basis of the gender the organization has to cater to the requisite problem domain.

So, for males the basic concept of work and reward is applicable.

Compensate the employee for every contribution he makes.

Recognize his effort and reward him appropriately the reward needn't be tangible even intangible rewards might work well at times.

Define a proper career path for the individual since the day he feels his growth is stagnated he might quite.

Counsel and coach him in order to help him see his career develop through the entire period of his stay in the company.

For females, since they are not as ambitious as males, providing the best environment would be the best option.

Provide them the option of flexi-timing if possible. Don't force them for over-work.

Mentor them to make a balance between their work and personal life, esp. for married women with small kids provide them baby sitters if possible.

Make sure there are no cases of physical or mental abuse since this reason is sufficient for a female employee to quit the very day.

Provide proper leaves, including provision for leave without pay, sabbaticals, and provide assistance during major health issues.

Writing Services

Essay Writing
Service

Find out how the very best essay writing service can help you accomplish more and achieve higher marks today.

Assignment Writing Service

From complicated assignments to tricky tasks, our experts can tackle virtually any question thrown at them.

Dissertation Writing Service

A dissertation (also known as a thesis or research project) is probably the most important piece of work for any student! From full dissertations to individual chapters, we’re on hand to support you.

Coursework Writing Service

Our expert qualified writers can help you get your coursework right first time, every time.

Dissertation Proposal Service

The first step to completing a dissertation is to create a proposal that talks about what you wish to do. Our experts can design suitable methodologies - perfect to help you get started with a dissertation.

Report Writing
Service

Reports for any audience. Perfectly structured, professionally written, and tailored to suit your exact requirements.

Essay Skeleton Answer Service

If you’re just looking for some help to get started on an essay, our outline service provides you with a perfect essay plan.

Marking & Proofreading Service

Not sure if your work is hitting the mark? Struggling to get feedback from your lecturer? Our premium marking service was created just for you - get the feedback you deserve now.

Exam Revision
Service

Exams can be one of the most stressful experiences you’ll ever have! Revision is key, and we’re here to help. With custom created revision notes and exam answers, you’ll never feel underprepared again.