UNIVERSITY POLITEHNICA OF BUCHAREST FACULTY OF ENTREPRENEU RSHIP, ENGINEERING AND BUSINESS MANAGEMENT MANAGEMENT OF THE DIGITAL ENTREPRISE… [616576]

UNIVERSITY POLITEHNICA OF BUCHAREST
FACULTY OF ENTREPRENEU RSHIP, ENGINEERING AND
BUSINESS MANAGEMENT
MANAGEMENT OF THE DIGITAL ENTREPRISE

Dissertation Paper
Information systems used for recruitment and selection
at Deloitte Romania

Scientific Coordi nator,
Prof. D r. Eng. Florin DĂNĂLACHE Graduate ,
Georgeta DRĂGHICI

BUCHAREST
2017

TABLE OF CONTENTS
Introduction ………………………….. ………………………….. ………………………….. ………………………….. ……… 3
Chapter 1. Human resources recruitment and selection ………………………….. ………………………….. 4
1.1. RECRUITMENT – an ongoing and fluent process ………………………….. ………………………….. …. 4
1.1.1 Planning the human resources recruitment ………………………….. ………………………….. … 5
1.1.2 Stages of human resources recruitment ………………………….. ………………………….. …….. 6
1.1.3 Sources of the human resources recruitment ………………………….. ………………………….. . 7
1.1.4 Methods of the human resources recruitment ………………………….. ………………………. 10
1.2. SELECTION OF HUMAN RESOURCES – importance and characteristics ……………………. 11
1.2.1 Methods and techniques for human resources selection ………………………….. ……….. 12
1.2.2 The selection process ………………………….. ………………………….. ………………………….. .. 14
1.2.3 Tests used in selection of human resources ………………………….. …………………………. 14

Chapter 2 . Informational Systems in h uman resources management ………………………….. ……… 16
2.1 HUMAN RESOURCES INFORMATIONAL SYSTEM METHODS ………………………….. …… 16
2.1.1. Mainframe softwares ………………………….. ………………………….. ………………………….. ……….. 17
2.1.2. Web -based techniques ………………………….. ………………………….. ………………………….. …….. 17
2.2 THE NEED FOR THE COMPUTERIZATION OF HUMAN RESOURCES MANAGEMENT
………………………….. ………………………….. ………………………….. ………………………….. ………………….. 18
2.3 THE PROCESS OF RECRUIMENT AND SELECTION TROUGH INFORMATIONAL
SYSTEMS ………………………….. ………………………….. ………………………….. ………………………….. …. 20
2.3.1 The use of informational softwares in recruitment and selection ………………………….. ……….. 20
2.3.2 The use of social media in recruitment and selection ………………………….. ………………………… 21
2.3.3 Recruiting and selection stages using HRIS ………………………….. ………………………….. ….. 21
Chapter 3 . Deloitte’s informational systems used for recruiting and selection case study ……. 25
3.1 .DESCRIPTION OF THE COMPANY ………………………….. ………………………….. ………………….. 25
3.2 DELOITTE‘S RECRUITING AND SELECTION STRATEGY ………………………….. ……………. 26
3.3 THE PROCESS OF RECRUITING AND SELECTION THROUGH INFORMATIONAL
SYSTEMS AT DELOITTE ………………………….. ………………………….. ………………………….. ………. 30
3.3.1 Recruiting process ………………………….. ………………………….. ………………………….. ………………. 33
3.3.2 Selection process ………………………….. ………………………….. ………………………….. ……………….. 37
3.3.3 Assessing the hiring process ………………………….. ………………………….. ………………………….. .. 44
Conclusion ………………………….. ………………………….. ………………………….. ………………………….. ………. 47
Bibliography ………………………….. ………………………….. ………………………….. ………………………….. …… 49
Appendices ………………………….. ………………………….. ………………………….. ………………………….. ……… 52

1
INTRODUCTION
With the increasing effect of globalization and technology, organizations have started to use
information systems in various functions and departments in the last decades. Human Resources
Management is one of the departments that mostly use management info rmation systems. Human
Resources Units in organizations now place much emphasis on sharing information, especially ones
that affect the workforce.
This paper aims to illustrate the process of hiring through modernized recruiting and
selection techniques. Organizations deem it fit to employ HRIS in their transactions. Human
Resource Information System (HRIS) is a computer -based system that is used to manage the
administration of HR processes and procedures. Its purpose is to become more efficient in providi ng
better information for decision making by overlooking to the theoretical basis.
Social and organizational changes in the financial environment are various and extensive.
According to this, it is imperative for human resource management (HRM) to be com prehensive,
high quality, fast, flexible and in track with forthcoming trends. The use of information and
communication technology becomes an imperative for HRM, as well for other company‘s activities.
One of the main issues of successful business is human resource management, and that
practice is under excessive influence of modern information technology. Human Resources
Information Systems (HRIS) are systems used to gather, record, store, analyze and save data relating
to an organization‘s human resources , but it is not simply a reduction of administrative actions. The
first part of the paper approaches theoretical information about both of the subjects under research:
the new methods of hiring and also the evolution of information systems in HR.
The signi ficance of HRIS system is multi layered, going from operational assistance in
collecting, storing and organizing data for reports, simplifying and hastening the processes and
controlling the available data, reducing labor expenses for HR departments, and giving timely and
various data to the management of the organization , based on which it is possible to t ake quali ty
planned decisions connected to human assets .
The purpose of this paper is to highlight the importance of HRIS and to give a
comprehensive insight of the subject. In the first part of the research, the focus in on the process of
recruiting through socia l media and the informational systems used in this process. This contributes
to effective decision making which may guide spending of huma n resources in an organization . To
meet the competition in the global market, firms need to balance their available re sources achieving
the desired profitability and survival. It is believed that management of human resources is the most
important function for every organization that wants to have a competitive edge over its competitors.
This paper therefore researched in to the benefits of HRIS on recruitment and selection, its
contributions to organizations and the challenges associated with its usage, by showing how middle
management uses human resources systems to monitor and analyze the recruitment and perform the
selection process in an organization . In the last part of the research, the concentration is fixed upon
illustrating the usefulness of data mining in selecting the most fitted employee by assessing the
overall characteristics.
The aim of this research is to r eview specific process for understanding the concept of HRIS,
to identify the benefits and contributions of recruitment and selection, the challenges of using this
informational system, profiled findings, conclusion an d recommendations of the study.

2
Chapte r 1. Human resources recruitment and selection – definition, principles,
objectives
1.1. RECRUITMENT – an ongoing and fluent process
The recruitment process is defined as t he procedure of finding and procuring the best –
qualified applicant (from inside or outside of an association) for an employment opportunity, in a
convenient and financially savvy way. The enrollment process incorporates investigating the
necessities of a vocation, pulling in workers to that occupation, screening and selecting candidates ,
procuring, and coordinatin g the new representative to the company.
A recruitment process is the system that associations use to fill work opportunities and
contract new ability into their association. Be that as it may, the opportunities ought to obviously in
a perfect world be loaded with qualified hopefuls and not just any individual who tri es to apply. The
enlistment capacity can be overseen through inside ability securing or completely or incompletely
outsourced to an enrollment organization. The cost effectiveness made by representative referrals
has likewise set more prominent inclusion a nd spotlight on the workers of t he enlisting divisions.
The recruitment process starts first with an opening made by another position, an end of a
worker, advancement , or a representative's renunciation. Next the position is assessed and the
expected set of responsibilities is composed to match it. After that, the capabilities for hopefuls are
composed. Next, the association posts the employment on a site, in a pamphlet, through a vocation
board, in the media, and so forth. The applications are then experi enced, and the ones from unfit
competitors are tossed .
1.1.1. Planning the human resources recruitment
With a specific end goal to expand proficiency in in hiring and retention and to ensure
consistency and compliance in the recruitment process, it is sugg ested the accompanying steps be
taken after. Subtle elements for every stride incorporate the base prescribed best practice to draw in
a gift ed and differing candidate group .
Recruiters need to keep in touch to changes in the labor market to make sure that their
recruitment efforts are not ineffective or directed at too small a group of work . Skill deficiencies
may occur without prior notice and recruitment and training processes need to be maintained
flexible. It is recommended for any organization to plan its labor force requirements, matching
available supply against forecast demand. To properly asses this, an audit of existing staff may be
needed. A straightforward strategy will help organizations to:
• assess future recruitment needs
• convey training programs
• create promotion and career development policies
• anticipate and, where possible, avoid redundancies
• develop a flexible personnel to meet future changing requirements
• control employees costs while ensuring salaries stay competitive
• assess future requirements for equipment, technology and business locations .
Management is accountable for generating the human resource plan, senior management for
supporting it. Implementation is likely to be most effective if it has the workforce suppor t, normally
attained through consultation. Making a human resource plan involves:

3
• forecasting staffing requirements against business objectives
• assessing the available supply of people to meet those requirements
• matching available supply against forecast demand
Employers also need to encourage a good ‗work -life balance‘ within the organization by
giving consideration to more flexible ways of working. Employees have the right to request flexible
working arrangements – including job sharing , part -time working, flexi -time, working from
home/teleworking and school time contracts – and employers must have a good business reason for
rejecting any application.
1.1.2. Stages of human resources recruitment
The recruitment process starts when you know you requ ire another person in the
Department, either in light of the fact that a current staff part has left, or in light of the fact that there
is new work to be finished. It doesn't complete until after the arrangement has been made. Each
stage of the human reso urces life cycle has its own challenges, opportunities, and benefits .

Fig. 1.1 .The stages of HR recruitment process (Taylor, 2005 )
Hiring decisions play a critical role in turnover, productivity, and growth. In order to succeed
in the recruitment phase of the HR recruitment stages ( Fig.1.1.) , your huma n resources department
needs to m ake a key staffing arrangement that incorporates understanding positions that should be
filled, what will be anticipated from a representative, a system for pulling in the most elite, and other
enlisting concerns .

1 •Identify Vacancy

2 •Prepare Job Description and person Specification

3 •Advertise

4 •Managing the Response

5 •Short -listing

6 •References

7 •Arrange Interviews

8 •Conduct The Interview

9 •Decision Making

10 •Convey The Decision

11 •Appointment Action

4
1.1.3.Sources of the human resources recruitment
The search of proper candidates and notifying them about the job openings in the enterprise
is the most important aspect of the recruitment process. The candidates may be existing inside or be
available outside the organization. Recruitment sources refers to the first point of the process of
identifying and encouraging prospecti ve employees to apply for jobs. The recruitment sources are
said to be posi tive in its approach as it seeks to attract as many candidates as possible.
There are two sources of recruitment:
 internal sources : Most fitted employees can be found within the organization . When a vacancy
appears in the organization , it may be suggested to an employee who is already on the pay -roll.
Internal sources , as listed below, include promotion, transfer and in certain cases downgrading .
When an advanced post is given to an worthy employee, it motivates all other company‘s
employees to work hard. T he employees can be informed of vacancies by internal advertisement.
 external sources : sources of recruitment are from outside the organization . An external source
involves recruiting candidates externally for better performance. External employees always feel
young and inn ovative for new companies, energetic enough to find business problems and fix it.

Fig. 1.2 .The types of HR recruitment sources
As listed in fig 1.2., the main recruitment sou rces used by a com pany to find proper
employees, th is task includes researching profiles that exactly fit the role. Among the sources listed
below, recruiters find the profiles on job boards, social sites, references, resume database, and
through head hunting techniques.
Sources of recruitment
Internal
Transfer
Promotion
Present employees
External
Advertisement
Employment Exchanges
Schools and Universities
Recommendation
Factory Gates
Casual Callers
Former Employees
Internet
Other sources

5
The Internal Sources:
1. Transfers: Transfer i nvolves changing of personnel from current jobs to other similar jobs.
These do not include any change in rank, responsibility or status . The amount of individuals
does not increase with transfers.
2. Promotions: Promotions means shifting of personnel to positions that have better status ,
complex responsibilities and more remuneration . The higher vacant positions may be filled
up from inside the organization . A promotion does not increase the number of personnel in
the organization . Promotion can motiv ate other existing employees to improve their
performance s so that they can also get a promotion.
3. Present Employees: The present employees of most companies are informed about the
possible vacant positions. The employees recommend their relations or person s closely
known of them. Organization is relieved of considering other potential candidates. The
individuals recommended by the employees may be commonly suitable for the jobs because
they know the re quirements of different positions .
All organizations have to use external sources for recruitment to higher positions when present
employees are not apt. More individuals are needed when expansions are carrying out.
The External Sources:
1. Advertisement s: It is a method of recruitment commonly used for skille d employees ,
secretarial and higher staff. Advertisement can be set in newspapers and specialized journals.
These advertisements attract candidates in great number of highly variable quality. Preparing
a proper advertisement is a specialized task. If a firm wants to hide its name, a ‗blind
advertisement‘ may be given requesting the applicants to apply through a specialized
advertising agency.
2. Employment Exchanges: For unqualified , semi -skilled, skilled, office posts etc., it is often
used as a source of recruitment. In certain circumstances it has been made compulsory for
the business to notify their vacancies to the employment exchange. In the past, employers
used to turn to these agencies only as a last optio n. The job seekers and job givers are
brought into contact by the employment exchanges.
3. Schools, Colleges and Universities : is direct recruitment from educational institutions for
certain jobs which need technical or specialized qualification ; has become a common
practice. A close tie between the firm and educational institutions helps in receiving suitable
candidates, j unior level executives or managerial trainees may be or may not be recruited in
this way.
4. Recommendation of Existing Employees : The current employees know equally the
company and the candidate being recommended. Therefore some companies encourage their
current employees to assist them in receiving applications from personnel who are known to
them. In certain cases rewards may also be given if applicants recommended by them are
actually selected by the firm. The down side is i f recommendation leads to nepotism , it will
damage the morale of employees.
5. Factory Gates: Certain workforces present themselves at the factory gate daily for
employment. This way of recruitment is very popular in third world countries for unskilled
or semi -skilled labor . The major drawback of this method is that the person selected may not
be suitable for the vacancy.

6
6. Casual Callers: Those workers who casually come to the enterprise for employment may
also b e considered for the vacant position . It is the most cost-effective method of
recruitment. In the advanced states , this technique of recruitment is very widespread .
7. Central Application File: A file of past candidates that weren‘ t selected previous may be
kept. In order to keep the file active , applications in the files must be checked at some
intervals.
8. Labor Unions: In some professions like construction, hotels, maritime industry etc.,
(activities with employment instability) all human resources usually come from unions. It is
beneficial from the managerial point of view because it saves expenses of recruitment.
Nevertheless , in other businesses , unions may be asked to recommend candidates either as a
benevolence gestu re or as a consideration towards the union.
9. Labor Contractors: The contractors stay themselves in touch with the labor market,
bring ing the employees at the specific places where they are required. They get commission
for the number of individuals supplied by them.
10. Former Employees: In instance s employees have been jobless or have left the factory at
their own, they may be taken back if they manifest their interest re joining the organization.
11. Internet /Social Media : nowadays is the most utilized mean of recruiting employees and it is
widely spread and used by big corporations and also by SME‘s. LinkedIn, Twitter and
Facebook are good alternative recruiting sources.
12. Other Sources : Separately from these main sources of exte rnal recruitment, there are some
other sources which are used by companies from time to time. These consist of special
lectures held by recruiter in different organizations , although apparently these lectures do not
have an impact on recrui tment directly.
Recruitment is the first step in the long -term commitment between management and the
employee. Recruitment is the process of attempting to locate and encourage potential applicants to
apply for existing or anticipated job openings.
Tab 1.1 . The main advantages/disadvantages of the recruitment sources
Recruitment Sources Internal External

Advantages 1) Easiest
2) Less time consuming
3) No deep interview required
4) Promoting to high position
increases the identity of
employees thereby long stay
with the company.
5) Competition among old
employees to occupy high level
when co -employee gets
promoted
6) Economy in Training Costs
7) No hasty d ecision
8) Improves morale of all the
employees
9) No Error in Selection , ease
to identify the bes t candidate 1) Create s new opportunities for
job seekers and experience
employees
2) Branding increases externally.
3) Rich source of candidates
4) Increases business strategy
because external candidates of
unique skills can take the business
to extra mil e.
5) No partiality among unique
employees.
6) Ability to recruit the candidates
with special skills employer need.
7) No need of training as
experienced external employees
are filled with the skills it needs.
8) New innovative ideas
9) High productivity

7
Recruitment Sources Internal External

Disadvantages 1) Discrimination may exist in
respective to language, religion
for promotion.
2) Dissatisfied co -employee
can quit the position because of
jealous or partiality
3) No opportunities for
external employees.
4) External candidates seeking
employment think about
discrimination
5) Lack of innovative ness 1) More time consuming
2) High cost
3) Sourcing the right quality
profiles becomes difficult
4) External candidates ask for
more salary and benefits
5) Candidates may quit the job
within short period if
uncomfortable with it .

For an organization it is ethical to recruit within the organization. Given the difficulties with
in-house recruitment methods (as viewed in Tab.1.1) , some managers may choose to use an external
recruitment agency instead, choosing such an agency may help organization recruiting the suitable
candidates, saving the employers the trouble of sorting out short list and concentrating on other
important functions, like selection decision, which still re sts with the organization, but also it
enables an organization to source talent which it may not otherwise have any access to, it can
approach the employees of their competitors, and it is up -to-dated the latest recruitment innovations.
1.1.4. Methods of t he human resources recruitment
Newest recruitment methods take account of online and social recruiting, mobile out -reach
and employment branding campaigns that are run ning in the same time with the efforts of the
marketing department. Develop ing a better strategy for recruiting for your company through daily
recruiting career tips, technology reviews, and following recruiting trends from the market are key
strategies for any organization .
Each recruitment method can be judged on its efficacy by v arious metrics, such as number of
sourced candidates, the total cost of hire, an d the long term success of hiring process. Some
examples:
 Headhunting – Headhunting is the practice of attracting employees from other companies by
offering them more compensati on and better benefits and circumstances . Some firms take a n
unclear view of this practice, while others take part in it broadly .. The disadvantage of getting
people through headhunting is that they are less likely to display a high level of loyalty .
 Socia l Networking – The employment potentials of online networking sites such as Twitter and
Facebook are being discovered by many corporations . Because of the interconnected nature of
these Web systems, a job opening message can be passed from person to person and reach
thousands of people within a short time. Because individuals with similar interests, careers and
skills tend to be linked to each other, a job publication can reach a large number of people who
are familiar wi th the work and are qualified for the opening .
 HRIS (human resources informational system). A HRIS, which is also identified as a human
resource information system or human resource management system (HRMS), is mainly a
juncture of human resources and info rmation technology through HR software. This allows HR
activities and processes to happen automatically . A HRIS may be regarded as a mean , through

8
software, for companies big and small to take care of a number of activities, including those
related to huma n resources, accounting, management, and payroll. A HRIS permits a company
to plan its HR expenses more efficiently , as well as to manage and control them without
requiring assigning too many resources to them.
Costs of Recruitment
Part o f recruitment planning includes budgeting the cost of finding applicants. Here is how
you can calculate the cost of recruitment :
cost per hire = advertising costs + recruiter costs + referral costs + s ocial media costs
+event costs

In addition, when assessing how effective the recruiting methods are, the value of yield ratio
can be determined . A yield ratio is the percentage of applicants from one source who make it to the
next stage in the selection process (e.g., they get an interv iew). By using these calculations, it can be
determine d the best place to rec ruit for a particular position. Some yield ratios may vary for
particular jobs, and a higher yield ratio must also consider the cost of that method, too. For an entry –
level job, c ampus recruiting may yield a better ratio than, say, a corporate recruiter, but the
corporate recruiter may have higher cost per hires.
The cost of a n informational system for HR can range anywhere from small amounts to
large ones, based on the size of the business and the needs.: a small business (10 -50 employees),
should look to see if the vendor o ffers a small business solution . Growing midsize businesses (50 –
1,000 employees) can benefit significantly from the more advance d HR features offered by client
server software. The organization can expect to pay more for more features, also can expect sizable
implementation costs, training costs and other associated costs , precise software modules, which
would satisfy to any of their HR needs. Therefore HRIS would soon be an integral part of HR
activities in all organization. After the recruiting process is finished , it can begin the selection
process .
1.2. SELECTION OF HUMAN RESOURCES – importance and characteristics
Selection is the process of picking up individuals (out of t he pool of job applicants) with
requisite qualifications and competence to fill job s in the organization. Selecti ng candidates involves
two main processes: short l isting and assess ing applicants to decide who should be made a job offer.
“Selection is the process of differentiating between applicants in order to identify and hire
those with a greater likelihood of success in a job.” (Beardwell, Holden, 2000).
It is important to use th e correct selection methods because:
 making a bad selection decision is costly in terms of the charge of the recruitment and selection
process itself, salary, management time and resources;
 if it is select ed some­one unsuitable , it might need to spend time managing th eir performance
and it might af fect other employees. They may also decide to leave meaning redoing the
recruitment process again;
 this is a good opportunity to assess skills and abilities, to help put together a n
induction/proba tion training pl an and for the future development of the employees .

9
1.2.1 . Methods and techniques for human resources selection
Different types of positions require different kinds of selection techniques. Choosing the
right techniques will help to recruit the best person for the position. The selection techniques used
will depend on the particular skills, attributes and knowledge required for the position. The HR
department must be able to match the selection method with the selection criteria that are key to the
position.
The next stage is to carry out a selection process so you can establish whic h of the shortlisted
applicants is most suitable for the role. Interviewing is the most commonly used method of assessing
prospective employees and all selection programs must include a panel interview as a minimum.
There are other methods available to assist the decision making process, for instance, psychometric
testing, aptitude tests, presentations, teaching demonstrations and work based exercises e.g. typing
exercise, data entry or creating a spreadsheet.
A diversity of methods are available and attention needs to be given as to which ar e suitable
for a particular position . The methods described here are:
 Online screening and shortlisting
 Interviews
 Psychometric tests
 Ability and aptitude testing
 Personality profiling
Online screening and shortlisting – Initial selection of application based on an evaluation
of a applicant's experience and qualifications compared to the job's requirements. On line systems are
now accessible which can filter ap plications automatically. An online selection tool screens
applications compared to set criteria through key word searches. It may also be responsible for
scoring tools. The use of online screening may be a convenient way of filtering rapidly through a
huge volume of applications or organizing applications, if there is a big amount of jobs that has been
advertised at the same time (for example through selecting applications with the same job reference
digit). This tactic necessitates that all applications are received online and hence having links with
the usage of the Internet as a recruitment device. Some organizations that use this tactic had to
develop extended submission forms in order to certify that the applicant includes all right
information in sufficient detail for their application to be chosen through the word search. Other
organizations only use this method as a first step in trying to select applicants and then go through
the applications once more manually , to double check that all proper applications have been
carefully chosen . As with the use of the Internet for promoting jobs, there is a preliminary cost in
applying online screening and shortlisting, such as the cost of software and training. These expenses
need to be well-adjusted against future probable benefits, such as improving the speed of the
recruitment and selection process.
Interviews – Structured interviews are the most effective type of interview s. The interview
procedure is made through identification of the main job requirements and a list of questions is
created . A selection of interviewers works through all sets of que stions with each candidate and
marks their answe rs. At the end of the interview, the total score is considered and the best candidate
is selected . If extra selection methods are preferred this is provided in the overall process at the end
and once more the best fitted applicant is offered the employment . Even when the interview is
structured, this does not mean that follow up questions cannot be asked to analyze more deeply into
an applicant's skill set and experience. An interview that does not do this, but instead sticks to a
Presentations
Group exercises
Assessm ent centers
References

10
inflexible set of questions, will not let the interviewer to obtain the information necessary to make a
appropriate hiring decision. Members involved in the interview process have to be trained in
interviewing skil ls and the sort of questions they should or should not ask.
Tests: A variety of characteristics are best tested over psychometric testing. This word is
often confusing but in core, means that a ―mental measure‖ is used. Hence , psyc hometric testing
covers a bility testing , aptitude testing and personality profiling. It is necessary that applicant s are
offered feedback on their performance and that applicants understand how the test information is
used in the course of decision making processes. Personality tests are usually measured in
combination with other information, as the results can be unreliable. Ability tests can also be used in
this mode , but many organizations use cut off marks so that all applicants who attain less t han a
definite mark are rejected , while the rest progress to the next phase . Organizations should be guided
before atten ding for interview that testing or profiling will be used as measure of the selection
process. T hey must also be provided in adv ance wit h a sample of the type of test that they are going
to take. Ability tests may cover a variety of areas including: overall intelligence; oral ability;
numerical ability; spatial ability; office capacity; grammatical skill; mechanical aptitude ; sensory
and motor skills .
Personality profiling – An individual's personality may affect their suitability for some
positions . However, personality profiling does not have a correct or incorrect answer and individuals
may be examined through identifying certain behaviors that are better suited for the job than others.
It is foremost important for those undertaking the selection process to be clear on the characteristics
required to perform good at the job.
Presentations – Presentations are often used as a se lection tool, mainly in senior experienced
jobs. The candidate is provided with a topic and given a timeframe to deliver a presentation on that
specific topic. In some organizations , the presentation topic is sent with the interview letter.
Group exercise s – Candidates are given a theme or a role play exercise and are invited to
discuss the theme or the role play in a group. During this, observers who are looking for specific
characteristics grant marks to each applicant . This evidently means that one candidate may need four
or more other candidates to make this a significant exercise and enough people to mark the
candidates correctly .
References – References are also used as a selection method. Sometimes , inadequate
references may affec t decisions to appoint individuals.
 Which methods are fitted for a particular job?
When choosing which selection method is most suitable , it is compulsory to have in mind
the job requirements , through analysis of the job or role description and person specification, and
what abilities , experience and skills are being search ed for. Some organizations make reference on
the individual specification , as to how they will consider that requirement during the selection
process, so t hat possible applicants are aware of the methods the society will use. Certainly , more
than one selection process may be ap propriate for a specific job. It is also important to assess the
success of the selection technique to make sure that it is effective. This can be completed through
looking for the views of applicants who have carried out the selection process or analyzing
recruitment statistics and turnover rates. It is also helpful to appraise the selection methods that were
used the last time that the position was vacant and check that they are still applicable and suitable to
the present selection process, particularly if the re have been changes to the job.

11
1.2.2. The selection process
A scientific and logical recruitment procedure leads to scientific selection of candidates. The
criteri a based on selecting a candidate for a particular job varies from company to company.
Therefore, the selection proce ss followed by different organizations can become lengthy as it is a
quest ion of getting the most suitable candidates for which various tests and interviews have to be
taken. The proce ss of selection should be systematic so that it does not leave any scope for
confusions and doubts about the choice of the selected candidate . Process/ selection steps:
1. Preliminary Interview : The purpose of preliminary interviews is basically to eliminate
unqualified applications based on information supplied in application forms. The basic
objective is to reject misfits. On the other hands prelimin ary interviews is often called a
courtesy interview and is a good public relations exercise.
2. Selection Tests: Jobseekers who past the preliminary interviews are called for tests. There
are various types of tests conducted depending upon the jobs and the co mpany. These tests
can be Aptitude Tests, Personality Tests, and Ability Tests and are conducted to judge how
well an individual can perform tasks related to the job .
3. Employment Interview: The next step in selection is employment interview. Here
interview is a formal and in -depth conversation between applicant‘s adequacy . It is
considered to be an excellent selection device. Interviews can be One -to-One, Panel
Interview, or Sequential Interviews. Besides there can be Structured and Unstructured
interviews, Behavioral Interviews, Stress Interviews.
4. Reference & Background Checks: Reference checks and background checks are
conducted to verify the information provided by the candidates. Reference checks can be
through formal letters, telephone conversations.
5. Selection Decision: After obtaining all the information, the most critical step is the selection
decision is to be made. The final decision has to be made out of applicants who have passed
preliminary interviews, tests, final interviews and reference check s.
6. Physical Examination : After the selection decision is made, the candidate is required to
undergo a physical fitness test. A job offer is made after the physical examination.
7. Job Offer: The next step in selection process is job offer to those applicants who have cut
across all the previous obstacles . It is made by using the method of letter of appointment.
8. Final Selection
9. Induction Process : The final stage of the process is to ensure the new start embarks on a
thorough induction to the company and their role. Line Managers have responsibility for
ensuring the effective induction of all new appointees.
1.2.3. Te sts used in selection of human resources
A section test can be defined as any instrument used to make a decision about a potential
employee. There are a number methods organizations use to de termine if an applicant has the
potential to be successful on the job. Selection tests are used to identify applicant skills that cannot
be det ermined in an interview proc ess. Properly designed selection tests are stan dardized, reliable
and valid in predicting an applicant‘s success on the job. Within a selection interview, it is difficult,
if not impossible, to eliminate subjective judgments . The use of objective standardized tests helps to
compare fairly the candidate wit h the requirements of the post.
Most jobs require a range of knowledge or aptitudes which cannot be tested properly in the
interview situation alone. It is good practice to look for other sources of evidence to support the
claims made by candidates in their applications and interviews.

12
The main goal of testing is to obtain data about candidates that help predict their possible
success as employees. Some of the advantages of testing include finding the most fitted person for
the job, attaining a high degree of job satisfaction for the candidate , and decreasing turnover. Some
of the usually used tests can be categorized as follows:
• Intelligence tests are aimed to measure mental capacity and to test memory, speed of
thought, and the capacity to see relationship links in complex problem situations.
• Proficiency and aptitude tests are made to determine interests, existing skills, and
potential for obtaining new skills.
• Vocational tests are aimed to point out a candidate's most appropriate occupation or the
areas in which the candidate's interests match the interests of exist ing employees.
• Personality tests are aimed to disclose the applicant's individual characteristics and the
way the applicant may interact with others, thus giving a measure of leadership potential.
The premise is that if scores on a test correlate with job performance, then it is economically
useful for the employer to select employees b ased on scores from that test. There has been an
increase in the use of these techniques. Ability, aptitude and personality questionnaires are used
mainly for manag erial posts, while literacy and numeracy tests are more popular for cle rical and
secretarial positions (Beardwell and Holden, 2000). Tests such as these should really be seen as
complementary to interviews, rather than replacing them.
Tab 1.2. The differences between the advantages and disadvantages of testing employees
Selection tests
Advantages Disadvantages
• Can p roduce valid inferences for a
number of organizational outcomes
(like performance, success in training).
• Can predict job performance
particularly for more complex jobs.
• Can be administered via paper and
pencil or computerized methods easily
to large numbers.
• Can be cost effective to administer.
• Does not require skilled administrators.
• Can reduce business costs by
identifying individuals for hiring,
promotion or training who possess the
needed skills and abilities.
• Will not be influenced by test taker
attempts to manage or fake responses • Are typically more likely to differ in
results by gender and race than other
types of tests.
• Can be time -consuming to develop if
not purchased off -the-shelf.
• May lead to individuals responding in a
way to create a positive decision
outcome rather than how they really are
(i.e., they may try to positively manage
their impression or even fake their
response).
• May be disliked by test takers if
questions are intrusive or seen as
unrelated to the job .
As it is revealed, t ests have a number of limitations (Tab 1.2.) , howev er. First, expert
psychologists agree that tests are not accurat e enough to be used as the sole measure of applicants'
characteristics but must be taken into account each individual' s entire history. Second, the te st
operator must know what tests do and what their limitations are . One of the main limitations is
uncertainty about whe ther tests are really applicable . Third, bef ore any test is broadly used, it should
be tried out, if likely on existing personnel, to see whether it is effective for employees whose skills
are already known. Fourth, it is also significant that tests be managed and interpreted by specialists
in the field. Finally, tests must not disc riminate and should be consistent with laws and guidelines.

13

Web-based
technology
Electronic
databases
LAN/PC based
Legacy
mainframe
systems Chapter 2.Informational Systems in human resources management
HR technology can be defined as any technology that is used to attract, hire, recall , and
maintain human resources, help HR administration, and improve HRM. This knowhow can be used
in different types of human resource informat ion systems (HRIS) and by different participant s, such
as managers, employees, and HR professionals. This tool can be accessed in different ways.
Informational Systems overcomes tradit ional restraints in a cost -effective manner and opens
opportunities to ingest, store and process data from new sources such as external social media data
inputs, market data, communications, interaction with both potential employees and customers via
digital channels, etc. By some estimates, more than 80% of the data within organizations is
unstructured and unfit for traditional processing (Spencer, 2014).
2.1. HUMAN RESOURCES INFO RMATIONAL SYSTEM METHODS
Human resources informational system is a system which seeks to merge the activities
associated with human resource management (HRM) and information technology (IT) into one
common database through the use of enterprise resource planning (ERP) software. The goal of HRIS
is to merge the different parts of human resources, including payroll, labor productivity, and benefit
management into a less capital -intensive system tha n the mainframes used to manage activities.
2.1.1. Mainframe softwares
A. First generation HRIS
Initially , HR systems were operated individually and did not in tegrate with any other
business related occupations . Features were added as needed. Data was usually stored on mainframe
computers, the reporting was very basic , and HR was the only keeper of the data. There is no
disbelief that technology has made it easier and faster to collect, assemble, and deliver information
and communicate with employees. More importantly, it has the potential to decrease the
administrative problems in the HR department so it is more able to focus on additi onal significant
HR activities, such as providing managers with the specific expertise needed.
Fig 2. 1. Level of access increases from HR staff having access through to all managers and employees having
access (Source : Julie Bulmash, 2006)

14
Combined with a lower cost of storage per gigabyte, this informational system enables
organizations to build, for example, a federated view of employees databases by shifting employee‘s
data from various separate business departments into a single infrastr ucture, and then to run
consolidated analytics and reporting on it.
B. LAN/PC based systems
In the next phase , there was a movement of th e information locally presented in the paper –
based systems to PCs and local area network (LAN) systems. The HR database s were capable to
create reports t hat listed tombstone data (list of employee) , meaning basic employee information.
Developments in database technology involved payroll and some basic forms of employee tracking. .
C. Electronic Database Systems
The next phase arose with the development of re lational database technology. A relational
database signifies that a quantity of data can be stored in more than one file, each one holding
diverse types of data. The various files can be linked so that information from separate files can be
used together. A relational database permits databases to be recognized in seve ral not the same
locations and the information linked. This knowhow provided organizations with the capac ity to
develop more compl ex reports that combined several data elements.
2.1.2. Web -based techniques
A. Web Based Technology
Nowadays , many firms have started to utilize HR technology. The benefits of automation are
becoming commonly known to HR and other areas of the industry . The concentration has shifted to
automating as many transactions as possible to attain efficiency and effectivene ss. With the Internet
growing , managers began to understand that HR has to become an integral part of the org anization.
For the most part, t he HR department remains the owner of HR information but others have
begun to be aware of the value of this information to the organization . The reports that HR is able to
create have become more refined . Currently , the most common of systems are still not Web based,
but some leading edge organizations have incorporated this technology.
B. Social media
Kaplan and Haenlein (2010) define social med ia as a group of Internet based applications
that build on the ideologica l and technological foundations of Web 2.0 , and that allow the creation
and exchange of user generated content.
These web -based services allow individuals to :
1. construct a public or semi public profile within a constrained system
2. generat e a list of other users with whom they share a connection, and
3. view and pass through their list of connections and those made by others inside the system
Social media tools and social networking sites ha ve modernized communication methods,
both priva tely and gradually , at work. Communication through social media includes the use of an
online platform or website (a social networking site) that allows individuals to communicate, usually
for a social purpose, over a range of servic es, most of which are web based and give opportunities
for general public to interact over th e internet, e.g. via e -mail and instant messaging.

15
2.2. THE NEED FOR THE COMPUTERIZATION OF HUMAN RESOURCES
MANAGEMENT
Human Resource Information System (HRIS) is a systematic way of storing data and
information for each individual employee to aid planning, decision making, and submitting of
returns and reports to the external agencies. HRIS can be concisely defined as int egrated systems
used to gather, store and investigate information regarding an organization‘s human resources.
Human resource information system are generally needed for collecting , classifying, processing,
recording and disassembling the information requi red for efficient and effective management of
human resource in an y organization. Necessity for such a system ascends due to some factors like:
1) Organizations that hire a very bulky number of people, it becomes necessary to develop
employee database for managing personnel matters .
2) In a geographically spread corporation every office needs timely and accurate information
for workforce management. If this information is stored in multiple location s, cost and
imprecision will in crease.
3) Modern day compensation package is complex consisting of many payment s and deductions .
A central stored available data can become useful for in real time decisions.
4) Organizations have to conform to several laws of the area. A compu terized informati on
system can store and retrieve data rapidly and correctly , allowing the organization to
conform to legal requirements.
5) With the support of computerization staffs information system, employer record and file s can
be incorporated and saved for cross refere ncing and forecasting.
6) Required flexibility for adaptation to changes occurring in the environment can be built into
computerized information system.

HRIS is one of the key components of the HR system. As a result, the importance of HRIS is
increasing with the rapid growth of the corporate sectors. But still there are a lot of barriers to the
success of HRIS. The increasing usage of HRIS emerged from the need to have and control
automated processes or activities like:
 Recruitment and Selection
 Workforce career cycle management
 24/7 data access to authorized managers
 Customized levels of access confidential data
 Access to real-time data with immediate updates
 Staff administration
 Benefits administration and Training
 Performance and development and Safety or health
 Organization al management
 Payroll and Compliance
Informational Systems technologies release organizations from the traditional accuracy vs.
cost challenge by enabling them to store data at the lowest level of detail, keeping all data history
under reasonable costs and with less effort. Companies are using informational systems as a way to
learn about potential employ ees' personalities and behavior . Professional use of social networking
services refers to the employment of a network site to connect with other professionals within a
given field of interest .

16
There are databases and the human resources department has the type of information collected,
where it is stored, how the data is used, and the category of system used for these purposes has
changed over time, but the need to collect data relating to employing , promoting, and firing
employees has not changed.
Tab. 2.1 . HRIS benefits and limitations
Human Resources Informational Systems
BENEFITS LIMITATIONS
• Reduction in duplication of efforts
leading to reduced cost.
• Ease in classifying and reclassifying
data.
• Better analysis leading to more
effective decision making.
• Higher accuracy of information/report
generated.
• Fast response to answer queries.
• Improved quality of reports.
• Better work culture.
• Establishing of streamlined and
systematic procedure.
• More transparency in the system.
• Employee -Self Management • It can be expensive in terms of finance
and manpower.
• It can be threatening and inconvenient.
• Thorough understanding of what
constitutes quality information for the
user.
• Computer cannot substitute human
beings.
• Barriers in user acceptance

One of the major factors is the limited understanding of benefits and cost. It is still difficult
to measure the return of investment in HRIS. This study has been initiated to understand the
benefits, cost and barriers of HRIS in a theoretical context.
 Why use a HRIS for recruitment?
The recru itment activities can be enhanced through the use of career expansion programs.
When designing such a program, the organization should collect work history and skill level
information on every one of its employees. Such information would include age, education level,
training, special abilities (e.g., foreign language spoken), and promotion track record, and should be
stored on a web based system . This information will let the organization to ident ify existing
employees who are qualified to accept jobs with superior responsibility levels. For example , in
departments where skilled managers are in short supply, a management replacement chart can be
made, that lists current managers, proposes prospecti ve replacements.
 Why use a HRIS for selection ?
One area of the selection process that is significant and sometimes under -valued is the
development of screening tools. The ultimate goal of selection is to hire the candidate who is the
best possible match fo r the job duties and the culture of the company. This makes using the right
selection tools and developing them effectively vital. Careful job analysis helps HR professionals
better align selection tools with the job. Only tools, criteria and interview que stions that help yield
the best hire should make it into the selection process. And an informational system can analyze,
compare and choose a much more suitable candidate because of the lack of human subjectivity.

17
2.3. THE PROCESS OF RECRUIMENT AND SELECTION TROUGH INFORMATIONAL
SYSTEMS
To recruit and select an employee through an informational system means to identify a form
of technology that enables HR professionals to integrate an organization‘s human resource s
strategies and processes in order to improve overall HR service delivery. Since the mid -1990s
organizations have been embracing ways to inco rporate electronic and computer functi ons into their
HR strategies.
2.3.1 The use of informational softwares in r ecruitment and selection
Organizations are always considering for better ways to manage expenses , provide better
service, and effectively man age human capital, and electronic HR has become important to helping
organizations accomplish these goals, as showe d in Tab 2.2. One of the most efficacious innov ations
is the migration of HRIS applications on an intranet. An intranet is a network that is interrelated
inside one organization, using Web technologies for sharing information
Tab 2.2 . Example s of softwares used for recruiting and selection
Informational System used for recruiting and selection
Enterprise -Wide Systems
An enterprise wide system (called an enterprise resource planning or ERP system) is
observed as a system that supports enterprise wide or cross functional requirements, than
a single part or group inside the organization. These ERP systems have their basis in
software that incorporates information from diverse applications (modules) into on e
universal database. This signifies that economic information can be related to HR
information through one database. The most widespread high end enterprise wide
systems are SAP, PeopleSoft, and Oracle.
SAP
SAP was founded as Systemanalyse und Programmentwicklung in 1972 by former IBM
employees in Mannheim, Germany. This abbreviation was changed to Systeme,
Anwendungen und Produkte in de r Datenverarbeitung, which signifies ―systems,
applications, and products in data processing . SAP is the world‘s third biggest software
company. In terms of revenue, SAP is the principal business application and ERP
solutions provider. The company‘s leading product is SAP R/3, the ―R‖ stands for real –
time data processing and the digit ―3‖ relates to a three -tiered system —database,
application server, and client.
SAP produces are used by more than 12 million individuals in more than 120 states , and
its target market has usually been major multinational companies. Only just lately SAP
has targeted small to medium sized societies .
SAP is composed of individual, integrated software modules that carry out several
organization al system tasks such as finance , controlling, project system, capitals
management, materials management, and sales dispersal . One of its main modules is
Human Resource Management Systems (HRMS/HRIS), that are very refined and SAP
offers a full range of func tionality, HR products, and Web based assistances .
PeopleSoft Inc.
PeopleSoft is software that provides HRMS, engineering , financial, enterprise
performance management, and student administration software solutions to large
companies and governments. The company was founded in 1987 by David Duffied and

18
Informational System used for recruiting and selection
Ken Morris. Its software is made up of modules, such as HRMS, that includes payroll, all
human resources jobs and benefits, financials, industrial , student administration, and
customer relationship management. PeopleSoft is known for its capacity to be easily
custom -built to fit the specific business needs of each client. PeopleSoft was acquired by
Oracle in 200 5. It allows employees to add dependents to health insura nce, change
payroll deductions, join in benefits programs, estimate pension benefits, and perform
retirement planning.
Stand -Alone HRIS
HRIS can also be stand alone. Not all organizations need an intri cate system. Some need s
for organizations include cost, the total of employees, the grade of efficiency, and the
firm‘s existing hardware and software. An effective HRIS needs stability between what it
can do from a technical viewpoint and how it can meet the needs of that business . These
requirements usually grow with the size of the organization.
Smaller organizations can use very basic software applications, like Microsoft Excel and
Access. And can only require payroll and benefits administratio n, period and presence ,
and employee scheduling utilities . Midsize firms usually need compliance tracking and
reporting, health claims administration, payroll, compensation, and benefit
administration. Managers may need facts on performance assessment , tim e and
attendance, succession forecasting , abilities testing, and employee scheduling, and
employees may use the system to help in career expansion and self -serve applications.
Midsize firms also need better data integration and enhanced backup and recovery ability .
In addition, they have many users and need local area netwo rk and server based
procedures .
Specialty Software
With so many options in the open market , software can be acquired for virtually any HR
related occupation in any industry, including training and development, performance
management, progression planning, and the construction of organization al charts, that
tend to shape a new form of informational systems used for HR.

2.3.2 The use of social media in recrui tment and selection
Old-fashioned methods of recruiting usually involve advertisements, paper a pplications and
CVs, or face to face networking, counting phone ca lls. T hese tools mainly rely on the so called
―spray and pray‖ method where big numbers of advertisements or appl ications are sent out (sprayed )
while both e mployers and job seekers wait (pray ) for feedback (Joos, 2008) .
With the appearance of the internet and social media more exactly , new method s have
entered the range of employment and recruitment. N ew conducts of recruitment have not yet
replaced the more traditional methodologies but rather added to existin g HR processes (Joos, 2008).
Social-media tools and techniques presently used are not a straight replacement for
traditional hiring procedures , but ra ther an enhancement to them. Furthermore , research suggests
that a change from tr aditional to online recruitment might not b e as easy as just altering the tools, as
it might necessitate more strategic changes in the recruitment pro cess (Parry and Tyson, 2008) .

19
There are a widespread variety of benefits associated with using social media in recruitment,
focusing mainly on c ost savings, increasing the number of applicants reached, and being able to
target recruitment at specific groups of potential candidates (Davison, Maraist and Bing, 2011).
Tab 2. 3. Examples of sites used for social networking recruiting
Social networking websites and tools
• Blog – a conve rsational/ informational website published on the inte rnet that is made of of
entries (posts ) usually showed in reverse chronological order , the most recent post appears
first. Usually a blog is run by a single author, recently multi author blo gs have emerged ,
with posts written by large numbers of authors and skillfully edited. Blogs contain regular
entries of commentary and portrayals of events, this is known as blogging .
• Facebook – a social ne tworking site where users create individual profiles, add other users
as friends and exchange messages, as well as receiving automatic notifications when
profiles are updating . In addition, users may join common interest groups, organized by
common characteristics (e.g. workplace).
• Google – an American multinat ional organization specialized in Internet related services;
most known its core search engine, but also a has large series of products including
Google+, the second biggest social networking site (as of January 2013).
• LinkedIn – a business related social networking website, mostly used for professional
networking. Users keep up a list of contact details of people with whom they have some
level of relationship, named connections, that can be used to shape up a contact list network,
follow diverse companies and find jobs, personnel, and business prospects .
• Podcast – a type of digital broadcasting containing a series of audio, video, PDF, or
electronic files subscribed to, downloaded or streamed online to an electronic device.
• Twitter – a common micro blogging social networking site that enables its users to send
and read openly visible messages named tweets. Tweets are text based posts of maximum
140 characters shown on the user‘s profile page. Users can subscribe to other users profiles .
• YouTub e – a (Google owned) video sharing network in which users may upload, share, and
view videos. An extensive variety of user generated video content is shown , including film
and music videos, as well as amateur content such as media blogging. Media companies
also offer some of their material on here.
• Wikipedia – a cooperative web based encyclopedia venture ; its 18 million articles and
almost all are easily editable by any user. A noticeable web site but not an example of social
networking site as such .
• Yahoo – an American international internet service corporation extensively known for its
social networking services and user generated content counting online discussion panels
(Yahoo! Groups), community Q&A site s (Yahoo! Answers) and photo sharing (Flickr).

The internet behaves itself good on finding and attrac ting university graduates, skilled
employees , managers, and executives, as table 2.3. suggest. These individuals have a tendency to be
computer literate, and technology usage is an essential part of their day -to-day customs, developing
and sustain ing relations at work as well as in their private lives.
 The b enefits of using social media in recruit ing
Social media networks can be used to identify possible applicants with definite
characteristics or qualifications and to reach nont raditional applicants, those who are not in the
typical target groups, or who have diverse backgrounds and e xperience, rapidly and easily:

20
 Cost-saving and targeting: online recruitment offers a n economical alternative to other
traditional means of recruiting.
 Fostering accurate expectations: Interactive instrument s such as Facebook, LinkedIn and Twitter
can show an important role in the overall process of recruitment, providing recruitment
information and adopting realistic jo b expectations among possible employees.
 Improving external communication: Social media can aid to increase communication in general ,
which is good for all organizations .
 Building the brand : It also exists a brand building element when using social networking
websites that is hard to quantify, but more and more important for societies .

 The risks and costs of using social media to recruit
While social media undoubtedly offers certain advantages to employers in terms of making
saving and targeting recruitmen t, using social media to hire also carries certain costs and risks . The
main risks of recruitment and selection using social media and informational system:
 the accuracy of online information used in HR decisions; invasion of applicant privacy;
 variability in form and amount of information available across an applicant group ;
 lack of openly identifiable theoretical concepts used in the selection process;
 the absence of data to sustain that the information used in selection is job related ;
2.3.3. Recruiting and selection stag es using HRIS
Developing a plan to make a campaign to recruit and select includes establishing a process
that should cover steps like gathering i nformation from the recruitment, training and growth, and
administrative substructures, such as number of open positions, types of positions, employee skills
and capabilities, job rates (salaries), retirement suitabili ty.
Fig 2.2. Guideline steps for social media hiring
Select the Right
HRIS for the Job
Create a Real Social
Media Presence
Include Employees
in Recruitment
Efforts
Use Social Media as
another Screening
Tool
Select the Social
Media Platforms
That Work for You
Start an online
recruiting campaign
Select the most
fitted candidate
using HRIS tools
Perform the hiring

21
As a summarization of all theor etical methodology, it can be stated that HRIS can help to
streamline recruiting by making it easier for employees to fill out applications from the comfort of
home – or wherever they may roam with their mobile phone if your HRIS includes mobile
accessibil ity. However, if a company simply tries to attract candidates using old school techniques
you may be missing the mark even with active HRIS recruitment functions in play. Taking the time
to link up HRIS to popular social media sites may help you to get the most out of your recruitment
processes.
 Select the Right HRIS for the Job
First and foremost, an organization will have to find an HRIS that offers modern recruiting
and applicant tracking options. If employees can‘t fill out an application online, the organization
going to have a much more difficult time attracting top talent and getting them through the door.
The best HRIS solutions should make it exceedingly simple to post job ads through social media and
automatically funnel interested employees into the proper place to apply when they click.
 Create a Real Social Media Presence
When companies create social media profiles just to try to create an online presence, it is
painfully obvious. Even if an organization‘s sole purpose of creating company so cial media
accounts is to use them for recruiting, maintaining the pages periodically and doing something to
make the pages unique and fun fosters a better company image and is more likely to attract smart
and savvy candidates. It is best to switch up the posts, as well, as a barrage of job postings can look
desperate.
 Include Your Employees in Recruitment Efforts
Employees have always been a great source for job candidates and that is not likely to
change. When social media is included, however, it opens u p whole new realms of possibility for
recruiting through employees and including employees in the search. Employees can share job
openings postings with their friends, message employers to check out certain applications, and help
in a myriad of other ways.
 Use Social Media as another Screening Tool
Social media gives a concise snapshot of people, so it can be very helpful when used as a
screening tool. While most social platforms have the option to make profiles private, not eve ryone
uses these options. If the company can find out more about a candidate to help you decide if that
person will be a good fit, it will only help improve the recruitment processes and save you time
weeding out people that it otherwise may have taken three interviews to filter out.
 Select the Social Media Platforms That Work for You
Since it takes time and effort to start and maintain social media pages, it doesn‘t make sense
to get on every available platform and start a page. Chances are good that just two or three social
media sit es really fit a company culture and work with current recruiting strategy. Instead of making
a costly researching activity , current employees can suggest what sites they a re on to perpetuate a
company‘s culture.

22
Chapter 3. Deloitte’s informational syste ms used for recruiting and selection
case study
3.1. DESCRIPTION OF THE COMPANY
Deloitte provides audit, consulting, financial advisory, risk management, tax and related
services to public and private clients spanning multiple industries. Present in more than 150
countries and territories, Deloitte brings world -class capabilities and high-quality service to clients,
delivering the insights they need to address their most complex business challenges.
Deloitte is part of the Big Four international group, which brings together the four largest
consulting firms, together with KPMG, Ernst & Young and PricewaterhouseCoopers. Deloitte‘s
more than 210,000 professionals are committed to becoming the standard of excellence.
 Deloitte Romania
Deloitte Romania is composed by Deloitte Audit S.R.L., Deloitte Tax S.R.L., Deloitte Consultancy
S.R.L., Deloitte Evaluation S.R.L. Deloitte Fiscal Representative S.R.L. and Reff & Associates,
which are affiliated with Deloitte Central Europe Holdings Limited.
 Headquarters in : Bucharest, Timisoara, Cluj.
 Number of employees in 2016: 614
 Turnover in 2016: 35 million euro (cumulated)

 Vision and Mission of Deloitte
Deloitte‘s vision is unchanging: they aspire to be the Standard of Excellence, the first choice of
the most sought -after clients and talent. Deloitte‘s mission is to make an impact that matters. Their
shared values succinctly describe the core principles that distinguish the Deloitte culture.
1. Integrity – Deloitte believes that nothing is more important than their reputation, and behaving
with the highest levels of integrity is fundamental to what they are, demonstrating a strong
commitment to sustainable, responsible business practices.
2. Outstanding value to markets & clients – Deloitte plays a critical role in helping both the capital
markets and our member firm cli ents operate more effectively with constant vigilance and
unrelenting commitment.
3. Commitment to each other – The firm‘s culture of borderless collegiality is a competitive
advantage , and they go t o great lengths to nurture and preserve it and to support th e people.
4. Strength from cultural diversity – Their business challenges are complex and benefit from
multidimensional thinking, working with people of different backgrounds, cultures, and thinking
styles helps the people grow into better professionals and le aders.

 Objectives and strategy
Deloitte‘s objective is to achieve profitable growth and strategically manage costs while
intelligently navigating current and future opportunities and risks. Increased globalization means
there are new markets for goods and services and pressure on emerging market companies to
compete on innovation as well as price.
The company is responding to these changes and has a plan in place that positions them for
future success and offers their people and their recruits more opport unities than ever before. Deloitte
calls this plan Vision 2020, and its objective is to make them the leading global professional services
organization by 2020.

23
By 2020 Deloitte has developed the following strategic objectives, built on three pillars:
1. Focu s on maintaining their position in the market by delivering exceptional client service
and maximizing opportunities in markets and services
2. Create the highest performing teams by attracting, developing and inspiring the best
people and committing to a cult ure of world -class teaming
3. Strengthen global, empower local by pressing our global advantage and empowering
local teams

 Target market: Deloitte‘s target market segments include private companies, mid -sized private
equity firms, next -gen companies, and mid -cap multinational companies.
 Services: All of Deloitte Romania entities provide services in five areas: audit, tax services,
consulting, risk assessment services and financial consulting.
 Market positioning: Currently, Deloitte is the number one leader in its industry field
worldwide, as well, according to wallstreet.ro, Deloitte Romania has the largest market share in
their service sector – 35%, compared with their competitors.
 Competitors: Deloitte's thing is that they are able to provide services is jus t about every
industry (they are largest professional services firm in the world) so a full list would be quite
extensive, but their major competitors are:

Tab 3.1. Deloitte‘s competitors according to specific industry
Strategy Technology Human Capital Audit
McKinsey
Bain & Co
BCG
Accenture
PwC IBM
Cap-Gemini
TCS
HP Enterprise
Services Towers Watson
Mercer
Aon Hewitt
Hay Group Ernst & Young
KPMG
Grant Thornton LLP

3.2. DELOITTE’S RECRUITING AND SELECTION STRATEGY
Deloitte Romania, an auditing and consulting company with approximately 600 local employees
in Bucharest, Timisoara and Cluj, has designed a growth strategy by 2020 that involves increasing
the number of current employees to more than 800. The company's plans include, among ot her
things, the overrun of a threshold of 800 employees and strong investments in developing employee
leadership skills.
 Recruiting strategy for 2020:
The company's recruitment strategy now targets both the attraction of young people and
candidates with more than 3 -4 years of experience on the labor market. Until now, they've been
employing almost exclusively young graduates, but for 3 -4 years they've started recruiting almost
young people and experienced people(Carmen Baibarac,2016).
 Target employee:
1. Young Graduates
2. Junior -Mid Level Experienced Candidates
Increasing customer demands and developing new business areas, such as accessing European
funds, are two of the reasons why the audit and consulting firm has chosen to recruit a similar
number of fresh graduates and people with experience.

24
 Employee profile:
More than 70% of Deloitte Romania employees are women. Also, five of the company's 15
partners are women. The average age of the company's employees is about 30 years.

Fig 3 .1. Division of Deloitte‘s employees according to gender
At the same time, Deloitte Romania has an annual staff turnover of 15%. The value of staff
fluctuation at the firm level is normal (Allen, 2012). Most depart from the audit area. A junior can
evolve int o a partner position in ten years.
 Division of employees:
About 500 of the company's employees work in Bucharest, and the rest in Timisoara and Cluj.
Also, 200 of Deloitte Romania employees work in the audit and tax departments, and the others
cover the co nsulting area and support functions. The junior candidate profile in the tax advisory area
is the same as the junior candidate profile in the audit.

Fig 3. .2. Division of Deloitte‘s employees in the regional offices
 Employee origin :
In the area of young candidates, Deloitte Romania recruits mainly graduates from the Academy
of Economic Studies (ASE) in Bucharest . About 80% of the juniors they employ are ASE graduates
(Pestler,2016). But we are also recruiting graduates of different economic engineerin g 29%
71% Gender division of employees
Women
Men
82% 12% 6% Regional division of employees
Bucharest
Timisoara
Cluj

25
specializations from the Polytechnic or graduates of the National School of Political and
Administrative Studies.

Fig 3 .3. University of origin of Deloitte‘s employees
 Challenges

One of the biggest challenges that Deloitte has to overcome in recruiting young people is the
decrease in the number of ASE graduates in recent years, the number of economics graduates has
decreased over the past three years by about 1,000 people (INS,2016 ). In addition to lowering the
number of graduates, their level of training is also a challenge. Above all, competition has also
increased for Deloitte. Until a few years ago, most of the ASE graduates entered the Big 4 or in
banks, there are now many empl oyers in the Shared Service Center area who absorb these graduates.
In the segment of experienced candidates, the company is experiencing recruitment problems for
certain specialist positions due to the limited number of market experts. When is found the specialist
the company needs, they have to c onvince him to come at Deloitte . And here comes another
challenge, because some experienced people have great difficulty in accommodating to a new job
because of the cultural differences that exist between Deloit te‘s organization and others.
 Goal:

Deloitte Romania will recruit between 90 and 110 people in the financial year 2017/2018, both
for new jobs and for replacing employees who choose to leave the company. About 75 of the new
employees will be university or master graduates (Tab 3.5) that will fill in the new created
vacancies.
Tab 3.2 . Deloitte hiring goal
Year Number of employees
New positions Fill-in positions
2017 -2018 75 25
2018 -2019 75 25
2020 Σ =200
77% 12% 6% 5% Deloitte Employees by Univerisity
of Origin
ASE
UPB
SNSPA
Other

26
 Campaigns:
The company runs all recruitment activities for young people under the generic title of "Talent
feel at home @Deloitte", with the following components:
 Tax Graduate Recruitment campaign
 Audit Graduate Recruitment campaign
 Deloitte Student Academy (DSA), an employer branding vehicle introduced by the company
in 2014. The DSA addresses students or students in the last year of school attending the
various workshops run by Deloitte Romania managers. At the same time, the young people
selected in the program can also see what a working da y actually looks like in the audit and
consulting company.
 Deloitte Internship Program In the financial year 2016/2017, the internship program began in
the fall of 2016 and will end in May 2017.
The internship program has such a long duration for several reasons. One of the reasons is that
the company wants to give young people a real chance to understand what Deloitte is doing.
Another is that the beginning of the calendar year corresponds to the peak of the audit season and
then it is the time when they really need extra resources. Altogether, they take about 30 interns
annually, 20 of which enter the audit department alone, approximately 90% of those completing the
internship program receive a job offer.
 Sources for recruiting:
The company uses various r ecruiting tools ranging from job advertisements to student campus
events. On the younger candidates segment, the company's most successful channel for recruiting is
the online career page.
 Financial benefits:
A young man who gets out of school is paid over 3,000 RON per month (more than 2,100 RON)
for an entry level job in an audit and consulting company. Deloitte is very attentive to the way
wages are on the market, because they have to make sure that they are competitive also from this
point of view.
All employees of Deloitte Romania receive a monthly fixed salary and an annual bonus, which
is granted according to certain performance indicators. The company increases wages on average by
about 10% annually. The salaries their employees receive firstly incre ase as a result of the natural
promotions that take place in the company. About 90% of Deloitte‘s people promote each other on a
different professional level. And any promotion brings increase in the salary.
 Hiring budget:
The company has an annual traini ng budget of about 1 million euros. It may seem like a lot, but
in order to grow professionally Deloitte invest s heavily, both on the professional and th e soft skills.
Almost all of Deloitte Romania's employees have or are in the process of obtaining an AC CA
(Advanced Chartered Certified Account), ADIT (Advanced Diploma in International Taxation) or
CIMA (Chartered Institute of Management Accounting) accreditation. Just for the juniors who
entered last year in the audit department, ACCA credentials cost 50, 000 euros.
Also, all employees of the company that occupy "manager -up" positions undergo specialized
training alongside their peers from similar positions in other countries. Training for executives takes
place in Brussels (Belgium) or Paris (France). Such a course costs at least 3,000 euros for each
person, with no accommodation and transportation costs.

27

3.3. THE PROCESS OF RECRUITING AND SELECTION THROUGH
INFORMATIONAL SYSTEMS AT DELOITTE
Main goal of the study:
The purpose of the research is to illustrat e the informational systems used in the processes of
recruiting and selection of the employees and to offer a clear, concessive image of the informational
tools that Deloitte is using in their hiring process and the means in which this process has become
digitalized and how the HRIS is used to attract, hire and evaluate new employees.
To accomplish this research it was used the process of recruiting and selection, which had
four main objectives. Those objectives were:
O.1 Recruiting and sourcing for the most fitted applicants through social media and web -based tools
O.2 Selecting the best candidates for the job using a data mining tool
O.3 Help to fill the roles at Deloitte and reach the target of 800 hires till 2020
O.4 Evaluate the process of hiring through informationa l systems used for recruiting and selection
Duration of the research: 14 September 2016 – 15 May 2017
Process stages:

Fig 3.4 . The hiring process using informational systems
This process has three key phases, including planning, recruitment, and employee selection.
In order to see how this process works it has to be divided into specific activities and task. This
information will help the company in taking the proper hiring development decisions to increase
their competitiveness in perspective of future hires.

28
As illustrated in table 3.3.1. , is in the best purpose of this research that the hiring process
should be divided in two distinct independent processes: the process of recruitment with its specific
tools and the selection process, also il lustrating the specific informational systems used.
Tab 3.3 . Description of the hiring process activities
Deloitte’s Hiring Process
Objective Activity Description

O1 A1.1 Defining the recruitment A1.1.1. Identify vacancy
A1.1.2.Prepare job description and specification
A1.2 Planning the recruitment A1.2.1. Defining the process
A1.2.2. Setting time -frames
A1.3 Defining the target group A1.3.1. Sourcing the candidate profile
A1.3.2. Advertise
O2 B2.1 Defining the selection
B2.1.1. Identify proper selection tools
B2.1.2.Prepare process
B2.2 Identify/collect candidate
resumes B2.2.1. Receive resumes
B2.2.2. Shortlist resumes
B2.2.3. Send shortlisted resumes to department
manager
B2.3 Establish interview B2.3.1. Agree to interview date
B2.3.2. Call candidate for interview
B2.3.3. Interview 1
B2.3.4 Interviewers complete feedback form
B2.3.5. Send feedback form to HR
B2.4 Testing candidates
B2.4.1. Plan for testing candidates
B2.4.2. Schedule test
B2.4.3.Test
B2.4.4. Interviewers complete feedback form after
testing
B2.4.5. Send feedback form to HR
B2.5 Select final candidates
B2.5.1. Identify/collect resumes after interviews
(batch 2)
B2.5.2. Receive resumes
B2.5.3. Shortlist resumes
B2.5.4. Identify /collect resumes after testing
(batch 3)
B2.5.5. Shortlist resumes
B2.5.6. Select candidate
O3 C3.1 Finalize the hiring process C3.1.1. Email HR regarding selected candidate
C3.1.2. Send offer letter
C3.1.3. Sign contract
O4 D4 Evaluation of the hiring
process D4.1.1. Assessing the hiring process
D4.1.2. Asses the recruiting process
D4.1.3. Asses the selection process
D4.1.4. Final closing of the hiring process

29
3.2.1. Recruiting process
Recruiting process stages:
 Defining the recruitment
1. Identify Problem
The labor market in Romania is quite a difficult one for employers with approximately only 10%
of the workforce looking for a new job, this includes people both with and without jobs (INS, 2016).
This makes recruitment a priority challenge for the business. Consulting company Deloitte Romania
has around 600 employees and a hiring target for 2017 of another 75 juniors.

 Proposed strategy for Deloitte
For Deloitte, recruitment is shifting more towards internet marketing, content management and
measuring traffic to the career site. Goals:
• Help to fill the roles at Deloitte and reach the target of 75 new hires in 2017
• Develop a strong employe r brand which attracts the "right" candidates for Deloitte
• Drive traffic to their career website to create a funnel of candidates and supports the
employer brand

2. Prepare Job Description and Specification
For the purpose of the study, the attention will f ocus on Deloitte‘s current hiring campaigns
for junior tax and audit officers. The campaig ns are running under the name :
1. Tax graduate recruitment @Deloitte
2. Audit graduate recruitment @Deloitte
The reason of choosing those two campaigns is that their main focus is on undergraduate or
recent graduate students with few or no work experience at all, this will allow of finding a large
amount of possible candidates that could fit the company hiring target. With this in mind, proper job
vacancies add is created t o illustrate the specific requests for this campaign (Appendix 3).
 Planning the recruitment process
3. Defining the process
The first step in any hiring process is to determine the need for a new or replacement
position in the company. The hiring cycle was br oken down into specific tasks for every activity of
the process, as it follows:
Tab 3.4 . The hiring process task duration
Deloitte’s Hiring Process
Objective Activity Description Responsible Duration
(min)

O1 A1.1 Defining
the recruitment A1.1.1. Identify problem HRM 0
A1.1.2.Prepare job description
and specification Recruiter 960
A1.2 Planning
the recruitment A1.2.1. Defining the process HRM 480
A1.2.2. Setting time -frames HRM 480
A1.3 Defining A1.3.1. Sourcing the candidate Recruiting manager 960

30
Deloitte’s Hiring Process
Objective Activity Description Responsible Duration
(min)
the target
group profile
A1.3.2. Advertise Recruiter 23040
O2 B2.1 Defining
the selection
B2.1.1. Identify proper
selection tools Staffing manager 480
B2.1.2.Prepare process Staffing manager 120
B2.2
Identify/collect
candidate
resumes B2.2.1. Receive resumes Staffing specialist 1
B2.2.2. Shortlist resumes Staffing specialist 1440
B2.2.3. Send shortlisted
resumes to department manager Staffing specialist 1
B2.3 Establish
interview B2.3.1. Agree to interview date Staffing manager 60
B2.3.2. Call candidate for
interview Recruiter 4500
B2.3.3. Interview 1 Staffing specialist 15000
B2.3.4 Interviewers complete
feedback form Candidate 1500
B2.3.5. Send feedback form to
HR Candidate 1
B2.4 Testing
candidates
B2.4.1. Plan for testing
candidates Staffing manager 960
B2.4.2. Schedule test Staffing manager 120
B2.4.3.Test Candidate 18000
B2.4.4. Interviewers complete
feedback form after testing Candidate 1500
B2.4.5. Send feedback form to
HR Candidate 1
B2.5 Select
final
candidates
B2.5.1. Identify/collect resumes
after interviews(batch 2) Staffing officer 1
B2.5.2. Receive resumes Staffing specialist 1
B2.5.3. Shortlist resumes Staffing specialist 1440
B2.5.4. Identify/collect resumes
after testing (batch 3) Staffing specialist 1
B2.5.5. Shortlist resumes Staffing specialist 1440
B2.5.6. Select candidate Staffing manager 1440
O3 C3.1 Finalize
the hiring
process C3.1.1. Email HR regarding
selected candidate Staffing manager 60
C3.1.2. Send offer letter HRM 960
C3.1.3. Sign contract Candidate 5400

31
Deloitte’s Hiring Process
Objective Activity Description Responsible Duration
(min)
O4 D4 Evaluation
of the hiring
process D4.1.1. Assessing the hiring
process HRM 1440
D4.1.2. Asses the recruiting
process Recruiting manager 960
D4.1.3. Asses the selection
process Staffing manager 960
D4.1.4. Final closing of the
hiring process HRM 1
Total amount of time 93.708

4. Setting time frames
In days, the time frame of the hiring process takes place in approximate 175 working days
(8h/day). The revised process includes some waiting time when it comes to directly interacting with
candidates, whose availability is an uncontrolled variable.
The number of people involved in this process is six, each with their specific tasks to per form:
 HRM (Human resources manager)
 Recruiting manager
 Staffing manager
 Staffing specialist
 Staffing officer
 Recruiter

 Defining the target group
5. Defining the candidate profile sourcing
The candidate profile is defined by a young professional with close to no work experience,
economics graduate or undergraduate, adaptable and with a relevant Bachelor‘s degree or
experience commensurate with job expectations for this position.
Tab 3.5 . Candidate profile features
Key Performance Indicator (KPI) Interval
Age 18-27 years
Studies Enrolled -Graduate
Work experience 0-2 years
English level A1-C2
Region Bucharest – Cluj – Timisoara – Other
Test results 0-100

The features described in Tab 3.v will be used do design the candidate profile whilst running
the hiring campaign on social media. To target the right employee is important that the KPI‘s should
be clearly defined in order to reach the intended audience.

32
6. Advertise
A large percentage of candidates for entry level jobs are seeking their first ―career jo b.‖
Many are young, either fresh out of school, looking to start a new career. Newspaper ads are not an
effective staffing source for most of today‘s applicants. Enlisting vacancy announcements online is
more effective and profitable than most traditional forms of advertising.

Tab 3.6 . Recruitment Strategy Planning for Deloitte
Strategy Gap Tactic
Career
website Not reaching at
employees that are
focused on the
company‘s brand  www.Deloitte.com has a designated Career page with all
job listings
 www2.workingatdeloitte. com is an website independent
to Deloitte‘s corporate website.
 Deloitte categorized the website in 3 categories:
o Students
o Young professional and
o Professional.
 They address different questions and provide different
content depending on the user.
 The design is quite simple and on all pages, they also
have testimonials from current employees.
 Recruitment staff profiles on the social media sites to
help them create communities which will communicate
with potential candidates.
 In the vaca ncy section Deloitte uses geolocation on a
Google map to target candidates in their specific
locations: Bucharest, Cluj and Timisoara..
Online
Recruiting Not reaching
majority of
applicants,
especially young
college grads  Social media is a big part of yo ung professionals‘ life,
this is the main reason for which online recruiting has
become the main recruiting method these days.
o Twitter is used to post jobs and events and alert people to
new blog posts. The strength of their Twitter presence
lies in the fact that all recruiters have a Twitter account…
And tweet as well.
o Facebook focuses on potential candidates. The company
is making contact with passive candidates this way by
sharing knowledge, building a relationship and keeping
them informed.
o LinkedIn is used to source more senior candidates. The
recruitment team can search for professionals by looking
at their network and recommendations. It also posts
vacancies on LinkedIn.
Campus
Recruiting
and Job Fairs Not enough
applicants with
specialized
econo mical work
degrees  Collaborate with university deans and professors to help
generate student interest in the field by creating
university workshops.
 Active partner of ASE Job & Internship Fair since 2007
 Develop stipend program partially covering college
tuition and other expenses of employee who agree to
further specialize in audit: AuditXcellence at Frankfurt

33
Strategy Gap Tactic
School of Finance & Management
 Participation at job fairs like Top Employers 2017 and
BuzzCamp – an event dedicated to students and young
profession als
Target
Recruitment –
job sites Hiring process
takes too long –
high quality
applicants are
looking elsewhere
for jobs  Placing vacancy announcements online is more effective
and economical than using most traditional forms of
advertising, thus Deloitte Romania collaborates with the
top 5 recruiting job sites:
o Ejobs.ro
o Bestjobs.ro
o Hipo.ro
Employees as
Recruiters Need to improve
overall applicant
pool  Issue periodic reminders to staff that vacancies exist and
their referrals are appreciated.
 Offer ―recruitment bonuses‖ to staff that refer applicants
who are eventually hired.
 Tie the bonus to the successful completion of the
probationary period.

After the completion of the recruitment process through the social media campaign, a total of
862 CVs was obtained. After contacting the candidates, 694 of them expressed their willingness to
continue the recruitment process further and be tested, according to their origin of application, those
candidates came from:
Tab 3.7. Resumes number according to adv ertising source
No. Advertising source Amount
1 deloitte.com 71
2 workingatdeloitte.com 37
3 Twitter 2
4 Facebook 174
5 LinkedIn 68
6 Campus recruitment 41
7 Job fairs 73
8 Job sites 196
9 Employees recommendation 32
Total 694

According to Tab 3.3.5, it can be seen that the most effective ways to attract candidates are
through the social media campaign on Facebook and specialized recruiting websites. At the same
time, the figures suggest that a Twitter campaign is practically us eless in attracting potential
employees in Romania, and Deloitte should avoid to use this source of recruitment in the future.

34
3.2.2. Selection process
Research methodology stages:
 Defining the selection
1. Identify proper selection tools
Hypothesis: cand idates with similar skills set and capabilities will perform similarly. All
information about the future employees is collected from data based software that use similar
technology and programming language and work on similar platforms. In consequence cand idates
under comparison here have similar capabilities and skill sets.
The development of this study was built after the Cross Industry Standard Process for Data
Mining (CRISP -DM) model. The CRISP -DM model was the most suitable for the purpose of the
study because it provides a standard guide to develop Data Mining selection development.
2. Prepare process
Data Collection : Data was collected by using various techniques such as form filling, testing,
interviewing and tracking down performance indicators about c andidates from different sources
such as: their resumes, their test results or their personal data from the Human Resources
Department.
Data Understanding: This is the first step in developing the problem. During this step the selection
process is formulat ed as a data mining problem, using Weka as the software for data mining, and
developing an algorithm to solve the problem, in order to expand the problem into quantifiable
variables.
The predictive attributes were extracted from the candidate‘s data collection form and were used
as performance indicators. They include:
1. AR – Age Range: Age attribute has the biggest impact on the future candidate‘s
performance. Due to the fact that the recruitment campaign is focused on young
professionals, the candidat es were labeled based on their age, the candidates had between 18
and 28 years old.
2. RAS – Reasoning and Analytical Skills: the values were collected through the test
assessment which involved analytical and logical -reasoning questions, the score was given
between 0 and 100.
3. SDQ – Specific Domain Qualification: This is the grade scored by the candidate during
his/her graduation (under -graduation) courses. It is categorized and normalized based on
their university grades from 6 to 10.
4. TE – Time efficiency: ma rked as a simple "Good" or "Bad" option based on the full
completion of the test, it shows if the candidate is considered efficient in completing his/her
work within the expected time -frame and it is represented in percentages of the test
completion in the given time, from 0 to 100.
5. SA –Score Average: the candidate‘s skills were tested using both simple and elaborate math
problems with multiple test cases. The entire score was granted for each problem only if all
the problems were performed well. The candi dates were labeled between 0 and 100.
6. CS – Communication Skills: This score for each of the employees was obtained from the
interview, grading similar to SA.
7. PE – Previous Experience: Obtained from the candidate‘s resumes and labeled with ―No‖ if
the candi date hadn‘t any previous work experience, ―Entry‖ if the candidate had maximum 2

35
years of experience and ―Experienced‖ if the candidates had more than 2 years in the field of
work, also there were some candidates that exceeded that value.
8. FLA – Foreign Lan guage Assessment: the candidate‘s English skills were tested. The
evaluation is similar to SA .
9. P – Performance Parameter: represents the dependent variable for the problem statement
which must be correlated to the independent attributes available from th e preliminary study
of relevant features. This was a numerical value between 0 -100 that was transformed into a
nominal one, based on performance intervals. The value was acquired by assessing the
quality of the overall result from 3 to 8, with :
o 3= Doesn‘t meet minimum standard
o 4= Needs improvement
o 5= Average
o 6= Meet Expectation
o 7= Exceed Expectation
o 8= Outstanding
These parameters mainly represent the skills set required to be possessed by a candidate.
Data Cleaning: The raw data had instances that were not valid due to few errors and anomalies that
had to be thrown away. First, the data was transferred to Excel sheets, then reviewed and modified.
Data cleaning and filling the missing values was performed before the se lection was applied to the
dataset to identify the key attributes obtained for the study‘s case. Afterwards, the files were
converted to .arff format for further using it in WEKA Data Mining toolkit.
Model Building: During this step, the selection was perf ormed, calibrated and applied the most
adequate data mining algorithms in order to obtain the models.
Tab 3.8 . Classification Algorithms Accuracy Rate
Number Technique Accuracy
1 J48 97.5216 %
2 JRip 93.3429 %
3 KMeans 90.2368%

Based on the study led , the data obtained was combined and summarized in a tabular form.
The algorithm used is JRip, executed using WEKA and R under "Test options". 10 – Fold Cross
validation was applied to increase the calibration mechanism of J48 decision tree.
Decision Tree
A decision tree is a decision support instrument that uses a tree -like diagram or model of
choices and their possible results, including chance event results, expenses, and utility. It is one
method to display a n algorithm.
Decision trees are regularly utilized as a part of operations research, as in the case of
selecting employees and displaying the existing patterns between them , Figure 3.x below shows the
tree generated using the J48 (also known as C4.5) algorithm. The overall res ults of the model are
presents in Appendix 6. As specified, score average and age attribute were the most effective
factors, with score average being the central node of the decision rule tree.

36

Fig 3.5 . Candidate selection decision tree
Based on the tre e description, after the score average being considered the most important
attribute (this is illustrated by SA being the tree nod), the following attributes had the foremost
important middling in the tree construction and showed that older candidates (mor e than 24 years
old) will have a better work performance due to their previous work experience and overall
knowledge . As in the case of the decision tree, the rule analysis is performed in order to see how the
attributes are associated. The result will ai d the company to target the candidates more specifically.
Classification Rules
The table below exemplifies the effects of attributes on performance. The score average had
the maximum gain percentage, which made it the opening node and most effective attribute
(Appendix 5) , just as the Decision tree.
Tab 3.9 . Classification Rules Generated By JRip Algorithm for Forecasting Performance
No Rule Antecedent Performance
Decision Recurrence
1 (SA >= 93) and (AR <= 25) => P=8 Outstanding 30
2 (SA > = 85) and (TE <= 82) => P=8 Outstanding 21/5
3 (SA >= 80) and (TE <= 74) and (TE >= 73) and (CS
<= 90) => P=8 Outstanding 6
4 (SA >= 68) and (CS >= 96) => P=7 Exceed Expectation 14
5 (SA >= 80) and (RAS <= 61 => P=7 Exceed Expectation 14
6 (SA > = 68) and (AR <= 20) => P=7 Exceed Expectation 43/15
7 (SA >= 74) and (TE <= 77) and (SA >= 79) and Exceed Expectation 9

37
No Rule Antecedent Performance
Decision Recurrence
(SDQ >= 8.44) => P=7
8 (SA >= 81) and (TE <= 82) => P=7 Exceed Expectation 7
9 (SA >= 74) and (AR <= 21) and (SA <= 75) =>P=7 Exceed Expectation 6
10 (SA >= 53) and (SA <= 59) and (AR >= 19) => P=5 Average 29/2
11 (SA >= 57) and (AR <= 18) => P=5 Average 19
12 (SA >= 63) and (TE >= 93) and (AR <= 26) and
(TE <= 96) => P=5 Average 8
13 (SA <= 74) and (SA >= 71) and (AR >= 22) => P=5 Average 6
14 (SA >= 64) and (AR <= 21) and (SA >= 70) => P=6 Meet Expectation 28
15 (SA >= 78) => P=6 Meet Expectation 68/27
16 (SA >= 61) and (AR <= 20) and (SA >= 64) => P=6 Meet Expectation 12
17 (SA >= 60) and (AR <= 19) => P=6 Meet Expectation 15/2
18 (SA >= 69) and (AR <= 22) and (SA >= 75) => P=6 Meet Expectation 9
19 (SA >= 43) and (AR <= 19) => P=4 Needs improvement 91
20 (SA >= 61) and (SA >= 64) and (RAS <= 75) =>
P=4 Needs improvement 38/11
21 => P=3 Doesn‘t meet minimum
standard 203/41

The study found out that several factors had a great effect on selecting the best future
employee. One of the most effective factors was SA which had the maximum gain ratio, which
made it the starting node and most effective attri bute. The attribute had positively affected the
performance with candidates who had more years of experience showing better performance than
those with less years of experience
Age attribute showed a positive effect on final performance indicator. This eff ect on
performance showed that younger candidates depicted poor performance. This could be due to
newly working candidates who did not have experience working in other companies. On the other
hand, older candidates may have had much experience that would i nfluence their performance. It
was observed that candidates between the age of 22 years and 28 years showed better performance.
Taking into account that the purpose of the hiring campaign is to attract young professional, an older
aged candidate would not fit this description regardless of his/hers overall performance indicator.
Age range has also a strong connection to FLA attribute, the younger candidates tend to have
better English skills but, when you correlate those two attribute with PE it shows that from the
overall poll of candidates, the ones with high FLA levels are also the ones with more than 3 years
previous work experience. This is simple to justify, nowadays corporate employees are using
English in their day to day tasks, thus improving their language level.
The SDQ attribute was in strong correlation with RAS attribute. Candidates with higher
academic qualification performed better at reasoning and analytical questions than the ones with
lower qualification. Figures from the experiments concluded that most ca ndidates with PhD and
Master‘s degree had outstanding performance and exceeded expectations respectively. Regarding
the impact on the performance attribute, SDQ levels showed that academic results don‘t influence
that much the candidate future performance but only if its correlated with previous work experience.

38
The Time efficiency TE attribute had a major impact on candidate performance and it is
strongly correlated with AR. The study has shown that older candidates tend to finish most of the
given tasks i n the specific time -frame than younger candidates.
Last, but not least, Communication skills CS had one of the biggest impact of the candidates
performance. The numbers show that candidates that have a high ease of communicating will better
perform that th e ones that have communication problems. Also CS is strongly linked with SA and
RAS, rational and analytical candidates tend to have a great communication skill set.
Clustering
In order to see how the candidates profiles are grouped to organize selection b atches, it has
to be performed the clustering process. To identify homogeneous groups of fitted candidates it is
used the KMeans algorithm. After performing the clustering (Appendix 7), KMeans i dentifies four
uniform clusters.
After analyzing the KMeans‘ r esults it was identified that the perfect match for the hiring
process is cluster number 0 because it has the highest level of overall attributes.
Tab 3.10 . Classification Clusters Generated By KMeans Algorithm
Cluster
Attribute Characteristics
0 1 2 3
AR 23-25 18-19 20-22 26-28
RAS 83.8372 33.2826 64.8485 64.4158
SDQ 7.8844 7.742 8.4246 7.477
TE 85.6512 59.6486 70.0303 73.4316
SA 88.6047 44.3514 68.8384 73.2474
CS 92.0853 49.9457 70.8889 72.3421
PE 1.6938 0 0.1869 0.3526
FLA 79.969 85.692 86.202 85.8526
P 8 4 5 6
Instances 129 ( 19%) 276 ( 40%) 99 ( 14%) 190 ( 27%)

According the clustering process, the selected candidate‘s profile is a 23 -25 years old, with
outstanding communication and analytical skills , that performs good within the given time -frame.
Also, that has a high overall score and no more than two years work experience. Regarding the
Kmeans result, the candidate that has this profile will ought to have the maximum future work
performance rate, thus ending the selection profile.
Model Evaluation
To measure the accuracy of the classifier, it has to be implemented some sensitivity and
specificity parameters. As for the detailed accuracy by class, all the values, except the false positives
(FP Rate) should be as closer to 1 as possible.
The following are the meaning of the variables used in the following equations.
 True positive = correctly recognized
 False positive = incorrectly recognized

39
 True negative = correctly rejected
 False negative = incorrectly rejected
1. True -Positive Rate (sensitivity) is the division of training samples predicted correctly by
model.
TPR =
(1)
where TPR represents the True -positive -Rate and as higher this value is, the better the model is.
2. False -Positi ve Rate (specificity) is the division of training samples predicted incorrectly by
model.
FPR =
(2)
where FPR represents the False positive Rate and lower this value, the better the model.
3. Area under ROC curve (Receiver Operating Charact eristics): is obtained by plotting TPR against
FPR.
The area under the plotted graph gives a good measure of the accuracy of the classifier. The
area can be as high as 1 (maximum accuracy) and as low as 0 (minimum accuracy). Since Area
under -ROC curve take s into consideration, both FPR and TPR values, this measure is preferred over
the parameters to compare accuracies between models.
Tab 3.11 . Accuracy measures for attributes classified by output class
Class FP Rate TF Rate Area under ROC
curve
JRip J48 JRip J48 JRip J48
Doesn‘t meet
minimum standard 0.958 0.916 0.129 0.057 0.919 0.954
Needs improvement 0.790 0.741 0.019 0.032 0.908 0.925
Average 0.925 0.977 0.015 0.024 0.882 0.939
Meet Expectation 0.934 0.848 0.089 0.101 0.903 0.898
Exceed Expectation 0.971 0.971 0.061 0.049 0.905 0.915
Outstanding 0.889 0.852 0.017 0.013 0.939 0.971
Weighted Average 0.933 0.925 0.063 0.050 0.908 0.931

It can be verified that both models present good classification values (more than 90% of the
instances were classified accordingly), but the model with the tree is slightly better at illustrating the
differences between the correctly classified instances.
Tab 3.12 . JRip confusion matrix
 a b c d e f < – classified as
 160 5 2 0 0 0 | a = Doesn’t meet minimum standard
 37 109 3 9 0 0 | b = Needs improvement
 14 4 52 27 2 0 | c = Average
 14 1 3 83 30 0 | d = Meet Expectation
 3 0 1 13 57 11 | e = Exceed Expectation
 0 0 0 1 5 48 | f = Outstanding

40
As for the confusion matrix, in this case it is not very relevant. But, it is more important that
the false positives be as near to zero as possible. As for false negatives, should be as close to zero as
possible in the extreme case of considering a poor candidate an outstanding one.
Tab3.13 . .J48 confusion matrix
 a b c d e f < – classified as
 153 11 1 2 0 0 | a = Doesn’t meet minimum standard
 28 117 4 9 0 0 | b = Needs improvement
 0 6 67 25 1 0 | c = Average
 2 0 9 98 22 0 | d = Meet Expectation
 0 0 0 20 57 8 | e = Exceed Expectation
 0 0 0 1 7 4 6 | f = Outstanding

Also, in this step it is intended to evaluate the models obtained in the previous step, as for the
quality and effectiveness. To evaluate classification algorithms there are several measures. The
following table shows the comparativ e values for the two models, JRip rules and trees, generated in
the previous step:
Tab 3.14 . Model effectiveness comparison of JRip and J48
Measure Rules
JRip Tree
J48
Correctly Classified
Instances 649
93.3429 % 678
97.5216 %
Incorrectly Classified
Instances 45
6.6571 % 16
2.4784 %
Kappa statistic 0.8709 0.8233
Mean absolute error 0.1218 0.0988
Root mean squared error 0.2631 0.2431
Relative absolute error 4.9219 % 4.4464 %
Root relative squared error 21.4719 % 20.0436 %
Total Number of Instances 694 694

Both models are very good based on kappa value (a close to 1 value indicates a very good
agreement between the model results and the instances), but the model with the tree is a little bit
better since it classifies correctly 97% of the instances, while the one of the rules classifies correctly
93% of them. Besides that, the errors, both absolute and relative, are smaller in the case of the trees.
The kappa statistics is bigger in the case of the rules, which indicate s a more stable model.
Deployment
The previous created model was implemented in Weka. The scope of this web application was to
map the results attained after modeling phase to code, by using class methods in PHP. The result of
the algorithms was in the for m of trees and rules, so this was converted to code in the form of if/else
statements. The statements were then transformed into PHP class method that accepted only the
splitting attributes i.e. Previous Experience, Age Range, Performance, etc. The class m ethod then

41
returned the final result of the evaluation, indicating the performance category of a candidate, and
the final selection decision was based on this number.
 Closing the selection
After the selection process was finalized, and a final number of 12 9 fitted candidates was
established, the next steps are in order:
 Email HR regarding selected candidate: in this stage the HR officer will contact them in
order to establish meetings, discuss the final offer and finalize this process.
 Send offer letter

3.3.3. Assessing the hiring process
 Assessing the recruiting process
For Deloitte the challenge was to prove that the results are measurable, and this was
accomplished by merely using Google analytics and integrating the results with the data mining
platform to measure where the candidates are coming from, how much money it takes to attract
those candidates and the conversion rate between traffic/applications and jobs.
Tab 3.15 . Deloitte websites traffic during the recruitment campaign (Sept 16 -May 17)
Source
KPI deloitte.ro workingatdeloitte.ro Increase
%
Visitors 54 024 154 163 177%
Unique visitors 47 425 108 018 130%
Average per day 178 240 130%
Returning visitors 11% 31% 182%

First the traffic, the figures speak for themselves, because Deloitte built this new website
from scratch workingatdeloitte.ro and now it largely surpasses the old Deloitte corporate website in
terms of traffic. As illustrated in tab 3.3.6., the traffic comes from the following platforms:
There is 134% more traffic coming from social media than from the other sources, mainly
job recruitin g platforms. This difference is easily explained by the presence of direct links to the
careers page on each of Deloitte‘s social media platforms, thus increasing the number of interactions
between candidates and the company.
Tab 3.16 . Traffic to career site from other platforms during the campaign (Sept 16 -May 17)
Online Platforms
Job Sites Social Media
ejobs.ro 901 Facebook 2615
bestjobs.ro 581 LinkedIn 327
hipo.ro 722 Twitter 14
Total Traffic 2204 Total Traffic 2956

Twitter is, so fa r, the least efficient network so is useless using it as a way to push jobs
forward, because it had no candidate company interaction whatsoever. Twitter is a conversation tool
probably not suited for a company like Deloitte in Romania. As a direct recruiti ng tool, Facebook

42
functioned well, posting jobs listing generated a good amount of applicants, and it can be used as a
community building tool, and thus Facebook seems to be one of the best options for Deloitte in thi s
case. And this page is ―liked‖ by 6,1 54 people which is the biggest community on social networks
for Deloitte Romania.
Deloitte had also significant success with LinkedIn as a direct recruitment tool, but can be used
as well as a way to raise their employer brand and engage with specific tar gets, using a wide range
of tools and tactics on LinkedIn like customizing the company profile with banners, videos and links
directing to the career website.
 Assessing the selection process
Deloitte‘s selection process evaluation identifies the internal capability and potential gaps,
candidate retention levels and the corresponding effectiveness of the process. It provides a
foundation for improving the selection process for future use. Firstly, the criteria used for measuring
is the applicants volume .

Tab 3.17 . Progress of the applicants‘ number during the hiring process
Stage Received
resumes Enrolled
candidates Selected
candidates Employed
candidates
Number 862 694 129 75

One of the indicators of the success of the selection process is the number of candidates who
applied for the position. A large numerical valuation of this indicator illustrates that the social media
campaign has had a big impact and has managed to attract a large number of potential candidates
862, with a hiring rate of 8.7%. Th e high number of candidates is explainable by the solely reason
that Deloitte is a big, multinational company that has brand awareness and is attractive for young
professionals.
It is a known fact that most of today's youth spend a lot of time on social me dia platforms so
it's no wonder they have a major impact on recruitment campaigns . The data presented in the table
below illustrates the origin of each candidate who has been employed, according to the platform
used to apply. This information is decisive i n running future campaign, because it shows what is the
most effective platforms for attracting the most fitted candidates.
Tab 3.18 . Employed candidates depending on the platform origin
No. Platform Amount Percentage
1 deloitte.com 5 6.66%
2 workingatdeloitte.com 11 14.66%
3 Twitter 0 0%
4 Facebook 23 30.66%
5 LinkedIn 13 17.33%
6 Campus recruitment 1 1.33%
7 Job fairs 2 2.66%
8 Job sites 16 21.33%
9 Employees recommendation 4 5.33%
Total 75 100%

43
To obtain this classification a new class attributes was introduced in the dataset previously
used for selection process through data mining. This new set of data illustrates the origin of each
candidate's resume in relation with the platform used for appl ying.
The information gained will help Deloitte to decide on the most appropriate strategy to use
for hiring and which online platform the company should focus on, the data showed that by far
Facebook leads as the most effective way to attract candidates who were later hired with a
percentage of 26.33, followed by specialized hiring sites with 21.33 percent, and LinkedIn.
As a conclusion of the assessment process, Deloi tte should focus its strategy for attracting
potential candidates from the following cha nnels: Facebook, LinkedIn and specialized recruiting
websites. They should also revise their age range. The selection process shows that both younger
group candidates 18 -19 years and older 27 -28 years do not fit their job requests so Deloitte should
focus their strategy on a smaller age group. This not only will improve the candidates‘ batch b ut will
also reduce the campaign cost.
Informational systems, when used appropriately, can be responsible firstly for the
opportunity of personal interaction, and ulti mately for personalizing the content. It allows
organizations to build a strong brand identity and a strong offer, but also it does involve on -going
commitment in the recruitment and selection strategy. A carefully coordinated informational system
recruitm ent campaign will enable any organization to attract the high quality candidates that any
organization needs.

44
Conclusion
HRIS is essential for any good organization to ensure effective people management and to
get a competitive edge in the corporate wor ld accordingly. In Romania, it is gradually getting
popularity and acceptability. The enterprise/business owners should consider it as an investment
rather than cost. Commitment of top management is required to remove the existing barriers to
success of HR IS. It will help management to make effective and better decision. The
communication between individual and line manager became faster. The HR services became self –
service rather than paper based transaction that resulted in more efficient HR department.
The times when HR professionals had only limited channels within their range to attract and
recruit new candidates are long gone. As soon as the technology developed, more and more
individuals have become increasingly active online, so it is in the best int erest of companies to use
this channel in their benefit, for attracting younger new talents. This research perfectly illustrated the
benefits and the easiness of running a social -media based recruiting campaign for an organization,
clearly pointing out tha t the purpose of it isn‘t just to attract candidates but also to attract individuals
and to make them interact with the organization, thus increasing the brand awareness.
To make sure you attract the right candidates, it‘s highly recommended to alternate the
innovative promoting channels with the conventional ones that offer you great exposure, both online
and offline. Traditional job panels for recruiting workforce are definitely not dismissed at Deloitte
Romania and they are still actively using them, but there has been a certain change in traffic.
Nowadays, most of the traffic generated from social media to their careers website is now greater
than the one from traditional job panel traffic.
Human resources signify the core of an economic activity. Rec ruitment of new workforce is
one of the most essential business processes that affect the quality of human capital within a
company. It is highly critical for companies like Deloitte to make sure the recruitment of right talent
to preserve a competitive ed ge over their competitors in the market.
Data -mining systems provide useful information from the past projects depending on which
the hiring -manager can make decisions for recruiting and selecting high -quality workforce. This
paper aimed to close that gap by developing a framework based on a social media recruiting strategy
and a data mining selection process to focus on the different conditions imposed for personnel
hiring.
During the study it was possible to see how certain factors had a major impact on the
selection process of the most suitable employee from a batch of candidates using data mining
techniques. They were: age, previous professional experience, school results, communication skills
and reasoning and analytical thinking, as well as the resul ts of the computerized assessment test.
These are the factors that can help in selecting the most suitable candidate by predicting the future
work performance.
The scope of this study can be further extended to a number of domains and suitable changes
need to be integrated when creating a selection framework for each one of them. This would involve
a preliminary research about finding the most applicable characteristics that can be directly
correlated to the performance of any given candidate in that partic ular field. Once this is completed,
a similar model can be used to create a feature ranking order and convert the most important
attributes that need to be measured to make a hiring decision by means of such a framework.

45
When data mining processes are adap ted into informal procedures like selecting an
individual, where question of subjectivity arises, and the software eliminates from the start this
complication , thus optimizing the process. T his type of framework can be extended into other
domains like asse ssing the future performance of the employee, determining the best motivational
factors for an employee, assessing the staffing needs, forecasting employee turnover and many
more.
In future work, it is compulsory for Deloitte to extend the selection of c andidates as a
continuous value instead of predicting performance. A comparative analysis of the classification
model and a prediction value based on past experience would aid into choosing a more robust data
mining model so that the software could be easi ly used by HR including the rules generated for
predicting performance of future employees.

46
Bibliography

1. Azevedo, A. & Santos, M.F. (2008). KDD, SEMMA and CRISP -DM: a Parallel
Overview . In Proceedings of the IADIS European Conference on Data Mining, DM2 008,
182-185.
2. Azevedo, A. & Santos, M.F. (2009). Business Intelligence: State of the Art, Trends, and
Open Issues . In Proceedings of the First International Conference on Knowledge
Management and Information Sharing – KMIS 2009, 296 -300.
3. Barber A., Recruit ing Employees: Individual and Organizational Perspectives ,
Thousand Oaks, CA, Sage, 1998.
4. Beardwell, I, Holden, L, Human Resource Management: A Contemporary Perspective ,
Ed. Prentice Hall, 201 0
5. Belcourt Monica, Sherman Arthur, Bohlander George, Snell Scott, Managing human
resources , Second Canadian Edition, ITP Nelson, Toronto, 1999.
6. Broughton, A., Ledermaier, B., Cox, A., ( 2013 ) – PDF. The use of social media in the
recruitment process [ONLINE] Available at: http://docplayer.net/203238 -2013 -andrea –
broughton -beth-foley -stefanie -ledermaier -and-annette -cox-ies.html. [Accessed 20 -02-2017].
7. Broughton, A., Ledermaier, B., Cox, A. (2017) Workplaces and Social Networking. The
Implic ations for Employment Relations [ONL INE] Available at:
http://docplayer.net/1719832 -Research -paper -workplaces -and-social -networking -the-
implications -for-employment -relatio ns-ref-11-11.html. [Accessed 20 -02-2017].
8. Bulls eye . 2017. Deloitte Reasoning Questions | Reasoning Test For Deloitte . [ONLINE]
Available at: http://cpt.hitbullseye.com/Deloitte -reasoning -test.php. [Accessed 2 -03-2017].
9. Bulmash, Julie – Strategic Planning for Human Resources , Pearson Prentice Hall , 2006.
10. Burloiu P. , (1997). Managementul resurselor umane , Editura Lumina Lex, Bucharest.
11. Byars Loyd, Rue Leslie, (1987). Human Resource Management , Homewood, Irwin,
Prentice Hall.
12. David G. Allen, (2012 ). Managing Employee Turnover: Dispelling Myths and Fostering
Evidence -Based Retention Strategies (Human Resource Management and Organizational
Behavior Collection). Edition. Business Expert Press.
13. Davison, Maraist and Bing , (2011). The Promise and Pitfalls of Using Social. Networking
Sites for HR Decisions , ED. Prentice Hall.
14. Dănălache, F., ( 2003 ). Managementul Resurselor Umane , Ed. Bren, Bucharest
15. eJobs Blog. 2017. Deloitte România în 2020: 800 de angajați și miza pe leadership –
eJobs Blog. [ONLINE] Available at: https://cariera.ejobs.ro/deloitte -romania -in-2020 -800-
de-angajati -si-miza -pe-leadership /. [Accessed 03 May 2017].
16. E-manageri – blogul managerilor de succes. 2017. August | 2012 | E-manageri – blogul
managerilor de succes. [ONLINE] Available at: https://emanageri.wordpress.com/2 012/08/.
[Accessed 10.12 .2015 ]
17. Fielding, H, 1988. Tom Jones (The Great Writers Library). Marshall Cavendish . [ONLINE]
Available at: http://www.bartleby.com/ebook/adobe/301.pdf Accessed 2 April 2017].
18. FlatWorld. 2017 . Human Resource Management 1.0 | Fla tWorld . [ONLINE]
Available: http://catalog.flatworldknowledge.c om/bookhub/2807?e=portolese dias_1.0 –
ch04_s03. [Accessed 20 -01-2016 ].
19. Fort, T, . 2017. Introduction in HRIS . Final Project. [ONLINE] Available at:
https://www.slideshare.net/VibhaKattige/final -project-66088645. [Accessed 20 -01-2017]

47
20. HR Payroll Systems. 2017. HRIS – Human Resources Information System . [ONLINE]
Available at: http://www.hrpayrollsystems.net/hris/. [Accessed 05-01-2017 ].
21. Human Resource Management. 2017. Human Resource Management » Selection Method.
[ONLINE] Available at: http://www.hrwale.com/recr uitment/selection -method/. [Accessed
15.01.2016]
22. Institutul National de Statistica. 2016. Romania in cifre. [ ONLINE] Available at:
http://www.insse.ro/cms/files/publicatii/Romania_in%20cifre%202011.pdf. [Accessed 9
April 2017].
23. John Kirimi, M., 2016. Application of Data Mining Classification in Employee
Performance Prediction. International Journal of Computer Applic ations Available at:
http://www.ijcaonline.org/archives/volume146/number7/kirimi -2016 -ijca-910883.pdf
[Accessed 12 March 2017].
24. Joos, J.G.(2008) , Management of the Human Resources and Millenials , 2008
25. Joos, J. G. (2008), Social media: New frontiers in hiring and recruiting . Empl. Rel.
Today, 35: 51 –59. doi:10.1002/ert.20188
26. Lefter V., Manolescu A., Deaconu Alexandrina, Rașca Lavinia, Popa I., Mihalache V. –
Managementul resurselor umane: studii de caz, probleme, teste , Editura Economică,
1999.
27. Lucian Claudiu Anghel, Arhitectura și caracteristicile sistemului informatic necesare
pentru managementul resurselor și plasamentelor bancare în contextul actual , Revista
Informatica Economica, nr. 2 (22)/2002
28. Milkovich, G. T., Boudreau J. W., Human Resource Management , Sixth edition Irwin,
Boston, 1991.
29. Mews, C., (2017) Selection Methods for Graduate Talent | GRB. [ONLINE] Available at:
https://w ww.grb.uk.com/recruiter -research/selection -methods. [Accessed 20 June 2017].
30. Mondy, R. Wayne; Noe, Robert M.; Premeaux, Shane R, (2001 )Human Resources
Management , Ed. Blackwell Publishing Ltd
31. Novac Emilia, (2003). Managementul resurselor umane , Editura Waldpress, Timișoara
32. Onete B. (2000) – Sisteme informatice , Edit ura A.S.E București, București.
33. Parry and Tyson , (2008). An analysis of the use and success of online recruitment
methods , Ed. Blackwell Publishing Ltd
34. Pelster, B, 2016. Rewriting the rules fo r the digital age . Deloitte University Press,
[Online]. 2, 144. Available at: http://deloitte/studies -on-employee -statistics -worldwide.com
[Accessed 4 May 2017].
35. Pisău Gheorghe, Toma C., Mihăilescu I., Elaborarea și introducerea sistemelor
informatice , Ed. Tehnică, 1975
36. Radu, I. (coord) – Informatică si management Ed. Universitara 2005;
37. Recruitment Strategy – HRM Guide. 2017. Recruitment Strategy – HRM Guide .
[ONLINE] Available at: http://www.simplehrguide.com/recruitment -strategy.html.
[Accessed 03 May 20 17].
38. Remy, S., Placement Paper of DELOITTE for All UG & PG. 2017. Admission Open:
Placement Paper of DELOITTE for All UG & PG. [ONLINE] Available at:
http://anantoacademy.blogspot.com/2016/10/placement -paper -of-deloitt e-for-all-ug.html.
[Accessed 20 -05-2017].
39. Resurse umane: Etapele recrutarii si selectiei. 2017. Resurse umane: Etapele recrutarii si
selectiei. [ONLINE] Available at: http://legislatiamuncii.manager.ro/a/148/resurse -umane –
etapele -recrutarii -si-selectiei.html. [Accessed 11.10 .2015 ]

48
40. Sărătean Elena, Managementul resurselor umane Suport al schimbării , Editura
Mirton.Timișoara, 2003.
41. Stan C. , Managementul resurselor umane , Editura Mirton, Timișoara, 2002.
42. Taking the Recruitment Industry Forward. 2017. Taking t he Recruitment Industry
Forward [ONLINE]
43. Targowski, Andrew S & Deshpanade, Satish P. The Utility and Selection of an HRIS .
Advances in Competitive Research, 2001.
44. University of Leeds Blog. Interviews and other selection methods . [ONLINE]
Available: http://hr.leeds.ac.uk/info/23/interviews_a nd_other_selection_methods/154/what_is
_selection_and_why_is_it_important_to_get_right [Accessed 2 -04-2016 ].
45. Ursacescu, M. (2002) – Sisteme informatice – o abordare între clasic si Modern Ed.
Economică 2002;
46. Walker, Alfred J. How the Web and Other Trends are Changing Human Resources ,
McGraw -Hill, 2004
47. Walker, Alfred J., (2011) Web -Based Human Resources . New York: McGraw -Hill
48. Watson Wyatt. (2002). e-HR: Getting Results Along the Journey , Survey Report. Watson
Wyatt Worldwide , 2002
49. Wilson, D. Communication and C ognition , Ed. Prentice Hall ,1995.
50. Zhao, X.(2008). A Study of Performance Evaluation of HRM: Based on Data Mining ,
FITME 2008, International Seminar on Future Information Technology and Management
Engineering
51. 92% Of Companies Use Social Media For Recruitment [INFOGRAPHIC] – Adweek. 2017.
92% Of Companies Use Social Media For Recruitment [INFOGRAPHIC] – Adweek.
[ONLINE] Available at: http://www.adweek.com/socialtimes/social -media –
recruiting/492059. [Accessed 3 -02-2016 ].
Available at: https://www.isixsigma.com/operations/human -resources/taking -recruitment –
industry -forward/. [Accessed 09 June 2016].

49
Appendices
Appendix 1
List of abbreviations
 CRISP DM Cross Industry Standard Process for Data Mining
 DM Data mining
 HR Human Resources
 HRIS Human Resources Information System
 HRM Human Resources Management
 HRMS Human Resources Management System
 IT Information Technology
 SAP Systems, Applications and Products
 SME Small and Medium Enterprise
 SNS Social Network Sites

50
Appendix 2
List of tables
NO
. TABLE PAGE
1 Tab 1.1. The main advantages/disadvantages of the recruitment sources 9
2 Tab 1.2. The differences between the advantages and disadvantages of
testing employees 15
3 Tab. 2.1. HRIS benefits and limitations 19
4 Tab 2.2. Examples of softwares used for recruiting and selection 20
5 Tab 2.3. Examples of sites used for social networking recruiting 22
6 Tab 3.1. Deloitte‘s competitors according to specific industry 26
7 Tab 3.2 . Deloitte hiring goal 28
8 Tab 3.3. Description of the hiring process activities 31
9 Tab 3.4 . The hiring process task duration 32
10 Tab 3.5 . Candidate profile features 34
11 Tab 3.6 . Recruitment Strategy Planning for Deloitte 35
12 Tab 3.7 . Resumes number according to advertising source 36
13 Tab 3.8. Classification Algorithms Accuracy Rate 38
15 Tab 3.9. Classification Rules Generated By JRip Algorithm for
Forecasting Performance 39
16 Tab 3. 10. Classification Clusters Generated By KMeans Algorithm 41
17 Tab 3.11 . Accuracy measures for attributes classified by output class 42
18 Tab3.12 . JRip confusion matrix 42
19 Tab3. 13. .J48 confusion matrix 43
20 Tab 3.14 . Model effectiveness comparison of JRip and J48 43
21 Tab 3.15 . Deloitte websites traffic during the recruitment campaign 44
22 Tab 3.16 . Traffic to career site from other platforms 44
23 Tab 3.17 . Progress of the applicants‘ number during the hiring process 45
24 Tab 3.18 . Employed candidates depending on the platform origin 45

List of figures
NO. FIGURE PAGE
1 Fig. 1.1.The stages of HR recruitment process (Taylor, 2005) 6
2 Fig. 1.2.The types of HR recruitment sources 7
3 Fig 2.1. Level of access increases from HR staff having access through
to all managers and employees having access (Bulmash, 2006) 16
4 Fig 2.2 . Guideline steps for social recruiting 23
5 Fig 3. 1. Division of Deloitte‘s employees according to gender 27
6 Fig 3.2. Division of Deloitte‘s employees in the regional offices 27
7 Fig 3.3. University of origin of Deloitte‘s employees 28
8 Fig 3.4. The hiring process using informational systems 30
9 Fig 3.5 . Candidate selection decision tree 39

51
Appendi x 3
Job Description
Tax Graduate Recruitment 2017 Deloitte Romania
Tax/ Audit Assistant
Job type: full-time
Job level: 0 – 2 years of experience
Location: BUCHAREST

Come work as a business analyst at a growing company that offers exceptional benefits with
opportunities to advance and learn alongside accomplished business leaders.

The Company:
Deloitte is the brand under which tens of thousands of dedicated professionals in
independent firms throughout the world collaborate to provide audit, consulting, financial advisory,
risk management, and tax services to selected clients.
We of fer great mentoring programs for our analysts with top execs.
We‘re on the cutting edge of technology – you‘ll be working with the best equipment and software.

The Position:
If you are still a student in your last year of study (either studying for bachel or or master
degree), you can already apply and follow all selection steps, with the aim of joining us after
graduation (estimated starting time September 2017).
Our approach is to understand a client's business, the risks it faces, and its internal contro ls. Based
on our understanding we assess the risk of material misstatement of the financial statements so that
we design effective audit procedures. Our professionals bring a comprehensive methodology that
leverages the firm's extensive professional and te chnical resources to our clients' financial reporting
process, helping them meet the growing demand for enhanced disclosure and accountability.

Qualifications:
• Bachelor or master degree (Accounting, Finance, Business Administration or other economic
major preferred, but not mandatory) obtained or to be obtained in 2017; Other non –
economic degrees (e.g. Engineering) are eligible provided the candidates demonstrate active
interest in finance and business matters;
• Previous exposure to business environment an d/or involvement in extracurricular activities
will be appreciated;
• Good academic credentials;
• Demonstrated problem solving skills, initiative and business awareness;
• Strong verbal and written communication skills (Romanian and English);
• Ability to priorit ize tasks and work on multiple assignments;
• Ability to work both independently and as part of a team with professionals at all levels;
• Willingness to travel for out -of-town engagements

Responsibilities
An Audit Assistant in our team:
• Develops an understanding of the Deloitte audit approach, methodology & tools;
• Establishes working relationships with client personnel;
• Performs analytical review of audit documents;

52
• Reviews client accounting and operating procedures and systems of internal controls;
• Identifies accounting and auditing issues; perform research to solve issues that arise;
• Applies concepts of risk assessment;
• Provides recommendations for improved controls and enhanced business efficiency.
• Exhibits a professional, business -like demeanor;
• Pursues ACCA certification.
•
Why Should You Apply?
• Very competitive benefits package.
• Excellent growth and advancement opportunities.
• Bonus program.
• Remote working opportunities.

Interested?
Please contact us.

53
Appendix 4
Evaluation Test
This test is a written exam that evaluates knowledge, reasoning and analytical skills and ability to
communicate in writing. It consists of 30 MULTIPLE CHOICE questions with one correct answer.
You should attempt all questions.
Test duration: 45 minutes
1. What will be the next term in the following series? A B B D C F D H E J
a. N
b. O
c. K
d. F
e. L
2. REPORT is related to UGSQUV in the same way as LONGER is related to
a. OQPIHT
b. OQQFGS
c. OQQIHT
d. QPQHIT
e. None of these
3. Which of the following will come next in t he series? a z a b y a b c x a b c d w a b c d
a. f
b. u
c. a
d. y
e. e
4. How many such pairs of letters are there in the word CONVULSE, which have as many
letters among them in the word as in the English alphabet?
a. Two
b. Three
c. Four
d. One
5. Andrew walked 2m towards east, took a right turn and walked 7m. He then took a left turn
and walked 5m before taking a left turn and walking 7m. He then took a final right turn and
walked 1m before stopping. How far is Andrew from the starting point?
a. 8m
b. 7m
c. 6m
d. 5m
e. 9m
6. 4, 16, 36, 64, 100, ?
a. 121
b. 142
c. 144
d. 120
7. Study the given information carefully and answer the given questions: A is the mother of B.
B is the sister of C. D is the son of C. E is the brother of D. F is the mother of E. G is the
granddaughter of A. H has o nly two children – B and C. How is C related to E?

54
a. Father
b. Son
c. Cousin brother
d. Cannot be determined
8. In the following series, one of the terms is wrong. Which? 102, 101, 98, 93, 86, 74, 66, 53
a. 101
b. 66
c. 74
d. 93
9. If '+' means ÷, '÷' means 'x', 'x' means ' -', '-' means '+', then (75 x 25) ÷ 2 +50 – 10 = ?
a. 26.67
b. 12
c. 77.5
d. 25
10. In a certain code RAJAN is coded as ‗91115‘. How is IQBAL written in that code?
a. 98213
b. 27582
c. 12968
d. 18213
e. None of these
11. The mean daily profit made by a shopkeeper in 30 days was 350. If the mean pro fit for the
first fifteen days was 275, then the mean profit for the last 15 days would be
a. 200
b. 350
c. 275
d. 425
12. There were 35 students in a hostel. If the number of students increases by 7, the expenses of
the mess increase by 42 per day while the average expe nditure per head diminishes by 1.
Find the original expenditure of the mess.
a. 480
b. 520
c. 420
d. 460
13. The ratio between the number of passengers travelling by I and II class between the two
railway stations is 1:50, whereas the ratio of I and II class fares between the same stations is
3:1. If on a particular day 1,325 were collected from the passengers travelling between these
stations, then what was the amount collected from the II class passengers?
a. 750
b. 1000
c. 850
d. 1250
14. A boat travels upstream from B to A and downstream from A to B in 3 hours. If the speed of
the boat in still water is 9 km/hour and the speed of the current is 3 km/hour, the distance
between A and B is
a. 4 km
b. 8 km
c. 6 km
d. 12 km

55
15. A man while returning from his factory travels 2/3 of the distance by bu s, ¾ of the rest partly
by car and partly by foot. If he travels 2 km on foot, find the distance covered by him.
a. 24 km
b. 22 km
c. 28 km
d. 26 km
16. The fuel indicator in a car shows 1/5th of the fuel tank as full. When 22 more liters of fuel
are poured in to the tan k, the indicator rests at the 3/4of the full mar k. Find tank‘s capacity .
a. 25 liters
b. 35 liters
c. 30 liters
d. 40 liters
17. A pump can be operated both for filling a tank and for emptying it. The capacity of the tank
is 2400 m3. The emptying capacity of the pump is 1 0m3 per minute higher than its filling
capacity. Consequently, the pump needs 8 minutes less to empty the tank than to fill it. Find
the filling capacity of the pump.
a. 45 m3/min
b. 40 m3/min
c. 50 m3/min
d. 55 m3/min
18. A sum of money is accumulating at compound intere st at a certain rate of interest. It simple
interest instead of compound were reckoned, the interest for the first two years would be
diminished by 20 and that for the first three years, by 61. Find the sum
a. 7000
b. 8000
c. 7500
d. 6500
19. In a kilometer race, A can gi ve B a 100 m start and C a 150 m start. How many meters start
can B give to C?
a. 50
b. 50/9
c. 8500/9
d. 500/9
e. None of these
20. The average age of all the student of a class is 18 years. The average age of boys of the class
is 20 years and that of the girls is 15 years. If the number of girls in the class is 20, then find
the number of boys in the class.
a. 15
b. 45
c. 30
d. 50
B. DIRECTIONS for the questions 21 to 22: In each of these questions, a statement is followed by
two assumptions numbered I and II. An assumption is somethi ng supposed or taken for granted.
Mark answer as :
a. if only Assumption II is implicit.
b. if only Assumption I is implicit.
c. if either Assumption I or II is implicit.
d. if neither Assumption I nor II is implicit.

56

21. Statement: Lack of stimulation in the first four years of life can have adverse consequences.
Assumptions:
I. A great part of the development of observed intelligence occurs in the earliest years of life.
II. 50 percent of the measurable intelligence at the age of 17 is already predictable by age of four.
a. 1
b. 2
c. 3
d. 4
22. Statement: Take this 'oven' home and you can prepare very tasty dishes which you were
unable to prepare earlier – an advertisement of X brand oven.
Assumptions:
I. The user knows the procedure recipe of tasty dishes but does not have a proper oven to cook.
II. Only ‗X' brand oven can cook very tasty dishes.
a. 1
b. 2
c. 3
d. 4
DIRECTIONS for 25. Pick the correct choice for each of the following sentence.
23. ____________ the room the door hit him on the brow.
a. while he was entering
b. As he entered
c. On entering
d. On having entered
DIRECTIONS for the questions 24 -25: Some of the sentences have errors and some have none.
Find out which part of the sentence has an error.
24. He returned back (1)/ when he sensed (2)/ that there was danger (3)/ No error(4)
a. 1
b. 2
c. 3
d. 4
25. The progress of th ese plants (1)/ has not been very encouraging (2)/ and they are in various
stage of decay (3)/ No error (4)
a. 1
b. 2
c. 3
d. 4
DIRECTIONS for the question 26 to 30: Read the passages given below and answer the questions
that follow each passage.
―I hope my reader wi ll be convinced, at his very entrance of this work, that he will find in the
whole course of it nothing prejudicial to the cause of religion and virtue, nothing inconsistent with
the strictest rules of decency, nor which can offend even the chastens eye in the perusal. On the
contrary, I declare that to recommend goodness and innocence has been my sincere endeavor in this
history. This honest purpose you have been pleased to think I have attained: and to say the truth, it
is likeliest to be attained in book s of this kind; for an example is a kind of picture, in which virtue

57
becomes, as it were, an object of sight, and strikes us with that loveliness, which Plato assures there
is in her naked charms.
Besides displaying that beauty of virtue which may attract the admiration of mankind, I have
attempted to engage a stronger motive to human action in her favor, by convincing men, that their
true interest directs them to a pursuit of her. Lastly, I have endeavored strongly to inculcate, that
virtue and innocence can scarce ever be injured but by indiscretion. A moral which I have the more
industriously labored, as the teaching it is, of all others, the likeliest to be attended with success;
since, I believe, it is much easier to make good men wise, than bad men go od.
For these purposes, I shall submit to the candid reader, with only two requests: first, that he
will not expect to find perfection in this work; and secondly, that he will excuse some parts of it, if
they fall short of that little merit whic h I hope m ay appear in others. I will detain you, sir, no longer.
Indeed I have run into a preface, while I professed to write a dedication. But how can it be
otherwise? I dare not praise you; and the only means I know of to avoid it, when you are in my
thoughts, are either to be entire ly silent, or to turn my subjects to some other subject.
Pardon, therefore, what I have said in this epistle, not only without your consent, but
absolutely against it; and give me leave, in this public manner, to declare that I am, with the highest
respect, and gratitude, Sir, Your most obliged, obedient, humble servant…. ―
Fielding, H, 1988. Tom Jones (The Great Writers Library). Marshall Cavendish.
26. The tone used by the author in the beginning of the passage can best be described as
a. mellow
b. deferential
c. polite
d. courteous
27. How has the author tried to cause a favora ble opinion of virtue from the readers?
a. By displaying virtue always wins
b. By showing that a thing gained by a way of virtue is better than a thing obtained by vice
c. The Govt. has recently decided to provide post qualification professional training to all
engineering graduates at its own cost
d. By substantiating his claim in proving to the readers that nothing got out of unfair means
can compensate for the loss of inward peace
28. In the first paragraph, 'Her naked charms' – 'her' here refers to
a. virtue
b. picture
c. objec t
d. book
29. The author seeks forgiveness because
a. he has detained the reader
b. he did not seek his patron's permission
c. he went against the wishes of his patron
d. he wrote a preface instead of a dedication
30. The preface written by the author is likely to be followed by
a. an essay on virtue and vice
b. a play on virtue and vice
c. a comedy that laughs at absurdities
d. a fable that stresses the importance and frailties of virtue

58
Appendix 5
JRip Algorithm Data mining Report
=== Run information ===
Scheme: weka.classifiers.rules.JRip -F 3 -N 2.0 -O 2 -S 1
Relation: deloitte -selection –
weka.filters.unsupervised.attribute.NumericToNominal -R12-
weka.filters.unsupervised.attribute.Remove -R9-11-
weka.filters.unsupervised.attribute.NumericToNominal -R9
Instances: 694
 Attributes: 9
 AR
 RAS
 SDQ
 TE
 SA
 CS
 PE
 FLA
 P
Test mode: 10 -fold cross -validation
=== Classifie r model (full training set) ===
JRIP rules:
===========
 (SA >= 93) and (AR <= 25) => P=8 (30.0/0.0)
 (SA >= 85) and (TE <= 82) => P=8 (21.0/5.0)
 (SA >= 80) and (TE <= 74 ) and (TE >= 73) and (CS <= 90) => P=8
(6.0/1.0)
 (SA >= 68) and (CS >= 96) => P=7 (14.0/1.0)
 (SA >= 80) and (RAS <= 61 => P=7 (14.0/1.0)
 (SA >= 68) and (AR <= 20) => P=7 (43.0/15.0)
 (SA >= 74) and (TE <= 77) and (SA >= 79) and (SDQ >= 8.44) => P=7
(9.0/1.0)
 (SA >= 81) and (TE <= 82) => P=7 (7.0/0.0)
 (SA >= 74) and (AR <= 21) and (SA <= 75) => P=7 (6.0/1.0)
 (SA >= 53) and ( SA <= 59) and (AR >= 19) => P=5 (29.0/2.0)

59
 (SA >= 57) and (AR <= 18) => P=5 (19.0/0.0)
 (SA >= 63) and (TE >= 93 ) and (AR <= 26) and (TE <= 96) => P=5
(8.0/1.0)
 (SA <= 74) and ( SA >= 71) and (AR >= 22) => P=5 (12.0/2.0)
 (SA >= 64) and ( AR <= 21) and (SA >= 70) => P=6 (28.0/1.0)
 (SA >= 78) => P=6 (68.0/27.0)
 (SA >= 61) and ( AR <= 20) and (SA >= 64) => P=6 (24.0/2.0)
 (SA >= 60) and ( AR <= 19) => P=6 (15.0/2.0)
 (SA >= 69) and (AR <= 22) and (SA >= 75) => P=6 (9.0/1.0)
 (SA >= 43) and (AR <= 19) => P=4 (91.0/1.0)
 (SA >= 61) and (S A >= 64) and (RAS<= 75) =>P=4( 38.0/11.0)
 P=3 (203.0/41.0)
Number of Rules : 21
Time tak en to build model: 0.4 7 seconds
=== Stratified cross -validation ===
=== Summary ===
 Correctly Classified Inst ances 649 93.3429 %
 Incorrectly Classified Instances 45 6.6571
%
 Kappa statistic 0.8709
 Mean absolute error 0.1218
 Root mean squared error 0.2631
 Relative absolute error 4.9219 %
 Root relative squared error 21.4719 %
 Total Number of Instances 694
=== Detailed Accuracy By Class ===
 TP Rate FP Rate Precision Recall F -Measure MCC ROC
Area PRC Area Class
 0.958 0.129 0.702 0.958 0.810 0.755 0.919
0.691 3
 0.790 0.019 0.916 0.690 0.787 0.747 0.908
0.818 4
 0.925 0.015 0.852 0.525 0.650 0.630 0.882
0.654 5
 0.934 0.089 0.624 0.634 0.629 0.542 0.903
0.666 6
 0.971 0.061 0.606 0.671 0.637 0.584 0.905
0.565 7

60
 0.889 0.017 0.814 0.889 0.850 0.837 0.939
0.787 8
 Weighted Avg. 0.933 0.063 0.754 0.733 0.730
0.680 0.908 0.702
=== Confusion Matrix ===
a b c d e f < – classified as
 160 5 2 0 0 0 | a = 3
 37 109 3 9 0 0 | b = 4
 14 4 52 27 2 0 | c = 5
 14 1 3 83 30 0 | d = 6
 3 0 1 13 57 11 | e = 7
 0 0 0 1 5 48 | f = 8

61
Appendix 6
J48 Algorithm Data mining Report
=== Run information ===
Scheme: weka.classifiers.trees.J48 -C 0.001 -M 2
Relation: Deloitte -selection –
weka.filters.unsupervised.attribute.NumericToNominal -R12-
weka.filters.unsupervised.attribute.Remove -R9-11-
weka.filters.unsupervised.attribute.NumericToNominal -R9
Instances: 694
Attributes: 9
 AR
 RAS
 SDQ
 TE
 SA
 CS
 PE
 FLA
 P
Test mode: 10 -fold cross -validation
=== Classifier model (full training set) ===
J48 pruned tree
––––––
SA <= 63
| SA <= 42: 3 (131.0/13.0)
| SA > 42
| | AR <= 20
| | | SA <= 52: 4 (77.0/1.0)
| | | SA > 52
| | | | SA <= 59

62
| | | | | AR <= 18
| | | | | | SA <= 56: 4 (14.0)
| | | | | | SA > 56: 5 (9.0)
| | | | | AR > 18
| | | | | | SDQ <= 6.85: 5 (4.0/2.0)
| | | | | | SDQ > 6.85: 5 (25.0)
| | | | SA > 59
| | | | | AR <= 19
| | | | | | AR <= 18: 5 (6.0)
| | | | | | AR > 18: 6 (15.0/2.0)
| | | | | AR > 19
| | | | | | SA <= 60: 3 (4.0)
| | | | | | SA > 60: 4 (13.0/6.0)
| | AR > 20: 3 (31.0/4.0)
SA > 63
| SA <= 79
| | AR <= 21
| | | SA <= 66
| | | | AR <= 20
| | | | | AR <= 18: 5 (3.0)
| | | | | AR > 18: 6 (20.0/1.0)
| | | | AR > 20: 4 (9.0/1.0)
| | | SA > 66
| | | | SA <= 69

63
| | | | | AR <= 20
| | | | | | AR <= 19: 5 (3.0/1.0)
| | | | | | AR > 19
| | | | | | | SA <= 67: 6 (4.0/1.0)
| | | | | | | SA > 67: 7 (7.0 )
| | | | | AR > 20: 5 (8.0/1.0)
| | | | SA > 69
| | | | | AR <= 20
| | | | | | SA <= 73
| | | | | | | SDQ <= 8.93: 6 (11.0/2.0)
| | | | | | | SDQ > 8.93: 7 (4.0)
| | | | | | SA > 73: 7 (18.0/3.0)
| | | | | AR > 20: 6 (30.0/7.0)
| | AR > 21
| | | SA <= 70: 4 (19.0/2.0)
| | | SA > 70
| | | | SA <= 74: 5 (12.0/2.0)
| | | | SA > 74
| | | | | SA <= 76
| | | | | | AR <= 23: 6 (8.0/2.0)
| | | | | | AR > 23: 3 (18.0/7.0)
| | | | | SA > 76
| | | | | | SDQ <= 9.27: 6 (25.0/9.0)
| | | | | | SDQ > 9.27: 7 (4.0/2.0)

64
| SA > 79
| | AR <= 25
| | | SA <= 84
| | | | F LA <= 99: 7 (40.0/13.0)
| | | | FLA > 99: 6 (2.0)
| | | SA > 84: 8 (51.0/5.0)
| | AR > 25
| | | SA <= 95: 6 (55.0/27.0)
| | | SA > 95: 7 (14.0/1.0)
 Number of Leaves : 33
 Size of the tree : 65
Time taken to build model: 0.05 seconds
=== Stratified cross -validation ===
=== Summary ===
 Correctly Classified Instances 678 97.5216
%
 Incorrectly Classified Instances 16 2.4784 %
 Kappa statistic 0.8233
 Mean absolute e rror 0.0988
 Root mean squared error 0.2431
 Relative absolute error 4.4464 %
 Root relative squared error 20.0436 %
 Total Number of Instances 694

=== Detailed Accuracy By Cl ass ===
 TP Rate FP Rate Precision Recall F -Measure MCC ROC
Area PRC Area Class
 0.916 0.057 0.836 0.916 0.874 0.834 0.954
0.859 3
 0.741 0.032 0.873 0.741 0.801 0.753 0.925
0.843 4

65
 0.977 0.024 0.827 0.677 0.744 0.712 0.939
0.779 5
 0.848 0.101 0.632 0.748 0.685 0.608 0.898
0.640 6
 0.971 0.049 0.655 0.671 0.663 0.615 0.915
0.610 7
 0.852 0.013 0.852 0.852 0.852 0.839 0.971
0.798 8
 Weighted Avg. 0.925 0.050 0.784 0.775 0.776
0.729 0.931 0.768
=== Confusion Matrix ===
a b c d e f < – classified as
 153 11 1 2 0 0 | a = 3
 28 117 4 9 0 0 | b = 4
 0 6 67 25 1 0 | c = 5
 2 0 9 98 22 0 | d = 6
 0 0 0 20 57 8 | e = 7
 0 0 0 1 7 46 | f = 8

66
Appendix 7
KMeans Algorithm Data mining Report
=== Run information ===
Scheme: weka.clusterers.SimpleKMeans -init 0 -max-candidates
100 -periodic -pruning 10000 -min-density 2.0 -t1 -1.25 -t2 -1.0 -N
4 -A "weka.core.EuclideanDistance -R first-last" -I 500 -num-slots
1 -S 10
Relation: Deloitte -selection –
weka.filters.u nsupervised.attribute.NumericToNominal -R12-
weka.filters.unsupervised.attribute.Remove -R9-11-
weka.filters.unsupervised.attribute.NumericToNominal -R9
Instances: 694
Attributes: 9
 AR
 RAS
 SDQ
 TE
 SA
 CS
 PE
 FLA
 P
Test mod e: evaluate on training data
=== Clusterin g model (full training set) === kMeans ======
Number of iterations: 9
Within cluster sum of sq uared errors: 514.3900893613613
 Initial starting points (random):
 Cluster 0: 28,90,8.96,90,85,90,4,85,4
 Cluster 1: 19,35,6.32,62,58,40,0,95,4
 Cluster 2: 20,65,7.98,61,61,63,0,96,4
 Cluster 3: 20,59,7.26,61,61,62,0,90,4

Missing values g lobally replaced with mean/mode

67
Final cluster centroids:
Cluster#
Attribute Full Data 0 1 2 3
(694.0) (129.0) (276.0) (99.0) (190.0)
==================================================================
AR 21.0893 24.2868 18.5761 21.5859 26.6316
RAS 55.7061 83.8372 33.2826 64.8485 64.4158
SDQ 7.7933 7.8844 7.742 8.4246 7.477
TE 69.7363 85.6512 59.6486 70.0303 73.4316
SA 63.9813 88.6047 44.3514 68.8384 73.2474
CS 66.8977 92.0853 49.9457 70.8889 72.3421
PE 0.438 1.6938 0 0.1869 0.3526
FLA 84.745 79.969 85.692 86.202 85.8526
P 5 8 4 5 6
Time taken to build model ( full training data) : 0.01seconds
=== Model and evaluation on training set ===
Clustered Instances
 129 ( 19%)
 276 ( 40%)
 99 ( 14%)
 190 ( 27%)