Page of1 24
“RESEARCH PROPOSAL”
“Employee Turnover in Unilever”
AUTHORED BY:
SUMMAYA SHARIF
Page of2 24
1.A Working Title: 3
2.BACKGROUND: 3
3.Literature Review: 3
4. Problem Statement: 4
4(a): Research Objective: 4
4(b):Research Questions: 4
1. Conceptual Framework: 5
6. Theoretical Framework: 5
7. Scope of the Study: 6
8. Relevance of the study: 6
9. RESEARCH DESIGN 7
9.RESEARCH DESIGN (explained) 7
9(a). PURPOSE OF STUDY: 7
9(b). EXTENT OF RESEARCHER INTERFERENCE: 8
9(c). STUDY SETTING: 9
9(d). RESEARCH STRATEGIES: 9
9(d) MEASUREMENT AND MEASURES: 10
9(e). UNIT OF ANALYSIS: 12
9(f). SAMPLING DESIGN: 12
9(G). TIME HORIZON: 13
9(h). DATA COLLECTION METHOD: 13
9(i). DATA ANALYSIS: 17
• HYPOTHESIS TESTING: 21
10. TIME FRAME OF THE STUDY: 23
11. BUDGET: 23
BIBLIOGRAPHY: 24
Page of3 24
1.A Working Title:
“Employee turnover in Unilever Pakistan”
2.BACKGROUND:
Unilever was established in Pakistan as a result of merger between Unilever and British
soap maker’s LEVER BROTHERS in 1984. Unilever is the largest fast moving
(FMCG’s) in Pakistan, with an experience of 50 years. It has a variety of brands in Food
Brands, Home Care and Personal Care. It is the market leader in 7 categories out of 8
categories. Unilever has employed 2200 people directly and 1000 indirectly by its
suppliers, vendors and distributors. Moreover, it operates in 100 different countries
across the globe. Unilever is working with its suppliers to support feasible approaches to
agriculture. Unilever is also learning from NGO’s and other organizations to contribute
towards the welfare of the people and the country.
3.Literature Review:
Employee turnover is the rotation of workers around the labor market; between firms,
jobs and occupations; and between the states of unemployment and employment
Abassi et al. (2002). Mondy (2010) clearly defined turnover rates as how many new
recruitments were hired to replace resigned employees. Price (1997) define employee
turnover as the ratio of the total numbers of members who have left the organization
during a considered period divided by the average number of employees during the
period.Unskilled labor is when individual’s level of satisfaction is not involved in the
issue. A person is able to do a particular work but yet he may not be satisfied. The
place, environment, or reduced activity might not suit him. A large number of notions
influences likes and dislikes. The worker moves for example in the state of unskilled
labor until he finds work that is according to his capabilities and skills (George, 1919).
Work-life balance is important for an individual psychological well-being, and that of high
self-esteem, satisfaction, and overall sense of harmony in life can be regarded as
indicator of a successful balance between work and family roles (Clark, 2000; Clarke et
al. 2004;Mark and MacDermid,1996).Interventions of work demand into personal life
(e.g: working during the week-end) resulted into a disturbing work life balance among
the employees (Hyman et al. 2003).Job stress and declining work life balance (eg: work
overload, job ambiguity) are the factors that triggers the chain of psychological states
that leads to employee turnover (Firth et al. 2003).
Page of4 24
Work-life balance is the ability to experience a sense of control and to stay productive
and competitive at work while maintain a healthy life style. It is achieving task with time
and attention (Jeff Davidson, 2014). Work life balance is to manage your work and
manage your life (Groysberg, 2014). Topel and Ward (1992) finds that thosewith better
economic conditions like jobs facilitating and with a wage growth rate are particularly
less likely to encounter with employee turnover than jobs that offer a low wage growth
rate.Campion (1991) cited in Tang suggests that the variable reason for high turnover
rates in organizations is higher wages and career opportunities employee seeks for.
Clark et al. (1998) argues that if performance related pay workers are more satisfied
with their jobs they may experience lower turnover than traditional times rate workers.
Griffeth et al. (2000) recorded pay and pay related variables have a modest effect on
turnover.There is a strong evident which state that age heterogeneity increase the rate
of employee turnover (Richard, 2002).People with an age of 30-45 and those over age
45 have more satisfaction with job security and desire to remain with their company as
compared to those age under 30. Also, for the under-30s, satisfaction with opportunities
to develop technical skills and pay linked to individual performance has a strong
negative relationship with willingness to change companies than for those over 45
(finegold,2002). Determining how to structure the employment relationship may be
difficult, however, since young people in today’s workforce are often describe as
generation wanting very different things from work than their senior person (Conger,
1998).
4. Problem Statement:
4(a): Research Objective:
The purpose of the study is:
‘’To investigate the factors influencing the employees turnover in Unilever Pakistan’’.
4(b):Research Questions:
1. To what extent unskilled labor leads to employee turnover?
2. Does work-life balance effects the employee turnover?
3. What contributions wage rate have towards employee turnover?
4. How much influence age have on employee turnover?
Page of5 24
5. Conceptual Framework:
6. Theoretical Framework:
In the above conceptual frame work the dependent variable for his study is Employee
turnover, which is the variable of primary interest. The four independent variables are
unskilled labor, age of the employee, wage rate and work-life balance. The mediating
variable is the satisfaction level of employees.
Employee turnover is defined as crossing of the membership boundary of an
organization due to voluntary resignation, termination and retirement. Unskilled labor
are the workers who do not have knowledge and experience of doing work. While, age
can be well defined as the lifespan of a person. A wage is monetary compensation paid
by an employer to an employee in exchange for work done. While work-life balance is a
concept of equally prioritizing between work; careers and ambition; and lifestyle; health,
leisure, family etc.
The more will be the unskilled labors in the organization, the higher will be the
employee turnover. The more the employees under the age of 30 in an organization the
higher would be the employee turnover rate. Wage and a positive work-life balance
holds a negative relationship with the employee turnover example; the more the wage
rate of the employee the lesser would be the employee turnover and the better the
work-life balance the lesser would be the employee turnover in the organization.
Satisfaction acts as a mediating variable, it surfaces exactly at the same time when the
independent variables starts affecting the dependent variable.
The unskilled labor whenever gets promoted to the job which is not of their capacity and
ability they leave the job, resulting in employee turnover, because they are unable to
find encouragement and motivation level while performing their work. People with age
less than 30, no matter how satisfy they become with their job as soon as they get
bored or get a better work environment they will move from existing job, which leads to
employee turnover. Due to the reason that they want to achieve more and unique from
life and their working career. If wage structure of a firm is not according to the quantity
Page of6 24
or quality of work done by employees, he/she will move out of the firm in order to
secure their career and future, leading to employee turnover. If Work-life balance of an
employee is not good like they are unable to give proper time to their families and
themselves due to the working situations than they finally choose to leave the job,
resulting in employee turnover.
7. Scope of the Study:
We will be studying the employee turnover. And the company we have chosen for our
research purpose is Unilever Pakistan situated in Karachi. We will be doing this
research for a month by reviewing literature and conducting interviews of 52% men and
42% women, who are working in the company currently. Moreover, we will also be
interviewing managers and CEO of the company. The study would investigate what are
the influential factors of employee turnover in the organization. Moreover, it would
provide development in the productivity and efficiency sector of the organization. The
study would also help to demonstrate the understandings of employee’s job satisfaction
level, loyalty and commitment level towards the organizations. The results of the study
can be generalized to all small to medium size firms facing the issue of employee
turnover.
8. Relevance of the study:
The relevance of the study is done to ensure the benefits that the research provides
through its finding. Those who would benefit from this study would be the organizations
who are striving to find a better solution to reduce employee turnover. It will actually
secure the direct cost that is due to the employee turnover for example, recruiting cost,
interviewing cost and time spend in orientations of employee. Moreover, the study would
help the company to be efficient and create efficiency in their performance and
productivity sector.
Page of7 24
9. RESEARCH DESIGN
9.RESEARCH DESIGN (explained)
9(a). PURPOSE OF STUDY:
Since the objective of our study is to investigate “the factors influencing employee
turnover”. The purpose of the study we have chosen be both descriptive and causal.
• DESCRIPTIVE:
The first half of our research objective is descriptive which help us in gathering
data in a way that elaborates persons, events or situations. Through descriptive
study we intend to gather the necessary quantitative and qualitative data
Page of8 24
which will describe the attribute of our chosen variables of cause that includes
age, wage rate, work-life balance and unskilled labor.
To describe how these variables are influencing the employee turnover we will
carry out the correlation study which will provide the theoretical basis for the
scientific or more generalized research.
• CAUSAL:
In the other half of our research objective, we intend to delineate the factors
(Age, Wage- rate, Work life Balance, Unskilled labors) with a scientific
approach so that we are able to state firmly that the described factors are the
actual causes of employee turnover.
Hypothesis Testing:
We are testing hypothesis against the factors that we described (Age, Wage rate-rate,
Work Life Balance, Unskilled labors) to scientifically hit our objective that these
factors are related with the employee turnover and the major causes of it.
• Directional hypothesis
The more the age, the less is the employee turnover
The more wage rate, the less is the employee turnover
The more unskilled labor, the more is the employee turnover
• Non-directional hypothesis
There is a positive relationship between work- life balance and employee turnover
9(b). EXTENT OF RESEARCHER INTERFERENCE:
Since the first half of our study is descriptive while the second half is causal study our
interference will be minimal to moderate.
• MINIMAL:
We intend to do a survey for collecting necessary quantitative and
qualitative data through questionnaires and interviews respectively. As the
minimal interference is there as it is a part of correlational study
(conducted in natural environment). Plus we also intend to ‘explicitly’
Page of9 24
describe the relationship between variables by correlating them without
manipulating any of our variable of interest.
• MODERATE:
After we have collected the necessary quantitative and qualitative data and described
the theoretical relationships between variables, We intend to study those factors,
that need to be studied on a scientific level as well as by manipulating those
necessary/selected independent variables of our study (Wage-rate and Work Life
Balance) and observe the effects of such manipulation on the dependent variable of
our study “Employee turnover” so that we can firmly establish a firm causal
connection between the variables of our study.
9(c). STUDY SETTING:
The study setting that we are going to use is non-contrived setting. As our study is
considered to be correlational study which will be done in natural environment. We
will carry out a field study for the correlational part of our study and field
experiments (Manipulation of some necessary Independent variables of our study) for
the causal part. Here we will be bringing alterations in four of the variables that
includes: age, wage- rate, work-life balance and unskilled labor. We will manipulate
the age of labor and work-life balance have effect on employee turnover. We will also
study how different wage rate and unskilled labor influence the employee turnover.
9(d). RESEARCH STRATEGIES:
The strategies that we are going to use during our research are:
• Survey:
We are going to carry out a survey via collecting information with the help of
questionnaires and getting them filled from the employees at Unilever. This survey
will help us in collecting all the necessary quantitative and qualitative data which
will, then, help us in explicitly and/or firmly describing the factors and their
respective relationships with the employee turnover so that we can compare them
together and come up with firm and generalized results.
• Survey instruments:
Surveys will be conducted through questionnaires. Surveys are conducted to inquire
respondents about the topic of our study in order to get desired responses pertaining
relevant to our study. Questionnaires are based on set of questions to which
Page of10 24
respondents or subjects give their answers. Questionnaires are considered to be less
expensive but time-consuming than data collected either through interviews or
observation. They are usually designed in order to gather large amount quantitative
data administered either personally, through mails or by making it available
electronically
• Observation:
The second strategy that we are going to follow to pursue our research is observation
by which we will gather and collect data on “how efficaciously the employees are
completing their assigned tasks” so that we can measure the efficiency of employees
in completing tasks (an element of unskilled labors) to check and state how skilled
they are.
9(d) MEASUREMENT AND MEASURES:
Measurement is assigning numbers or symbols to certain attributes, properties or
characteristics of an object as per the certain specified rules
• Operational definition:
We intend to study the effects of Age, Wage- rate, Work Life Balance, Unskilled labors
on employee turnover and since the factors are very nebulous to measure (except
age, which doesn’t need to be operationalized) so, we operationalized it in to
different dimensions, to reduce the abstractness to the minimum level, which will
ease the study of the effects that they have on employee turnover. The
operationalization is done below:
Page of11 24
IN THE ABOVE FIGURE, all the four variables we intend to measure are operationalized
in to multi-dimensions.
• ITEMS (MEASURE)
Items I intend to include in the questionnaire in order to measure their influence on
employee turnover are: Age, work life balance, wage-rate and unskilled labor and
since the extent of abstractness is high in each of them so I have operationalized it
(as shown in the figure above).
• SCALING:
The scales that I intend to use to measure my chosen variables of study are primary
scales (i.e. nominal and interval) and the category and Likert scale, since they’re easy
to construct and measure the attitudes nicely and precisely.
Likert scale:
We are using a 5-point Likert scale to easily and precisely measure the
attitudes and to know the degree of agreement of our respondents.
Interval scale:
Since the Likert scale uses the interval scale so we are actually using the
interval scale to measure the relation between variables of our study to firmly
state the causal connection between them, by the Likert scale in the
questionnaire.
Category Scale:
We have used the category scale to elicit a single response from respondent
which uses the Nominal scale.
Nominal Scale:
We have used the nominal scaling in obtaining the personal data of respondents
and in the category scale used in the questionnaires.
• Categorizing:
Since we are using Nominal scale in Category scale ergo, we have categorized
our items, on these scales, on the basis of Age, Work-life balance, Wage rate/
Salaries, and Unskilled labor so that the responses analyzed on the basis of
these items help in investigation of their influence on employee turnover.
Page of12 24
• Coding:
We have coded the dimensions of our variables, Age, Wage rate/Salaries, Work
life balance and Unskilled labor, as D1, D2(a), D2(b), D3(a), D3(b), D4(a), D4(b)
respectively; and these represent the hours given to work, hours given to
family/private life, salary that Unilever is offering, Salary that the competitors
are offering, Educational level of labors and efficiency in completing the tasks
respectively.
Moreover, we’ve also coded every item in our measure, with the help of the
scales that we have chosen, as 1, 2, 3, 4, 5 for the Likert Scale representing
Strongly disagree, disagree, neutral, agree, Strongly agree respectively, while
the different categories of the Category scale are assigned their own respective
subjects; like assigned 1 for the first category, 2 for the second and so on and
so forth.
9(e). UNIT OF ANALYSIS:
GROUPS:
I have chosen ‘groups’ as my unit of analysis and hence the population that I
will be studying will be groups of employee of Unilever possessing
heterogeneity across it and homogeneity within it.
9(f). SAMPLING DESIGN:
I intend to choose probability sampling as my sampling design in which I intend to
study the random subjects from strata of elements in my population (employees
within Unilever) via disproportionate stratified random sampling.
• DISPROPORTIONATE STRATIFIED RANDOM SAMPLING:
Since Unilever employees many thousands of employees therefore to increase
the efficacy of the work we intend to draw random subjects, from strata, that
are disproportionate to the number of elements in strata and that is how we
will ensure the representativeness of our sample.
We will divide the population in to different mutually exclusive groups on the
basis of the variables of our study, i.e. to answer the following questions
(related straight to our research questions), We will first divide all the
population of employees within Unilever according to the age groups,
Page of13 24
employees having high work life balance and low work life balance, wage/
salaries and finally according to ‘how skilled they are?’;
1-Are the elderly employees contributing more in employee turnover than the adults?
2-Are the employees having low work life balance causing more employee turnover
than the ones having high work life balance?
3-Are the employees getting low wages contributing more in turnover than the ones
getting high wages?
4-Are the unskilled labors influencing the company’s employee turnover rate more
than the skilled ones?
• SAMPLE SIZE:
Considering the Krejice and Morgan’s (1970) table to ensure a good decision, I intend
to study a sample of 384 employees so that the studied sample size is representative
of the population.
9(G). TIME HORIZON:
The study time would be purely longitudinal. Since I intend to carry on with the field
experiments so I will gather the information before manipulating the necessary
independent variables and after manipulating by pair testing technique to study the
effects of such manipulation which will help in firmly stating what variable is
contributing how much in employee turnover.
9(h). DATA COLLECTION METHOD:
The methods that I intend to use would be questionnaires and observation on a broad
perspective. To gather all the necessary quantitative and qualitative data for our
research.
• Questionnaire:
Following is the questionnaire that we intend to use in our research for collecting all
the necessary quantitative and qualitative data to further process it to gain the
necessary results and the theoretical basis for the field experiments on work life
balance and wage rate/salaries.
Page of14 24
“Employee turnover in Unilever Pakistan”
Dear Participant January 1, 2017
This questionnaires is designed to study the employee turnover in Unilever Pakistan.
The information you provided will help us better understand the factors of employee
turnover in organization. Because you are the one who can give us a correct picture of
how you can experience your work life, I request you to respond to the questions frankly
and honestly.
Your response will be kept strictly confidential. Only members of the research team will
have access to the information you give. In order to ensure the utmost privacy, we have
provided an identification number for each participant, this number will be used by us
only for follow-up procedures. The numbers, names, and the competed questionnaires
will not be made available to anyone other than the research team. A summary of the
results will be mailed to you after the data are analyzed.
Thank you very much for your time and cooperation. I greatly appreciate the help of
your organization and yourself in furthering this research endeavor.
Cordially,
Students Bahria University
About Yourself
Age Under 20 20-30 30-50 50-65 Over 65
Marital
status
Single Married Windowed Divorced/
Separated
Other
(Specify):
Job status Top mgt Middle mgt First Level
mgt
Non-
managerial
——
Highest
Completed
Education
High Scool College/
Diploma
Graduate Masters Other
Present
Workshift
First Second Third —— ——
No of Years
Worked
<1 1-2 3-5 5-6 Over 6
Other
organization
s. Worked
for before
this.
None One Two Three 4 or more
Page of15 24
Work-life Balance
Wage Rate
Strongly Agree Agree Neutral Disagree Strong Disagree
1 2 3 4 5
The demand of my work interfere with home
and family life 1 2 3 4 5
The working hours does not tired me
1 2 3 4 5
The amount of time my job does takes up
does not makes it difficult to fulfill family
responsibility
1 2 3 4 5
Things I want to do at home do not get done
because of the demands my job puts on me 1 2 3 4 5
My job produces strain that makes it difficult
to fulfill family duties 1 2 3 4 5
Due to work related duties, I have to make
changes to my plans for family activities 1 2 3 4 5
The job takes up makes it difficult to attend
social gatherings. 1 2 3 4 5
Your Current
main
Employment
status
Full Time Part Time Student in Full
Time
Student in Part
Time
How many paid
Jobs you
currently have?
ONE TWO THREE+ —
Total yearly
income before
taxes and
other
deduction is:
<30,000 PKR 30,000-50,000
PKR
70-90,000PKR OVER 90,000
Page of16 24
Unskilled Labor
Strongly Agree Agree Neutral Disagree Strong
Disagree
1 2 3 4 5
The job I do is of my interest
1 2 3 4 5
The job is my dream job.
1 2 3 4 5
I have to face constant failures in my
current job 1 2 3 4 5
I am able to fulfill all of my job
requirements 1 2 3 4 5
Is this your dream job
1 2 3 4 5
I understand my job sequence
1 2 3 4 5
If I get a job with better working position I
will leave this job. 1 2 3 4 5
I find difficulties in understanding my job.
1 2 3 4 5
I am to complete my task on time
1 2 3 4 5
Days you work
per week
1 Day 2 Days 3 Days 4 Days 5 Days 6 Days 7 Days
Hours worked
a day
<5hrs 5-7 7-9 More
than 9
— — —
Page of17 24
The questionnaires in the survey may not be all-embracing and comprehension and
may not therefore have afforded you an opportunity to report some things you may
want to say about your job, the organization, or yourself .Please make any additional
comments needed in the space provided.
FEEDBACK:
How did you feel as a respondent about completing this questionnaire?
I. Excited
II. Happy
III. Neutral
IV. Sad
V. Confused
VI. Angry
9(i). DATA ANALYSIS:
I intend to analyze the quantitative and qualitative data collected via questionnaires
and interviews by getting the ‘Feel for data’, testing the ‘Goodness of data’ and lastly
via ‘Testing Hypothesis’.
• FEEL FOR DATA:
Page of18 24
Since the scales that we have chosen are Likert, category, nominal and interval
therefore the chosen methods to get the feel for data are ‘descriptive statistics
(Frequencies, Measures of central tendencies, and Measures of dispersion) to check
the degree of agreement and responses over the scales, and relationship between
variables.
o Descriptive statistics:
-FREQUENCIES:
Measuring frequency will help us in getting a sense about the data to further process
it by providing us a great deal of basic information, e.g.This will indicate the number
of employees agreeing or disagreeing over the scale, number of employees belonging
to the same age group which will help in segregating them to carry out field
experiments, no of employees getting same wages/salaries, and the number of
employees who have a low or high work life balance.
We intend to visually display our results of frequencies in the form of pie charts and
bar charts.
-Measures of central tendencies:
We intend to measure the central tendency of data via mean, median and
mode.
MEAN:
We intend to take an average or find the mean of our gathered data to get a clear cut
picture of the data and to know where the actual central data is lying. This will
enable us in knowing the average age of employees in an organization and the
average degree of agreement of employees on work life balance and unskilled labor
(via questionnaire) or simply what on average is actually influencing employee
turnover. That is how calculating mean will help us in measuring the average of the
interval scales used.
MEDIAN:
Median is the central item of all the data gathered and we intend to calculate it to
know the midst values of nominal and interval scales.
This along with the mode help us in determining the skewness of the data gathered,
so we’ll be able to know whether or not the data gathered is ‘Normally distributed’
when we’ll compare the median and mode with the mean.
Page of19 24
MODE:
This will help us in calculating the highest frequencies of the variables of our study
and hence, we’ll be able to know what aged employees are contributing the most in
employee turnover? Or which employee is working more hours per day? Etcand that is
how it will help us in getting feel for the nominally scaled data.
-MEASURES OF DISPERSION:
We intend to measure the dispersion of data via standard deviation, since it’s the
most commonly used and the most reliable way of accurately calculating the
dispersion or how far is the data lying from mean.
STANDARD DEVIATION:
This will give us a more accurate figure of dispersion of data which we will help us in
manipulation of the necessary independent variables since we will know the accurate
rate of dispersion of the of the variables. This also helps in knowing how much
observations fall within one standard deviation, 2 standard deviations and 3 standard
deviations so we can compare these with the normal distribution (68%, 95%, and 99%
respectively) to get a more clearer picture of data dispersion.
-RELATIONSHIP BETWEEN VARIABLES:
Since we are doing a causal study so we need to have some tests done for the
relationship b/w the variables of our study, to firmly state a causal connection.
Therefore, we intend to study the relationship of nominally scaled data of ours via the
non-parametric test and the interval data via Pearson correlation matrix.
• NON-PARAMETRIC TEST:
The non-parametric that we are going to use is Chi Square Test (X2) to examine
the relationship between nominally scaled variables in our study. This will help
us in knowing ‘how the age is related to employee turnover?’, ‘how wage and
employee turnover are related?’
• PEARSON CORRELATION MATRIX:
To examine the relationship between the variable at an interval scale, I intend
to choose Pearson correlation matrix, so that we can know, ‘how the work life
Page of20 24
balance and employee turnover are related?’ and ‘how the unskilled labors and
employee turnover are related?’
• GOODNESS OF DATA:
After we have acquired the feel for data, we can test the reliability and validity of the
measures, to test the goodness of data or to attest to the scientific rigor.
-RELIABILITY:
We will test the stability and consistency of our data to confirm the generalization,
consistency and stability of our measures.
Since our interference will be moderate (with all the manipulation of necessary
independent variables) after gathering and analyzing data with a minimal
interference hence the “test-retest reliability” and “parallel form reliability” can be
the best correlational measures of examining reliability or the stability of our
measures across time and the error variability caused by wordings and ordering.
-TEST-RETEST RELIABILITY:
We intend to study a set of respondents (randomly selected subjects/
employees from each stratum) via questionnaire containing the above
described items to measure their influence on employee turnover with a
minimal interference but, tentatively, six months later (in which we will
analyze the gathered data) the same questionnaire will be administered to the
same respondents and that is how we will collect the scores at two different
times to check the correlation between them.
The higher will be the correlation, the better will be out test-retest reliability
or stability of our measures across time.
-PARALELL-FORM RELIABILITY:
We also intend to check the error variability of our measure by changing the
wordings and the order or sequencing of questions in one of the two
comparable sets of measures and then correlating the final responses of data
collected from both the sets.
The higher the correlation, the greater will be the parallel-form reliability or
the minimum will be the error variance caused by wording, ordering and
sequencing.
-INTERNAL CONSISTENCY TESTS:
It is an indicative of the homogeneity of the items in the measure or how the
items can be hung together as a set and are capable of independently
measuring the same concept so that the respondent attach the same meaning
Page of21 24
to every item. This can be seen by examining whether the items and the
subsets of items in the measuring instrument are highly correlated and the test
that we chose is “The Cronbach’s Coefficient Alpha”.
• CRONBACH’S COEFFICIENT ALPHA:
Since we have chosen the multipoint-scaled items like Age, Work-life balance,
Wage rata/Salaries, and Unskilled labor; which have multiple points like 1,2,3,4,5
etc respectively; therefore, the Cronbach’s reliability coefficient can be the best
way of indicating how well the items in a set are positively correlated to one
another.
The closer it is to 1, the higher will be the internal consistency reliability.
-VALIDITY:
However, the measures that we have chosen are well-validated (as per the
literature reviews) but we still want to ensure that the dimensions theorized (as
above) are efficaciously tapped by the items in our measure. Therefore, we
intend to establish the criterion related validity.
CRITERION RELATED VALIDITY:
We intend to establish this by testing for the power of the measure to
differentiate the individuals who are known to be different. This will
basically ensure the strata that we will form, for example; if the criterion
related validity is high then the strata formed will be valid since
heterogeneity across the groups or the homogeneity within the groups will
be ensured.
• HYPOTHESIS TESTING:

1. H0: There is no relationship between Age and employee turnover.
H1= OTHERWISE
i. H0: ᵨ= 0
ii. H1: ᵨ≠ 0
2. H0= There is no relationship between Wage rate and employee
turnover
H1= OTHERWISE.
Page of22 24
i. H0: ᵨ= 0
ii. H1: ᵨ≠ 0
3. H0= There is no relationship between Work-life balance and employee
turnover
H1=OTHERWISE
I. H0: ᵨ= 0
II. H1: ᵨ≠ 0
4. H0= There is no relationship between unskilled labors and employee
turnover.
H1= OTHERWISE.
• H0: ᵨ= 0
• H1: ᵨ≠ 0
5. H0= The means of the strata are not equal.
H1= The means of the strata are equal.
• H0: µage ≠ µwage ≠ µWLB ≠ µunskilled.lab
• H1: µage = µwage = µWLB = µunskilled.lab
6. H0= The manipulation of wage did not produce change in employee
turnover rate.
H1= The manipulation of wage produced a change in the employee
turnover rate.
• µD = µ1 -- µ2
• H0: µD = 0
• H1: µD ≠ 0
Page of23 24
7. H0= The manipulation of Work-life balance did not produce change in
employee turnover rate.
H1= The manipulation of Work-life balance produced a change in the
employee turnover rate.
• µD = µ1 -- µ2
• H0: µD = 0
• H1: µD ≠ 0
8. H0= The items in the measure are not related to one another
H1= The items in the measure are positively correlated to one
another.
• H0: r = 0
• H1: r ≥ 0
10. TIME FRAME OF THE STUDY:
The time frame of this study is two years excluding the external factors, because we
intend to study random subjects, from each stratum, before and after manipulation of
the necessary independent variables
11. BUDGET:
This study has a budget of ___________ RS/= since the chosen sampling method is a
tad costly and we also intend to manipulate some necessary independent variables
such as work life balance and wage rate/salaries to reach at more generalized results
Page of24 24
BIBLIOGRAPHY:
Bidisha Lahkar Das, Dr.Mukulesh Baruah, Employee Retention, volume 14, issue 2
Prof.Dr.uschi Backes-Gellner and Stephan Veen, The impact of aging and age diversity
on company performance
Boyan Jovanovic (1979), Job matching and the theory of turnover, Journal of political
economy, vol 87, No.5
Henry Ongori (2007), A review of the literature on employee turnover, African journal
of business management
Lalith Munasinghe, Barnard College, Wage growth and the theory of turnover,Journal
of labor economics, vol 18, No.2
Jonathan S.Leonard and David I.Levine, The effect of Diversity on turnover:A large
case study
Patrick L. O’Halloran, Performance pay and turnover, Journal of Economic studies, vol
39, issue.6, pg 653-674
Andrew Wiess, Efficiency wages: Models of unemployment, layoffs and wage
dispersion. Princeton University Press, 1990. 97–103.

Research proposal

  • 1.
    Page of1 24 “RESEARCHPROPOSAL” “Employee Turnover in Unilever” AUTHORED BY: SUMMAYA SHARIF
  • 2.
    Page of2 24 1.AWorking Title: 3 2.BACKGROUND: 3 3.Literature Review: 3 4. Problem Statement: 4 4(a): Research Objective: 4 4(b):Research Questions: 4 1. Conceptual Framework: 5 6. Theoretical Framework: 5 7. Scope of the Study: 6 8. Relevance of the study: 6 9. RESEARCH DESIGN 7 9.RESEARCH DESIGN (explained) 7 9(a). PURPOSE OF STUDY: 7 9(b). EXTENT OF RESEARCHER INTERFERENCE: 8 9(c). STUDY SETTING: 9 9(d). RESEARCH STRATEGIES: 9 9(d) MEASUREMENT AND MEASURES: 10 9(e). UNIT OF ANALYSIS: 12 9(f). SAMPLING DESIGN: 12 9(G). TIME HORIZON: 13 9(h). DATA COLLECTION METHOD: 13 9(i). DATA ANALYSIS: 17 • HYPOTHESIS TESTING: 21 10. TIME FRAME OF THE STUDY: 23 11. BUDGET: 23 BIBLIOGRAPHY: 24
  • 3.
    Page of3 24 1.AWorking Title: “Employee turnover in Unilever Pakistan” 2.BACKGROUND: Unilever was established in Pakistan as a result of merger between Unilever and British soap maker’s LEVER BROTHERS in 1984. Unilever is the largest fast moving (FMCG’s) in Pakistan, with an experience of 50 years. It has a variety of brands in Food Brands, Home Care and Personal Care. It is the market leader in 7 categories out of 8 categories. Unilever has employed 2200 people directly and 1000 indirectly by its suppliers, vendors and distributors. Moreover, it operates in 100 different countries across the globe. Unilever is working with its suppliers to support feasible approaches to agriculture. Unilever is also learning from NGO’s and other organizations to contribute towards the welfare of the people and the country. 3.Literature Review: Employee turnover is the rotation of workers around the labor market; between firms, jobs and occupations; and between the states of unemployment and employment Abassi et al. (2002). Mondy (2010) clearly defined turnover rates as how many new recruitments were hired to replace resigned employees. Price (1997) define employee turnover as the ratio of the total numbers of members who have left the organization during a considered period divided by the average number of employees during the period.Unskilled labor is when individual’s level of satisfaction is not involved in the issue. A person is able to do a particular work but yet he may not be satisfied. The place, environment, or reduced activity might not suit him. A large number of notions influences likes and dislikes. The worker moves for example in the state of unskilled labor until he finds work that is according to his capabilities and skills (George, 1919). Work-life balance is important for an individual psychological well-being, and that of high self-esteem, satisfaction, and overall sense of harmony in life can be regarded as indicator of a successful balance between work and family roles (Clark, 2000; Clarke et al. 2004;Mark and MacDermid,1996).Interventions of work demand into personal life (e.g: working during the week-end) resulted into a disturbing work life balance among the employees (Hyman et al. 2003).Job stress and declining work life balance (eg: work overload, job ambiguity) are the factors that triggers the chain of psychological states that leads to employee turnover (Firth et al. 2003).
  • 4.
    Page of4 24 Work-lifebalance is the ability to experience a sense of control and to stay productive and competitive at work while maintain a healthy life style. It is achieving task with time and attention (Jeff Davidson, 2014). Work life balance is to manage your work and manage your life (Groysberg, 2014). Topel and Ward (1992) finds that thosewith better economic conditions like jobs facilitating and with a wage growth rate are particularly less likely to encounter with employee turnover than jobs that offer a low wage growth rate.Campion (1991) cited in Tang suggests that the variable reason for high turnover rates in organizations is higher wages and career opportunities employee seeks for. Clark et al. (1998) argues that if performance related pay workers are more satisfied with their jobs they may experience lower turnover than traditional times rate workers. Griffeth et al. (2000) recorded pay and pay related variables have a modest effect on turnover.There is a strong evident which state that age heterogeneity increase the rate of employee turnover (Richard, 2002).People with an age of 30-45 and those over age 45 have more satisfaction with job security and desire to remain with their company as compared to those age under 30. Also, for the under-30s, satisfaction with opportunities to develop technical skills and pay linked to individual performance has a strong negative relationship with willingness to change companies than for those over 45 (finegold,2002). Determining how to structure the employment relationship may be difficult, however, since young people in today’s workforce are often describe as generation wanting very different things from work than their senior person (Conger, 1998). 4. Problem Statement: 4(a): Research Objective: The purpose of the study is: ‘’To investigate the factors influencing the employees turnover in Unilever Pakistan’’. 4(b):Research Questions: 1. To what extent unskilled labor leads to employee turnover? 2. Does work-life balance effects the employee turnover? 3. What contributions wage rate have towards employee turnover? 4. How much influence age have on employee turnover?
  • 5.
    Page of5 24 5.Conceptual Framework: 6. Theoretical Framework: In the above conceptual frame work the dependent variable for his study is Employee turnover, which is the variable of primary interest. The four independent variables are unskilled labor, age of the employee, wage rate and work-life balance. The mediating variable is the satisfaction level of employees. Employee turnover is defined as crossing of the membership boundary of an organization due to voluntary resignation, termination and retirement. Unskilled labor are the workers who do not have knowledge and experience of doing work. While, age can be well defined as the lifespan of a person. A wage is monetary compensation paid by an employer to an employee in exchange for work done. While work-life balance is a concept of equally prioritizing between work; careers and ambition; and lifestyle; health, leisure, family etc. The more will be the unskilled labors in the organization, the higher will be the employee turnover. The more the employees under the age of 30 in an organization the higher would be the employee turnover rate. Wage and a positive work-life balance holds a negative relationship with the employee turnover example; the more the wage rate of the employee the lesser would be the employee turnover and the better the work-life balance the lesser would be the employee turnover in the organization. Satisfaction acts as a mediating variable, it surfaces exactly at the same time when the independent variables starts affecting the dependent variable. The unskilled labor whenever gets promoted to the job which is not of their capacity and ability they leave the job, resulting in employee turnover, because they are unable to find encouragement and motivation level while performing their work. People with age less than 30, no matter how satisfy they become with their job as soon as they get bored or get a better work environment they will move from existing job, which leads to employee turnover. Due to the reason that they want to achieve more and unique from life and their working career. If wage structure of a firm is not according to the quantity
  • 6.
    Page of6 24 orquality of work done by employees, he/she will move out of the firm in order to secure their career and future, leading to employee turnover. If Work-life balance of an employee is not good like they are unable to give proper time to their families and themselves due to the working situations than they finally choose to leave the job, resulting in employee turnover. 7. Scope of the Study: We will be studying the employee turnover. And the company we have chosen for our research purpose is Unilever Pakistan situated in Karachi. We will be doing this research for a month by reviewing literature and conducting interviews of 52% men and 42% women, who are working in the company currently. Moreover, we will also be interviewing managers and CEO of the company. The study would investigate what are the influential factors of employee turnover in the organization. Moreover, it would provide development in the productivity and efficiency sector of the organization. The study would also help to demonstrate the understandings of employee’s job satisfaction level, loyalty and commitment level towards the organizations. The results of the study can be generalized to all small to medium size firms facing the issue of employee turnover. 8. Relevance of the study: The relevance of the study is done to ensure the benefits that the research provides through its finding. Those who would benefit from this study would be the organizations who are striving to find a better solution to reduce employee turnover. It will actually secure the direct cost that is due to the employee turnover for example, recruiting cost, interviewing cost and time spend in orientations of employee. Moreover, the study would help the company to be efficient and create efficiency in their performance and productivity sector.
  • 7.
    Page of7 24 9.RESEARCH DESIGN 9.RESEARCH DESIGN (explained) 9(a). PURPOSE OF STUDY: Since the objective of our study is to investigate “the factors influencing employee turnover”. The purpose of the study we have chosen be both descriptive and causal. • DESCRIPTIVE: The first half of our research objective is descriptive which help us in gathering data in a way that elaborates persons, events or situations. Through descriptive study we intend to gather the necessary quantitative and qualitative data
  • 8.
    Page of8 24 whichwill describe the attribute of our chosen variables of cause that includes age, wage rate, work-life balance and unskilled labor. To describe how these variables are influencing the employee turnover we will carry out the correlation study which will provide the theoretical basis for the scientific or more generalized research. • CAUSAL: In the other half of our research objective, we intend to delineate the factors (Age, Wage- rate, Work life Balance, Unskilled labors) with a scientific approach so that we are able to state firmly that the described factors are the actual causes of employee turnover. Hypothesis Testing: We are testing hypothesis against the factors that we described (Age, Wage rate-rate, Work Life Balance, Unskilled labors) to scientifically hit our objective that these factors are related with the employee turnover and the major causes of it. • Directional hypothesis The more the age, the less is the employee turnover The more wage rate, the less is the employee turnover The more unskilled labor, the more is the employee turnover • Non-directional hypothesis There is a positive relationship between work- life balance and employee turnover 9(b). EXTENT OF RESEARCHER INTERFERENCE: Since the first half of our study is descriptive while the second half is causal study our interference will be minimal to moderate. • MINIMAL: We intend to do a survey for collecting necessary quantitative and qualitative data through questionnaires and interviews respectively. As the minimal interference is there as it is a part of correlational study (conducted in natural environment). Plus we also intend to ‘explicitly’
  • 9.
    Page of9 24 describethe relationship between variables by correlating them without manipulating any of our variable of interest. • MODERATE: After we have collected the necessary quantitative and qualitative data and described the theoretical relationships between variables, We intend to study those factors, that need to be studied on a scientific level as well as by manipulating those necessary/selected independent variables of our study (Wage-rate and Work Life Balance) and observe the effects of such manipulation on the dependent variable of our study “Employee turnover” so that we can firmly establish a firm causal connection between the variables of our study. 9(c). STUDY SETTING: The study setting that we are going to use is non-contrived setting. As our study is considered to be correlational study which will be done in natural environment. We will carry out a field study for the correlational part of our study and field experiments (Manipulation of some necessary Independent variables of our study) for the causal part. Here we will be bringing alterations in four of the variables that includes: age, wage- rate, work-life balance and unskilled labor. We will manipulate the age of labor and work-life balance have effect on employee turnover. We will also study how different wage rate and unskilled labor influence the employee turnover. 9(d). RESEARCH STRATEGIES: The strategies that we are going to use during our research are: • Survey: We are going to carry out a survey via collecting information with the help of questionnaires and getting them filled from the employees at Unilever. This survey will help us in collecting all the necessary quantitative and qualitative data which will, then, help us in explicitly and/or firmly describing the factors and their respective relationships with the employee turnover so that we can compare them together and come up with firm and generalized results. • Survey instruments: Surveys will be conducted through questionnaires. Surveys are conducted to inquire respondents about the topic of our study in order to get desired responses pertaining relevant to our study. Questionnaires are based on set of questions to which
  • 10.
    Page of10 24 respondentsor subjects give their answers. Questionnaires are considered to be less expensive but time-consuming than data collected either through interviews or observation. They are usually designed in order to gather large amount quantitative data administered either personally, through mails or by making it available electronically • Observation: The second strategy that we are going to follow to pursue our research is observation by which we will gather and collect data on “how efficaciously the employees are completing their assigned tasks” so that we can measure the efficiency of employees in completing tasks (an element of unskilled labors) to check and state how skilled they are. 9(d) MEASUREMENT AND MEASURES: Measurement is assigning numbers or symbols to certain attributes, properties or characteristics of an object as per the certain specified rules • Operational definition: We intend to study the effects of Age, Wage- rate, Work Life Balance, Unskilled labors on employee turnover and since the factors are very nebulous to measure (except age, which doesn’t need to be operationalized) so, we operationalized it in to different dimensions, to reduce the abstractness to the minimum level, which will ease the study of the effects that they have on employee turnover. The operationalization is done below:
  • 11.
    Page of11 24 INTHE ABOVE FIGURE, all the four variables we intend to measure are operationalized in to multi-dimensions. • ITEMS (MEASURE) Items I intend to include in the questionnaire in order to measure their influence on employee turnover are: Age, work life balance, wage-rate and unskilled labor and since the extent of abstractness is high in each of them so I have operationalized it (as shown in the figure above). • SCALING: The scales that I intend to use to measure my chosen variables of study are primary scales (i.e. nominal and interval) and the category and Likert scale, since they’re easy to construct and measure the attitudes nicely and precisely. Likert scale: We are using a 5-point Likert scale to easily and precisely measure the attitudes and to know the degree of agreement of our respondents. Interval scale: Since the Likert scale uses the interval scale so we are actually using the interval scale to measure the relation between variables of our study to firmly state the causal connection between them, by the Likert scale in the questionnaire. Category Scale: We have used the category scale to elicit a single response from respondent which uses the Nominal scale. Nominal Scale: We have used the nominal scaling in obtaining the personal data of respondents and in the category scale used in the questionnaires. • Categorizing: Since we are using Nominal scale in Category scale ergo, we have categorized our items, on these scales, on the basis of Age, Work-life balance, Wage rate/ Salaries, and Unskilled labor so that the responses analyzed on the basis of these items help in investigation of their influence on employee turnover.
  • 12.
    Page of12 24 •Coding: We have coded the dimensions of our variables, Age, Wage rate/Salaries, Work life balance and Unskilled labor, as D1, D2(a), D2(b), D3(a), D3(b), D4(a), D4(b) respectively; and these represent the hours given to work, hours given to family/private life, salary that Unilever is offering, Salary that the competitors are offering, Educational level of labors and efficiency in completing the tasks respectively. Moreover, we’ve also coded every item in our measure, with the help of the scales that we have chosen, as 1, 2, 3, 4, 5 for the Likert Scale representing Strongly disagree, disagree, neutral, agree, Strongly agree respectively, while the different categories of the Category scale are assigned their own respective subjects; like assigned 1 for the first category, 2 for the second and so on and so forth. 9(e). UNIT OF ANALYSIS: GROUPS: I have chosen ‘groups’ as my unit of analysis and hence the population that I will be studying will be groups of employee of Unilever possessing heterogeneity across it and homogeneity within it. 9(f). SAMPLING DESIGN: I intend to choose probability sampling as my sampling design in which I intend to study the random subjects from strata of elements in my population (employees within Unilever) via disproportionate stratified random sampling. • DISPROPORTIONATE STRATIFIED RANDOM SAMPLING: Since Unilever employees many thousands of employees therefore to increase the efficacy of the work we intend to draw random subjects, from strata, that are disproportionate to the number of elements in strata and that is how we will ensure the representativeness of our sample. We will divide the population in to different mutually exclusive groups on the basis of the variables of our study, i.e. to answer the following questions (related straight to our research questions), We will first divide all the population of employees within Unilever according to the age groups,
  • 13.
    Page of13 24 employeeshaving high work life balance and low work life balance, wage/ salaries and finally according to ‘how skilled they are?’; 1-Are the elderly employees contributing more in employee turnover than the adults? 2-Are the employees having low work life balance causing more employee turnover than the ones having high work life balance? 3-Are the employees getting low wages contributing more in turnover than the ones getting high wages? 4-Are the unskilled labors influencing the company’s employee turnover rate more than the skilled ones? • SAMPLE SIZE: Considering the Krejice and Morgan’s (1970) table to ensure a good decision, I intend to study a sample of 384 employees so that the studied sample size is representative of the population. 9(G). TIME HORIZON: The study time would be purely longitudinal. Since I intend to carry on with the field experiments so I will gather the information before manipulating the necessary independent variables and after manipulating by pair testing technique to study the effects of such manipulation which will help in firmly stating what variable is contributing how much in employee turnover. 9(h). DATA COLLECTION METHOD: The methods that I intend to use would be questionnaires and observation on a broad perspective. To gather all the necessary quantitative and qualitative data for our research. • Questionnaire: Following is the questionnaire that we intend to use in our research for collecting all the necessary quantitative and qualitative data to further process it to gain the necessary results and the theoretical basis for the field experiments on work life balance and wage rate/salaries.
  • 14.
    Page of14 24 “Employeeturnover in Unilever Pakistan” Dear Participant January 1, 2017 This questionnaires is designed to study the employee turnover in Unilever Pakistan. The information you provided will help us better understand the factors of employee turnover in organization. Because you are the one who can give us a correct picture of how you can experience your work life, I request you to respond to the questions frankly and honestly. Your response will be kept strictly confidential. Only members of the research team will have access to the information you give. In order to ensure the utmost privacy, we have provided an identification number for each participant, this number will be used by us only for follow-up procedures. The numbers, names, and the competed questionnaires will not be made available to anyone other than the research team. A summary of the results will be mailed to you after the data are analyzed. Thank you very much for your time and cooperation. I greatly appreciate the help of your organization and yourself in furthering this research endeavor. Cordially, Students Bahria University About Yourself Age Under 20 20-30 30-50 50-65 Over 65 Marital status Single Married Windowed Divorced/ Separated Other (Specify): Job status Top mgt Middle mgt First Level mgt Non- managerial —— Highest Completed Education High Scool College/ Diploma Graduate Masters Other Present Workshift First Second Third —— —— No of Years Worked <1 1-2 3-5 5-6 Over 6 Other organization s. Worked for before this. None One Two Three 4 or more
  • 15.
    Page of15 24 Work-lifeBalance Wage Rate Strongly Agree Agree Neutral Disagree Strong Disagree 1 2 3 4 5 The demand of my work interfere with home and family life 1 2 3 4 5 The working hours does not tired me 1 2 3 4 5 The amount of time my job does takes up does not makes it difficult to fulfill family responsibility 1 2 3 4 5 Things I want to do at home do not get done because of the demands my job puts on me 1 2 3 4 5 My job produces strain that makes it difficult to fulfill family duties 1 2 3 4 5 Due to work related duties, I have to make changes to my plans for family activities 1 2 3 4 5 The job takes up makes it difficult to attend social gatherings. 1 2 3 4 5 Your Current main Employment status Full Time Part Time Student in Full Time Student in Part Time How many paid Jobs you currently have? ONE TWO THREE+ — Total yearly income before taxes and other deduction is: <30,000 PKR 30,000-50,000 PKR 70-90,000PKR OVER 90,000
  • 16.
    Page of16 24 UnskilledLabor Strongly Agree Agree Neutral Disagree Strong Disagree 1 2 3 4 5 The job I do is of my interest 1 2 3 4 5 The job is my dream job. 1 2 3 4 5 I have to face constant failures in my current job 1 2 3 4 5 I am able to fulfill all of my job requirements 1 2 3 4 5 Is this your dream job 1 2 3 4 5 I understand my job sequence 1 2 3 4 5 If I get a job with better working position I will leave this job. 1 2 3 4 5 I find difficulties in understanding my job. 1 2 3 4 5 I am to complete my task on time 1 2 3 4 5 Days you work per week 1 Day 2 Days 3 Days 4 Days 5 Days 6 Days 7 Days Hours worked a day <5hrs 5-7 7-9 More than 9 — — —
  • 17.
    Page of17 24 Thequestionnaires in the survey may not be all-embracing and comprehension and may not therefore have afforded you an opportunity to report some things you may want to say about your job, the organization, or yourself .Please make any additional comments needed in the space provided. FEEDBACK: How did you feel as a respondent about completing this questionnaire? I. Excited II. Happy III. Neutral IV. Sad V. Confused VI. Angry 9(i). DATA ANALYSIS: I intend to analyze the quantitative and qualitative data collected via questionnaires and interviews by getting the ‘Feel for data’, testing the ‘Goodness of data’ and lastly via ‘Testing Hypothesis’. • FEEL FOR DATA:
  • 18.
    Page of18 24 Sincethe scales that we have chosen are Likert, category, nominal and interval therefore the chosen methods to get the feel for data are ‘descriptive statistics (Frequencies, Measures of central tendencies, and Measures of dispersion) to check the degree of agreement and responses over the scales, and relationship between variables. o Descriptive statistics: -FREQUENCIES: Measuring frequency will help us in getting a sense about the data to further process it by providing us a great deal of basic information, e.g.This will indicate the number of employees agreeing or disagreeing over the scale, number of employees belonging to the same age group which will help in segregating them to carry out field experiments, no of employees getting same wages/salaries, and the number of employees who have a low or high work life balance. We intend to visually display our results of frequencies in the form of pie charts and bar charts. -Measures of central tendencies: We intend to measure the central tendency of data via mean, median and mode. MEAN: We intend to take an average or find the mean of our gathered data to get a clear cut picture of the data and to know where the actual central data is lying. This will enable us in knowing the average age of employees in an organization and the average degree of agreement of employees on work life balance and unskilled labor (via questionnaire) or simply what on average is actually influencing employee turnover. That is how calculating mean will help us in measuring the average of the interval scales used. MEDIAN: Median is the central item of all the data gathered and we intend to calculate it to know the midst values of nominal and interval scales. This along with the mode help us in determining the skewness of the data gathered, so we’ll be able to know whether or not the data gathered is ‘Normally distributed’ when we’ll compare the median and mode with the mean.
  • 19.
    Page of19 24 MODE: Thiswill help us in calculating the highest frequencies of the variables of our study and hence, we’ll be able to know what aged employees are contributing the most in employee turnover? Or which employee is working more hours per day? Etcand that is how it will help us in getting feel for the nominally scaled data. -MEASURES OF DISPERSION: We intend to measure the dispersion of data via standard deviation, since it’s the most commonly used and the most reliable way of accurately calculating the dispersion or how far is the data lying from mean. STANDARD DEVIATION: This will give us a more accurate figure of dispersion of data which we will help us in manipulation of the necessary independent variables since we will know the accurate rate of dispersion of the of the variables. This also helps in knowing how much observations fall within one standard deviation, 2 standard deviations and 3 standard deviations so we can compare these with the normal distribution (68%, 95%, and 99% respectively) to get a more clearer picture of data dispersion. -RELATIONSHIP BETWEEN VARIABLES: Since we are doing a causal study so we need to have some tests done for the relationship b/w the variables of our study, to firmly state a causal connection. Therefore, we intend to study the relationship of nominally scaled data of ours via the non-parametric test and the interval data via Pearson correlation matrix. • NON-PARAMETRIC TEST: The non-parametric that we are going to use is Chi Square Test (X2) to examine the relationship between nominally scaled variables in our study. This will help us in knowing ‘how the age is related to employee turnover?’, ‘how wage and employee turnover are related?’ • PEARSON CORRELATION MATRIX: To examine the relationship between the variable at an interval scale, I intend to choose Pearson correlation matrix, so that we can know, ‘how the work life
  • 20.
    Page of20 24 balanceand employee turnover are related?’ and ‘how the unskilled labors and employee turnover are related?’ • GOODNESS OF DATA: After we have acquired the feel for data, we can test the reliability and validity of the measures, to test the goodness of data or to attest to the scientific rigor. -RELIABILITY: We will test the stability and consistency of our data to confirm the generalization, consistency and stability of our measures. Since our interference will be moderate (with all the manipulation of necessary independent variables) after gathering and analyzing data with a minimal interference hence the “test-retest reliability” and “parallel form reliability” can be the best correlational measures of examining reliability or the stability of our measures across time and the error variability caused by wordings and ordering. -TEST-RETEST RELIABILITY: We intend to study a set of respondents (randomly selected subjects/ employees from each stratum) via questionnaire containing the above described items to measure their influence on employee turnover with a minimal interference but, tentatively, six months later (in which we will analyze the gathered data) the same questionnaire will be administered to the same respondents and that is how we will collect the scores at two different times to check the correlation between them. The higher will be the correlation, the better will be out test-retest reliability or stability of our measures across time. -PARALELL-FORM RELIABILITY: We also intend to check the error variability of our measure by changing the wordings and the order or sequencing of questions in one of the two comparable sets of measures and then correlating the final responses of data collected from both the sets. The higher the correlation, the greater will be the parallel-form reliability or the minimum will be the error variance caused by wording, ordering and sequencing. -INTERNAL CONSISTENCY TESTS: It is an indicative of the homogeneity of the items in the measure or how the items can be hung together as a set and are capable of independently measuring the same concept so that the respondent attach the same meaning
  • 21.
    Page of21 24 toevery item. This can be seen by examining whether the items and the subsets of items in the measuring instrument are highly correlated and the test that we chose is “The Cronbach’s Coefficient Alpha”. • CRONBACH’S COEFFICIENT ALPHA: Since we have chosen the multipoint-scaled items like Age, Work-life balance, Wage rata/Salaries, and Unskilled labor; which have multiple points like 1,2,3,4,5 etc respectively; therefore, the Cronbach’s reliability coefficient can be the best way of indicating how well the items in a set are positively correlated to one another. The closer it is to 1, the higher will be the internal consistency reliability. -VALIDITY: However, the measures that we have chosen are well-validated (as per the literature reviews) but we still want to ensure that the dimensions theorized (as above) are efficaciously tapped by the items in our measure. Therefore, we intend to establish the criterion related validity. CRITERION RELATED VALIDITY: We intend to establish this by testing for the power of the measure to differentiate the individuals who are known to be different. This will basically ensure the strata that we will form, for example; if the criterion related validity is high then the strata formed will be valid since heterogeneity across the groups or the homogeneity within the groups will be ensured. • HYPOTHESIS TESTING: 1. H0: There is no relationship between Age and employee turnover. H1= OTHERWISE i. H0: ᵨ= 0 ii. H1: ᵨ≠ 0 2. H0= There is no relationship between Wage rate and employee turnover H1= OTHERWISE.
  • 22.
    Page of22 24 i.H0: ᵨ= 0 ii. H1: ᵨ≠ 0 3. H0= There is no relationship between Work-life balance and employee turnover H1=OTHERWISE I. H0: ᵨ= 0 II. H1: ᵨ≠ 0 4. H0= There is no relationship between unskilled labors and employee turnover. H1= OTHERWISE. • H0: ᵨ= 0 • H1: ᵨ≠ 0 5. H0= The means of the strata are not equal. H1= The means of the strata are equal. • H0: µage ≠ µwage ≠ µWLB ≠ µunskilled.lab • H1: µage = µwage = µWLB = µunskilled.lab 6. H0= The manipulation of wage did not produce change in employee turnover rate. H1= The manipulation of wage produced a change in the employee turnover rate. • µD = µ1 -- µ2 • H0: µD = 0 • H1: µD ≠ 0
  • 23.
    Page of23 24 7.H0= The manipulation of Work-life balance did not produce change in employee turnover rate. H1= The manipulation of Work-life balance produced a change in the employee turnover rate. • µD = µ1 -- µ2 • H0: µD = 0 • H1: µD ≠ 0 8. H0= The items in the measure are not related to one another H1= The items in the measure are positively correlated to one another. • H0: r = 0 • H1: r ≥ 0 10. TIME FRAME OF THE STUDY: The time frame of this study is two years excluding the external factors, because we intend to study random subjects, from each stratum, before and after manipulation of the necessary independent variables 11. BUDGET: This study has a budget of ___________ RS/= since the chosen sampling method is a tad costly and we also intend to manipulate some necessary independent variables such as work life balance and wage rate/salaries to reach at more generalized results
  • 24.
    Page of24 24 BIBLIOGRAPHY: BidishaLahkar Das, Dr.Mukulesh Baruah, Employee Retention, volume 14, issue 2 Prof.Dr.uschi Backes-Gellner and Stephan Veen, The impact of aging and age diversity on company performance Boyan Jovanovic (1979), Job matching and the theory of turnover, Journal of political economy, vol 87, No.5 Henry Ongori (2007), A review of the literature on employee turnover, African journal of business management Lalith Munasinghe, Barnard College, Wage growth and the theory of turnover,Journal of labor economics, vol 18, No.2 Jonathan S.Leonard and David I.Levine, The effect of Diversity on turnover:A large case study Patrick L. O’Halloran, Performance pay and turnover, Journal of Economic studies, vol 39, issue.6, pg 653-674 Andrew Wiess, Efficiency wages: Models of unemployment, layoffs and wage dispersion. Princeton University Press, 1990. 97–103.