ASSIGNMENT – BB108

BUSINESS
STATISTICS

STUDENT NAME
STUDENT ID
In this assignment, a data set has been produced depicting the employee records of a global
corporation, which is dividing the employees based on their Gender, Marital Status, Age,
Years of experience, location, departments, salaries, and then gives them a job satisfaction
score before training and after training.

The main purpose of this survey was to gauge the job satisfaction of the employees in the
company on a scale of 1 -5, 1 being extremely dissatisfied and 5 being extremely satisfied.
This can be achieved by executing statistical hypotheses which can be tested on the given
data.[ CITATION Gre08 \l 1033 ]

Basic feature of a statistical variable

A data set of one statistical variable has two features:

(1) the variable can assume different values and
(2) some values appear more often while others less often. Descriptive statistics is about to
describe these two aspects at same time, using tables, graphs and numerical measures.

Frequency tables:

We use frequency to denote the number of times a particular value appears in the data set.
We use a frequency table to summarise the data.
Job Relative Cumulative
Satisfaction Frequency Frequency Frequency
1 93 0.31 0.31
2 112 0.37 0.68
3 69 0.23 0.91
4 24 0.08 0.99
5 2 0.01 1.00
Sum 300 1.00  

In the table above we see that the Scores of Job Satisfaction can assume 5 different values
and most often the value 2 (37%), then 1 (31%) and less often 3 (23%), 4 (8%) and 4 (1%).
Using mathematical notation, the frequency table can be written as follows[ CITATION
Her08 \l 1033 ]:

p ( Job Satisfaction=1 )=0.31 p ( Job Satisfaction=2 )=0.37

p ( Job Satisfaction ≤3 )=0.91 p ( 2≤ Job Satisfaction ≤ 3 )=0.60

The last formula above reads: the proportion for the variable Job Satisfaction to assume the
value between 2 and 3 is 60%.
Bar charts Pie charts
Frequency of Job Satisfaction Score Relative Frequency of Job Satisfaction Score
120

100

80

60

40

20

0
1 2 3 4 5 1 2 3 4 5

In the charts above we see that Job Satisfaction can assume 5 different values and most
often the value 2 then 1 and less often 3, 4 and 5.

Histogram

For continuous variable, we need to group the data first and then create the frequency
table. The reason is that theoretically we may not have two identical data for continuous
variables. Then the frequency would always be one if we do not group the data.[ CITATION
Mas87 \l 1033 ] For the variable Age, we have:

Frequenc Relative Cumulative Frequenc Relative Cumulative

Age Age
y y
frequency Frequency frequency Frequency
20-25 15 0.05 0.05 20 - 30 54 0.18 0.18
26-30 39 0.13 0.18 31 - 40 78 0.26 0.44
31-35 33 0.11 0.29 41 - 50 96 0.32 0.76
36-40 45 0.15 0.44 51 - 60 72 0.24 1
41-45 42 0.14 0.58 Sum 300 1  
46-50 54 0.18 0.76
51-55 47 0.157 0.917
56-60 25 0.083 1
sum 300 1  

Age Distribution Age Distribution

60 120
50 100
40 80
30 60
20 40

10 20
0
0
20-25 26-30 31-35 36-40 41-45 46-50 51-55 56-60 20 - 30 31 - 40 41 - 50 51 - 60
A histogram differs from a bar chart in that (1) it consists of continuous bars and (2) is not
the height of the bars that representing the frequencies but the area of the bars, because
when we combine two groups into one the areas of the two bars can add together to reflect
that frequencies of the two groups add together, not the heights of the bars.

Numerical measures

Description of data of one variable is about using a few numbers to characterize the
distribution of the frequencies across all the possible values of the variable.

Measures of central location:

For given data x 1 , x 2 , x 3 ,… , x N , we have

x 1+ x 1 +… + x N 1 N
Sample Average: x́= = ∑ xi (1)
N N i =1

Sample Median:

Order the data in increasingly: data x 1 ≤ x 2 ≤ x3 ≤ … ≤ x N , median of this data set is the value
of the data element in the middle, i.e. 50% percent of the data are less than the median and
50% are larger than the median.

If N is uneven, x Median =x( N+1 )/ 2 (2a)

x N /2 + x N /2+1
if N is even x Median = . (2b)
2

Percentiles

Median of a data set is also called 50 percentile of the data set, because at least 50% of the
data are less than or equal to x Median and at least 50 percent of the data are larger or equal to
x Median . We denote 50 percentile by x ⌈ 0.5 ⌉ . We have x Median =x [ 0.5 ]. Beside the 50 percentile,
many other percentiles are also often used[ CITATION Rob11 \l 1033 ]. We call the 25
percentile lower quartile and 75 percentile the upper quartile. For the lower quartile x ⌈ 0.25⌉ ,
we have at least 25% of the data less than or equal to x ⌈ 0.25⌉ and at least 75% of the data are
larger or equal to x ⌈ 0.25⌉ . With two different percentiles we can easily know how many
percent of the data lie within these two different percentiles. E.g. between the lower
quartile x ⌈ 0.25⌉ and the upper quartile x ⌈ 0.75⌉ , there are at least 50% of the data. Formally
p ( x [ 0.25 ] ≤ x ≤ x [ 0.75 ] ) =50 %

This above equation says, 50% of the data x lie within the interval [ x [ 0.25 ] , x [ 0.75] ]. Similarly, we
have

p ( x [ 0.16 ] ≤ x ≤ x [ 0.84 ] )=68 %.

Mode: is the value of the data point with highest frequency for categorical data. For
numerical data, mode is the interval (group) with highest frequency.
Measures of variation (dispersion):
( x1 −x́ ) 2+ ( x2 −x́ ) 2+ …+ ( x N −x́ )2 1
N
Sample variance 2
sX =
N−1
= ∑ ¿¿
N −1 i=1
(3)

Sample SD SD ( x )=s X = √ s2X (4)

Range L=x MAX− x MIN (5)

A pair of two statistical variables will assume different combinations of possible values and
each with different frequencies. Descriptive statistics of two statistical variables is about
description of all the possible combinations of the values and the corresponding frequencies
as same time. i.e. the frequencies across two dimensions. [ CITATION Tho02 \l 1033 ]We use
two-way frequency tables, double bar chats and sample covariance and sample correlation
coefficients to describe two statistical variables.

Two-way tables and double bar chart are often used to describe two nominal variables.

Promotion
Yes No
Marginal
Gender Frequency
Male 75 39 114
Female 116 70 186
Marginal
Frequency 191 109 300

The two-way table above can be used to answer the question: is promotion gender biased?
To answer this question, we calculate the relative frequency with in the promotion groups:

Promotion
Yes No Marginal
relative
Gender Frequency
Male 0.393 0.358 0.38
Female 0.607 0.642 0.62
Marginal
Frequency 1.00 1.00 1.00

The relative frequencies within the promotion groups are obtained by dividing the
75
frequency in each cell by the corresponding column sum: e.g. 0.393= .
191
Looking at the group of promoted, the proportion of Male is slightly higher than the Male
proportion in the not promoted group. We may conclude that the promotion is biased
towards males. [ CITATION Vro64 \l 1033 ]

Looking at the data, this conclusion seems to be OK. However, how sure are we about this
conclusion? The results above could be due to two newly promoted male staff because in
this round of performance review, it happened that no female stuff were so outstanding.
However, usually it is expected that female stuff performs at least as well as their male
colleagues. What is the proper way to draw a conclusion from such data? We will learn this
in Inferential Statistics.

The two-way frequency table can be presented as bar chart:

Frequency of Promotion and Gender

140
120
100
80
60
40
20
0
Yes No

Graphical Presentation of relationship between two continuous numerical variables

Scatter plot is often used to describe the relationship between two numerical variables:
Years of experience vs. Age
40
35
30
25
20
15
10
5
0
15 20 25 30 35 40 45 50 55 60 65

The scatter plot is a good visual tool to look at the relationship between variables. If the two
variables move in the same direction, the data cloud would scatter around an upwards
sloping line. If the two variables move in the opposite direction the data cloud would scatter
around a downwards sloping line.

Numerical measures for two variables:

Sample covariance
∑ ( xi −x́ )( yi − ý ) (6)
s XY = i=1
N−1

If the sample covariance is positive, this implies the two variables move in the same
direction. If the sample covariance in negative, these two variables move in the opposite
direction. However, the magnitude the covariance depends on the units in which the
variables are measured. It cannot reflect the strength of the relationship. To remove the
effect of the measuring units, statisticians normalize the sample covariance by dividing it
with the sample standard deviations. This ratio is called Sample correlation coefficient
which in independent from the measuring units:
N

s ∑ ( x i−x́ )( y i− ý )
ρ XY = XY = i=1
Sample correlation coefficient
sX sY N N

(7)
√ ∑ ( x i− x́ ) ( x i− x́ ) ∑ ( y i− ý ) ( y i− ý )
i=1 i =1

The sample correlation coefficients are always between -1 and 1. If the correlation
coefficient is close to 1, the two variables have strong positive linear relation, if it is close to
-1, the two variables have a strong negative linear relation. If it is close to 0, there is no
linear relations between the two variables.

Dispositional Theory Another well-known work fulfillment hypothesis is the Dispositional

Hypothesis. It may be a exceptionally general theory that recommends that individuals have
natural miens that cause them to have tendencies toward a certain level of fulfillment, in
any case of one’s work. This approach got to be a notable explanation of work fulfillment in
light of prove that work fulfillment tends to be steady over time and over careers and
occupations. Inquire about too demonstrates that indistinguishable twins have similar levels
of work satisfaction. A noteworthy show that limited the scope of the Dispositional
Hypothesis was the Center Selfevaluations Model.

Judge contended that there are four Core Self-evaluations that decide one’s mien towards
work fulfillment: self-esteem, general self-efficacy, locus of control, and neuroticism. This
show states that higher levels of self-esteem (the esteem one places on his/her self) and
common self-efficacy (the conviction in one’s own competence) [ CITATION Hac99 \l
1033 ]

Two-Factor Hypothesis (Motivator-Hygiene Theory) Frederick Herzberg’s Two calculate

hypothesis (too known as Spark Cleanliness Hypothesis) attempts to clarify fulfillment and
inspiration within the work environment. This hypothesis states that fulfillment and
dissatisfaction are driven by distinctive components – inspiration and cleanliness
components, respectively. An employee’s inspiration to work is persistently related to work
fulfillment of a subordinate. Motivation can be seen as an internal constrain that drives
people to accomplish individual and organization objectives. Persuading variables are those
viewpoints of the work that make individuals need to perform, and give individuals with
fulfillment, for case accomplishment in work, recognition, promotion openings.

These propelling variables are considered to be inherent to the work, or the work carried
out. Cleanliness components incorporate viewpoints of the working environment such as
pay, company approaches, supervisory hones, and other working conditions. While
Hertzberg's demonstrate has invigorated much investigate, analysts have been incapable to
reliably experimentally demonstrate the demonstrate, with Hackman & Oldham proposing
that Hertzberg's original definition of the demonstrate may have been a methodological
artifact. Moreover, the theory does not consider person contrasts, alternately foreseeing all
representatives will react in an indistinguishable way to changes in motivating/hygiene
variables. At long last, the show has been criticized in that it does not indicate how
motivating/hygiene variables are to be measured. (Herzberg 2008).

Following factors were identified to be affecting Job satisfaction:

Recommendations

On the premise of over discoveries a few proposals can be given which are as follows: The
pioneers of the organization have the obligation for making a tall level of job satisfaction.
The point of authority ought to be to make strides the execution of man and machine, to
move forward quality, to extend yield, and at the same time to bring pride of workmanship
to individuals. A spurring environment is one that gives laborers a sense of pride in what
they do. Directors can progress inspiration inside their organizations by following this
process:
1. Give a positive working environment Compensate and recognition.
2. Include and increment representative engagement
3. Create the aptitudes and potential of your workforce Assess and degree work fulfillment
Conclusion

Work fulfillment could be a intensely investigated zone of request. This assignment has focused
specifically on one angle of work fulfillment, unequivocally, fulfillment with pay. The aim was to
assess the degree to which an employee work fulfillment is decided by other variables mentioned
over.

Most of the organisations don't see worker cooperation as a driver of way better worker execution;
the consider highlights that these factors are among the contributing factors towards Work
fulfillment. Changes in organizational variables, such as pay scales, worker input in policy
development, and work environment might be made in an exertion to extend organizational
commitment which in turn will lead to representative commitment and satisfaction. According to the
comes about the work fulfillment of bank officers is altogether dependent upon pay, advancement
openings, rewards, connection with boss and co-worker.

Agreeing to the comes about the work fulfillment of officers is altogether dependent upon pay,
advancement openings, rewards, connection with boss and co-worker. The main findings of this
assignment are as takes after. It is obvious that the subordinate variable satisfaction with pay has the
anticipated positive effect on work fulfillment. There's a few prove that workers who are
disappointed with their bosses are more touchy to their pay in deciding on whether they are fulfilled
with their work.

References

Green, C., & & Heywood, J. (2008). Does performance pay increase job satisfaction?
Herzberg, F. (2008). Work and the nature of man.
Maslow, A. (1987). Motivation and personality,. New York: Harper & row.
Robbins, D. (2011). “Organization Development.
Thomson, A. (2002). Strategic HR”, Strategic Human Resource Planning.
Vroom, V. (1964]). Work & Motivation,. New York.
Oldham, H. &. (1999). Nature and Cause of job Satisfaction.