Impact of Initiatives on Employees: Statistical Analysis

The report aims to highlight the impact of certain initiatives that Margaret has undertaken. Both descriptive and inferential statistical techniques have been deployed in this regards. The confidence interval estimation for average age of employees and the social media time spent based on the sample chosen tends to indicate that the sample selected represents the population.

Business Modeling and Analysis
STUDENT NAME/ID
[Pick the date]
1

Executive Summary
The report aims to highlight the impact of certain initiatives that Margaret has undertaken. Both
descriptive and inferential statistical techniques have been deployed in this regards. The
confidence interval estimation for average age of employees and the social media time spent
based on the sample chosen tends to indicate that the sample selected represents the population.
Also, it is estimated that the social media usage is independent of gender. Further, the new
initiatives have led to increase in peer support but do not lead to any significant change in stress
level. Besides, time on social media does not influence the job satisfaction of employees but
tends to have a significant negative relation with productivity.
2

Introduction
The objective of this report is to analyse the effectiveness of the various initiatives that have been
undertaken by Margaret. Various variables such as job satisfaction, peer support have been
measured after and before these initiatives so that statistical analysis can be performed for
estimation of their respective performance. In order to measure the same, a random sample of
100 employees has been used instead of the population of 810 employees. This sample data can
be found in the Appendix section. Both descriptive and inferential statistical techniques have
been deployed.
Task 1
A random sample of 100 observations has been derived from the given population 810
observations and would be used for requisite analysis. The random sample has been represented
in the appendix. The variables of interest are highlighted below.
 Gender of employee
 Age of employee (Years)
 Time spent on social media (Hours per day)
 Productivity (Number of billable hours)
 Job satisfaction (Scale 1 to 10 where 1 represents lowest level of satisfaction and 10
represents highest level of satisfaction)
 Stress level (Scale 1 to 10 where 1 represents least stress level and 10 represents highest
stress level)
 Peer support (Scale 1 to 10 where 1 represents least peer support and 10 represents highest
peer support)
3

Task 2
The numerical summary and graphical representation of the variables is highlighted below.
Gender
It is apparent from the above shown pie chart that the distribution of gender in the sample has
been found as 45% and 55% for male and female employee respectively. It indicates that number
of female employees is higher than male employee which represents that in the population of 810
employees, the dominant portion of the employee working in the company could be female
employee assuming a representative sample (Eriksson and Kovalainen, 2015).
Age
4

Based on the above shown histogram, it can be concluded that distribution of age does not follow
normal distribution which is evident from the non-symmetric shape of the histogram and
5

presence of skew. Further, the measures of central tendency, mean, median and mode are not
same which also confirms the non-normal distribution of the variable. Further, the maximum
number of employees lies within the age group of 18-25 or older age within the age group of 53-
60 years.
Time Spent on SM
6

Based on the above shown bar chart, it can be said that maximum number of employees which is
45 employees from a sample of 100 employees spent 0 to 1 hours per day on social media.
Further, only 7 employees from the sample have spent maximum time which is 3 to 4 hours per
day on social media. Gradual decrease has been observed in the frequency of employees with
respect to the increase in the total hours spent on social media per day.
Productivity
7

It is apparent that variable productivity of employees is not normally distributed because
measures of central tendency are not coincided and also, the shape of histogram is not
symmetric. Moreover, the extent of variation in the data is significantly high which is evident
from the high value of coefficient of variance (Hillier, 2016).
Job satisfaction
8

The non-symmetric shape of bar chart indicates that variable job satisfaction does not follow
normal distribution. The maximum number of employees which is 18 are least satisfied with
their job. Also, 43 employees from the sample of 100 fall in the job satisfaction level 1 to 3
which indicates that nearly 43% of the employees are least satisfied with their job. However,
there are 14 employees from the sample who fall within the highest level of satisfaction which is
10.
9

Stress level
From the above, it can be seen that stress level of employee is not equally distributed. Maximum
number of employees of the company falls in the least stress level and only one employee from
the sample of 100 has highest level of stress. Assuming that the derived random sample is the
truly representative of the population, it can be concluded that only a minimum number of
employees fall in the highest level of stress while significant portion of the employees fall in the
least stress level.
Peer support
10

It is
apparent
from the
above
that
variable is not normally distributed which is evident from asymmetric shape. Further, high
variation has been reported in the employee in regards with the peer support.
Task 3
Based on the computed 95% confidence interval, it can be said with 95% confidence that average
age of the employee would fall within the range of 39.03 to 44.80 years. The population average
age of the employee is 42.36 years which falls within the 95% confidence interval. Therefore, it
11

can be concluded that sample is a true representative of the population of 810 employees because
the true mean of the population lies within the 95% confidence interval (Flick, 2015).
It can be said with 95% confidence that per day average time spent on social media by employee
would fall within the interval of 1.023 hours and 1.423 hours. The population average time spent
on social media comes out to be 1.17 hours per day which falls within the 95% confidence
interval. Therefore, it can be concluded that sample is a true representative of the population of
810 employees because the true mean of the time spent on social media in case of population lies
within the 95% confidence interval computed for sample of 100 employees (Hair et. al., 2015).
Task 4
Hypothesis testing is termed as inferential statistical technique to comment on the characteristics
of the population based on the results obtained from sample data. In hypothesis testing, a null
hypothesis would be compared with the alternative hypothesis based on the p value computed
with respect to the test statistics. The claim is defined in the alternative hypothesis. The rejection
would take place for null hypothesis when the p value is lower than significance level (Hillier,
2016). It cannot be said that female employees has spent more time on an average on social
media as compared with the male employees. Further, it can be said that employees have
received more support from their peers after the implementation of the new initiative. Also, the
stress level of employees has not decreased after the new initiative.
Task 5
Case 1:
Independent variable = Time spent on social media
12

Dependent variable = Productivity
Scatter plot
A strong negative relationship has been observed between the variables which is apparent from
the negative slope and high value of R square. It implies that as the employee spends more time
on social media, then the productivity of the employee would decrease. The maximum
productivity has been achieved when the employee has spent zero hours per day on social media.
Further, the value of R square is 0.7855 which indicates that 78.55% of variation in the
productivity of employee would be explained by the variation in the time spent on social media. ,
Based on regression model, it can be concluded that slope is significant and thus, cannot be
assumed to be zero. Therefore, linear negative relationship is present between time spent on
social media and productivity of the employee (Hastie, Tibshirani and Friedman, 2014).
13
0 0.5 1 1.5 2 2.5 3 3.5 4
0
5
10
15
20
25
30
f(x) = − 5.24537249398985 x + 23.4270905601496
R² = 0.785501409389044
Scatter Plot
Time on SM (Hrs per day)
Productivity

Case 2:
Independent variable = Time spent on social media
Dependent variable = Job satisfaction
 Scatter plot
0 0.5 1 1.5 2 2.5 3 3.5 4
0
2
4
6
8
10
12
f(x) = − 0.318909278868393 x + 5.34002604805605
R² = 0.0107589962919593
Scatter Plot
Time on SM
Job Satisfaction
A weak negative relationship has been observed between the variables which is apparent from
the negative slope and low value of R square. Further, the value of R square is 0.0108 which
indicates that 1.08% of variation in the job satisfaction of employee would be explained by the
variation in the time spent on social media. Based on regression model, it can be concluded that
slope is not significant and thus, can be assumed to be zero. Therefore, no linear relationship is
present between time on social media and job satisfaction of the employee (Flick, 2015).
Conclusion
14

Based on the above discussion, it can be concluded that 95% confidence interval estimated with
regards to the mean employee age and mean time spent on social media tend to capture the actual
respective population means and thereby indicate that the chosen sample is representative of the
population. Also, the peer support for employees has improved post implementation of measures
but the same cannot be said about decrease in stress level. Besides, with regards to social media
time, gender has no role. Besides, time spent on social media tends to have significant negative
impact on productivity but the same is not true for employee satisfaction.
One of the potential shortcomings of this exercise is that the sampling technique may not result
in representative sample leading to wrong conclusions. Another possible issue can be in relation
to the accuracy of the data. Besides, there can be potentially ethical implications involved since
informed consent of the employees has not been obtained and certain employees may have
participated in the survey rather reluctantly.
Appendix
15

Task 1
The respective sample of 100 employees has been drawn through simple random sampling and is
given below.
Task 3
16

The excel output for the calculation of 95% confidence interval for the two cases is highlighted
below.
Here, z value has been taken into consideration because population standard deviation is known.
(a) Average age of employees
(b) Average total time spent on SM (social media)
Task 4
The claim of the company has been checked with the help hypothesis testing which is given
below.
17

1) Claim – Female employees has spent more time on an average on social media as compared
with the male employees.
H0: μFemale = μMale
Ha: μFemale > μMale
Significance level = 5%
The one tailed p value comes out to be 0.3873 which is higher than significance level and hence,
insufficient evidence is present to reject the null hypothesis. Hence, it cannot be said that female
employees has spent more time on an average on social media as compared with the male
employees.
2) Claim – Stress level of the employees has decreased after the implementation of the new
initiative.
H0: Stress level of employees has not decreased after the new initiative.
18

Ha: Stress level of employees has decreased after the new initiative.
Significance level = 5%
The one tailed p value comes out to be 0.1087 which is higher than significance level and hence,
insufficient evidence is present to reject the null hypothesis. Hence, the conclusion can be made
that the stress level of employees has not decreased after the new initiative.
3) Claim – Employees have received more support from their peers after the implementation of
the new initiative
H0: Employees have not received more support from their peers after the implementation of the
new initiative
Ha: Employees have received more support from their peers after the implementation of the new
initiative
19

Significance level = 5%
The one tailed p value comes out to be 0.0072 which is lower than significance level and hence,
sufficient evidence is present to reject the null hypothesis and to accept the alternative
hypothesis. Hence, the conclusion can be made that Employees have received more support from
their peers after the implementation of the new initiative.
Task 5
Case 1: Independent variable = Time spent on social media
Dependent variable = Productivity
 Linear Regression Model
20

 Coefficient of correlation (R) = -0.8863
Coefficient of determination (R2) =0.7855
 Hypothesis testing
H0: β= 0 i.e. slope is not significant.
H1: β≠0 i.e. slope is significant.
From the ANOVA table highlighted above, it can be seen that the test statistic (F value) comes
out to be 358.87 and the corresponding p value (significance F) comes out to be 0.00. Assuming
a significance level of 5%, it can be said that p value is lower than significance level and hence,
sufficient evidence is present to reject the null hypothesis and to accept the alternative
hypothesis. Hence, it can be concluded that slope is significant and thus, cannot be assumed to be
zero. Therefore, linear relationship has been present between time on social media and
productivity of the employee.
Case 2: Independent variable = Time spent on social media
21

Dependent variable = Job satisfaction
 Linear Regression Model
 Coefficient of correlation (R) = -0.1037
Coefficient of determination (R2) = 0.0108
 Hypothesis testing
H0: β= 0 i.e. slope is not significant.
H1: β≠0 i.e. slope is significant.
From the ANOVA table, it can be seen that the test statistic (F value) comes out to be 1.0658 and
the corresponding p value (significance F) comes out to be 0.3044. Assuming a significance level
of 5%, it can be said that p value is higher than significance level and hence, insufficient
evidence is present to reject the null hypothesis and to accept the alternative hypothesis. Hence,
it can be concluded that slope is not significant and thus, can be assumed to be zero. Therefore,
22

no linear relationship is present between time on social media and job satisfaction of the
employee.
23

References
Eriksson, P. and Kovalainen, A. (2015) Quantitative methods in business research. 3rd ed.
London: Sage Publications.
Flick, U. (2015) Introducing research methodology: A beginner's guide to doing a research
project. 4th ed. New York: Sage Publications.
Hair, J. F., Wolfinbarger, M., Money, A. H., Samouel, P., and Page, M. J. (2015) Essentials of
business research methods. 2nd ed. New York: Routledge.
Hastie, T., Tibshirani, R. and Friedman, J. (2014) The Elements of Statistical Learning. 4th
ed. New York: Springer Publications.
Hillier, F. (2016) Introduction to Operations Research. 6th ed. New York: McGraw Hill
Publications.
24