Foundation Skills in Data Analysis: SuperPart Employee Survey Report

This report presents an analysis of employee survey data collected from SuperPart. The analysis, conducted using statistical tools, reveals key insights into the workforce demographics, including age distribution and the absence of young employees, the correlation between age, occupation, weekly work hours and income. The analysis shows a high overall satisfaction level among employees, and the relation with unionisation and weekly work hours. The report also assesses potential attrition risks and makes recommendations for the company to address these concerns, including incentives and manpower planning. Furthermore, the report includes statistical estimations on weekly working hours and unionization levels, offering a comprehensive overview of the employee survey data and its implications for the organization.

Foundation Skills in Data Analysis
[Type the document subtitle]
STUDENT ID:
[Pick the date]

Foundation Skills in Data Analysis
MEMORANDUM
Date: 21st September 2017
To: Hannah Zhou, Vice-President of Human Resources
From: Sandy Stedwell, Research and Analysis Department
Subject: Re:Analysis of SuperPart Employee survey Data
Dear Hannah,
In line with the tour request as per the memorandum on August 24, the data analysis of the
sample data which our team had collected is completed. In this analysis, a host of statistical
tools and techniques have been highlighted in order to draw reliable conclusions for the
organisation as a while. The first aspect which has been observed is the high concentration of
employees in the middle ages especially ranging from 30 years to 50 years. A worrisome
aspect is that the proportion of young employees especially in age group of 18-25 years
seems lacking. Also the underlying relation of occupation, weekly work hours, before tax
income with age has been explored but is found to be without any statistical significance. A
positive aspect for the company is in the form of high satisfaction level which has been
observed. Also, the satisfaction level does not tend to vary in a statistically significant manner
with unionisation. Further, the hours of work per week is not related to the underlying
satisfaction level of employees. Like the industry, going ahead, it is expected that the
company also would face attrition concerns for which suitable measures need to be
introduced by the company. The claim in relation of non-unionised workers giving weekly
hours in excess of 45 hours has not been found true. Besides, statistical support has not been
found for the unionisation level to be lesser than the national average level. Further, it is
estimated that the overall sample size which equals 400 employees should suffice for future
survey requirement to be fulfilled.

Foundation Skills in Data Analysis
The detailed findings of the statistical analysis in relation to the various issues raised are
shared below.
1) Based on the statistical analysis highlighted in the attached excel, it is apparent that the
mean employee age for the sample is 39.41 years. But since the various age groups
belonging to different ages is essential for your analysis, usage of quartiles is preferred. As
per this, about one-fourth or 25% of employees have been found to be less than 32 years in
age. This statistic along with the high mean employee age hints at the inability of the
company at attracting young aged employees between 18-25 age group. This is also
supported by about one –fourth of employees aged between 32 and 38 years. Also, the
distribution of age has a positive skew which further hints at the presence of exceptionally
old employees but no such indication occurs for very young which is a concerning factor.
Also, it is noticeable that half of the sample employees have their ages between 32 years
and 46 years and most of these would be found in the middle management causing some
crowding. Considering that three-fourths employees in the sample set are still less than 46
years, hence retirement should not a major concern for the company. Clearly, many of the
employees are still pretty far from their retirement age and hence managing and nurturing
the existing talent would be more pivotal. However, the young employees must also join
the organisation and it needs to be a key focus are otherwise the career progression would
typically be difficult especially on account of similar age levels of a host of employees.
Also, it seems quite strange that there is so much under-representation of the young people
and clearly relevant steps need to be taken in this regard as this should be a priority area
moving ahead.
2) The key concern is to highlight the relation of age with some particular variables.
(a) The association between the occupation and the employee age is not very significant.
This is quite evident from the age distribution of those workers who are working as
labourers. It may be expected that considering the strenuous nature of the job, it
should be dominated by young people but this is not the case as contribution of
sample employees in 18-31 age bracket and 44-56 age bracket is almost the same.
Also, the representation the age group 44-56 has a very low representation in terms of
roles related to management which is surprising. Therefore, the above evidence

Foundation Skills in Data Analysis
clearly reflects absence of any clear trend with regards to age and thus occupation
seems to be determined by the underlying skill rather than age.
(b) A weak but negative association is observed between weekly hour worked and
employee age. As a result, it indicates that the hours that the employee work is
independent of the age to a large extent as other factors would be considered more
pivotal such as the workload, manpower availability, cost considerations and the
underlying productivity level.
(c) There is a weak positive association observed between the variables of interest i.e. the
employee age and the income before tax. This is quite expected since occupation and
age were not related and hence senior management are not manned by employees
higher in age. However, it is noteworthy that experience does matter and leads to
higher salary with indifference to the underlying profession to which the employee
belongs. Hence, to an extent there is an expectation that the age would have a positive
influence on the salary levels of the employees. Having said that there are other more
significant factors which drive salary and thus reduce age to a relatively insignificant
variable.
3) (a) The employees at the company seem to have a very high satisfaction level. The given
sample represents that about 46% employees display a high satisfaction and 43% display a
moderate level of satisfaction with the company. Thus almost 90% of the workforce of the
company belongs to satisfied category which is a big positive for the company as it would
enable higher employee productivity and reduce attrition related costs. The employees
who are highly dissatisfied with the company tend to comprise around 4% of the total
sample strength. But, the company needs to make proactive efforts in order to resolve
certain genuine issues that these employees may have so that the morale of the workforce
remains high.
(b) On the basis of the relevant statistical tool, it has come to light that job satisfaction is
not significantly driven by the employee unionisation. This is very positive for the
company since if an employee is a member of union, it does not imply that he/she would
be more satisfied. This clearly highlights that management is fair in dealing with
employees which is why the employees do not necessarily have to join the union to put

Foundation Skills in Data Analysis
their point forward. The management is cognisant of the benefits of a satisfied workforce
and thus aims towards the same.
c) It does not seem that there are any significant difference in the satisfaction level of
employees who tend to have different weekly work hours. This is testimony to the fact that
higher satisfaction level is not derived from fewer amounts of work hours but from other
aspects related to the organisational culture and management policies as a result of which
the employees are satisfied even after working for long hours. This is clearly a big
positive for the company which it can leverage for creation of a competitive edge over
peers.
4. Based on the information provided, attrition may be a grave concern for the industry going
ahead with an estimated 10% attrition rate. In wake of this concern, the objective is to analyse
the likely chances for the company to also face such a concern arising. Based on the current
employee strength of 9800 employees, more than 980 employees leaving the company would
amount to an attrition rate in excess of 10%. Based on the computations in excel, it may be
concluded that there seems to be a 25% chance that more than 1,000 employee would leave
the company. Clearly this poses a significant risk to the business operations in the form of
loss of talent. Hence, it is essential that the company must introduce incentives particularly in
form of share options with lock in period which would be available over a long time frame so
that there is lesser attrition. Additionally to make up for the loss of talent, the manpower
planning must be performed by the HR so that there is no shortage of trained manpower.
Also, the key reasons behind employees leaving the organisation would also need to be
explored so that appropriate corrective measure can be taken to the extent possible to address
the concerns of the employees who wish to leave
5. a) In line with the sample employee information, estimates for the organisation have been
obtained using the inferential statistics. It can be concluded with 95% certainty that mean
weekly working hours for all employees would occupy a range of 44.45-46.42 hours.
However, in relation to the non-unionised employees, the corresponding range works out as
44.44 – 46.54 hours with 95% certainty. In relation to the unionised employees, weekly
working hour range works out to be 42.34 – 47.95 hours with 95% certainty. It is apparent
that the range of unionised employees is higher than their non-unionised counterparts which
may be attributed to a larger value of variance for these employees.

Foundation Skills in Data Analysis
b) The given sample data has been used to estimate with 95% certainty the weekly hours
worked on an average for different values. The non-unionised employee proportion who
tends to clock more than 40 weekly hours is expected to range in 40.55%-51.29% with 95%
certainty. Also, the non-unionised employee proportion who tends to clock more than 50
weekly hours is expected to range in 15.36% - 23.92% with 95% certainty The above
proportions clearly indicate that more than 55% of the employees who are not part of the
union tend to work in excess of 40 hours on a weekly basis.
c) The given claim has been examined through the lens of inferential statistics by considering
the available sample employee data. The claim about non-unionised employee given weekly
hours in excess of 45 hours has been found to be statistically insignificant. Therefore, the
sample data does not provide enough statistical evidence for the claim to be supported and
hence it would be rejected in light of the available evidence.
d) Using the provided sample data, the key aim was to determine unionisation level in the
company and thereby comment on the senior management claim of unionisation being lower
than the national average in case of the company. But the statistical evidence in this regards
does not provide sufficient evidence to support the claim made by the management. Hence, it
would be correct to conclude that unionisation levels in the company are comparable to the
national average and cannot be concluded to be significantly less.
6. The least amount of sample size that is required is primarily driven by three main aspects
namely, the dispersion in the underlying quantity to be measured, margin of error desires
along with the confidence level. As per the sample information available, for weekly hours
estimation to be within the desired limits, it is estimated that atleast 388 employee sample
data is required. However, for meeting the determination of relevant proportion, the sample
requirement as computed through the same formula is must higher, Therefore, based on the
computations, it would be fair to estimate a 400 employee sample size may not suffice for the
future research considering the desired accuracy in estimation.