Gender Gap in Workforce Management

In order to address this issue, the current research is being conducted where the central question would be to access whether gender gap is indeed present and also lacks statistical understanding the role played by occupation in the gender gap dynamics. The given data contains information about four key variables namely gender, occupation, gift related deduction and amount of annual salary. Also, there are quantitative variables present in the form of gift related deduction and annual salary amount which are numerical in nature and measured on an interval scale (Flick

STATISTICAL MODELLING
STUDENT ID:
[Pick the date]

Section 1: Introduction
a) With regards to workforce management, a crucial concern related to issue of gender gap.
This essentially refers to the lesser salary that is given to females when compared with
males and thereby amounts to discrimination driven by gender which is not permissible
under law. Studies in Australian context have confirmed presence of gender gap which in
the long run could have implications for national growth as there is less than optimum
participation from a particular gender (Livsey, 2017). In order to address this issue, the
current research is being conducted where the central question would be to access whether
gender gap is indeed present and also understand the role played by occupation in the
gender gap dynamics.
b) Dataset one is essentially comprises of secondary data about 1000 taxpayers who are
randomly selected from the data published by ATO. The given data contains information
about four key variables namely gender, occupation, gift related deduction and amount of
annual salary. There are categorical variables present in the form of occupation and gender
which are measured using a nominal scale. Also, there are quantitative variables present in
the form of gift related deduction and annual salary amount which are numerical in nature
and measured on an interval scale (Flick, 2015).
The first five cases of the unique dataset 1 provided are highlighted below.
c) There is also a primary dataset named Dataset 2 which comprises of survey results that
have been collected from the respondents themselves. Keeping in mind the research
question, only information related to the gender and annual salary have been recorded.
Also, it was felt that occupation and deduction as gift would not serve much purpose
owing to limited sample size. The sampling technique that has been used in convenience
sampling which raises potential issues in this case as it is highly possible that the sample is
not representative of the population that is of interest. A low sample size may further
lower the reliability and thus the results of dataset 1 should be held in higher credibility as

compared to that of dataset 2 despite the latter being primary data (Eriksson and
Kovalainen, 2015).
Section 2: Descriptive Statistics
a) The distribution of gender across occupations is graphically presented as follows.
A key visible observation is that across occupations, the variability of the two genders is quite
high. For some occupations such as code 5, females tend to form 75% workforce while in
other occupations such as code 7, this representation falls to below 5%. Hence the
representation of the two genders in different occupations tends to vary in a significant
manner.
b) The distribution of gender in terms of annual salary is graphically illustrated as follows.

The apparent conclusion that one can draw from the above chart is that the average salary of
females is lower than that of males which indicates towards the existence of gender gap in the
provided Dataset 1. It is evident for lower salary levels i.e. below $ 50,000 per annum, there
is marginal majority of the females but as these salary levels jump, there is a continuous fall
in female representation indicating the concentration of lower salaries for females.
c) The distribution of gender in terms of annual salary is numerically illustrated as follows.
About 75% of the sample female population has an annual salary level not exceeding $
50,000. In contrast, the corresponding proportion for sample male population stands at about
55% and thus there is a significant gap. Further, there are only 3 females present in the
sample earning annual salary in excess of $ 150,000. In comparison, the males stand at 17 or
more than 5 times. The above salary gap needs to be critically analysed especially in wake of
differing proportion of gender distribution across occupations so as to draw accurate
conclusions.

d) Deduction as gifts and income level relationship is represented through the use of
following scatter diagram.
The relation between Sw_amt and Gift_amt lacks statistical significance which can be
concluded based on random positioning of the scatter point without any particular pattern
along with value of R2 being in the vicinity of zero. This implies that correlation coefficient
would alos be quite near to zero which would imply that no relationship exists between salary
and gift related deduction (Hair et. al., 2015).
Section 3: Inferential Statistics
a) Based on the given sample data and pivot table in excel, the four occupations with highest
median salary levels are 2,7,3 and 1. The 95% confidence interval estimates of female
representation in the above occupations are highlighted as follows.

With a confidence of 95%, one can state that proportion of females in occupation 1 would be
expected to range between 0.2572 and 0.4678.
With a confidence of 95%, one can state that proportion of females in occupation 2 would be
expected to range between 0.4541 and 0.6069.
With a confidence of 95%, one can state that proportion of females in occupation 3 would be
expected to range between 0.1119 and 0.2710.

With a confidence of 95%, one can state that proportion of females in occupation 7 would be
expected to range between 0.0000 and 0.1024.
The above estimates clearly reflect that females seem to have limited job opportunities in two
of four highest paying occupations. Of particular worry is the gross under-representation of
females is occupation code 7.
b) The required hypotheses are defined as follows.
Test Statistics – Z since approximately normal distribution
Number of tails – One (Right)
Preferred Testing Approach – P value
As p value (0.0027) less than α, hence Ho rejection (Hillier, 2016).
Conclusion: Given sample data supports that more than 80% of workforce engages in
occupation with code 7 comprises of males.

c) The required hypotheses are defined as follows.
.
Test Statistics – T since unknown population standard deviation
Number of tails – Two
Preferred Testing Approach – P value
As p value (0.00) less than α, hence Ho rejection (Flick, 2015).
Conclusion: Given sample data (i.e. dataset 1) supports gender gap presence in Australia.
d) The required hypotheses are defined as follows.
Test Statistics – T since unknown population standard deviation
Number of tails – Two
Preferred Testing Approach – P value

As p value (0.604) greater than α, hence Ho is not rejected (Hair et. al., 2015).
Conclusion: Given sample data (i.e. dataset 2) does not support gender gap presence in
Australia.
Section 4: Conclusion
a) The dataset 1 has provided evidence in favour of gender gap in Australia. However, the
precise reason for the same has not been identified from the given study. It might be
possible that this would be on account of higher representation of males across
occupations which on average offer higher salaries. Also, it has been found that no
relationship exists between the annual salary amount and the gift related deduction.
b) The underlying research raises certain questions which need to be addressed through
future research. One of these is to ascertain the precise reason of gender gap by comparing
the average salary levels of the two genders across occupations on the basis of each
occupation which would provide more accurate information. Also, research is required for
low representation of females in certain occupations with emphasis on potential entry
barriers and attempts to find viable solutions.

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.
Hillier, F. (2016) Introduction to Operations Research 6th ed. New York: McGraw Hill
Publications.
Livsey, A (2017) Australia's gender pay gap: why do women still earn less than men?
[online] Available at
https://www.theguardian.com/australia-news/datablog/2017/oct/18/australia-gender-pay-gap-
why-do-women-still-earn-less-than-men (Assessed on May 25, 2018)