Business Analytics Discussion: Hypothesis Testing and Salary Analysis

This discussion post analyzes a scenario where a company's HR department is concerned about potential salary discrepancies between male and female employees. The student proposes using hypothesis testing to determine if a significant difference exists in salaries. The student outlines the key question, sets up the null and alternative hypotheses, and suggests using a two-sample t-test to compare the means of male and female employee salaries. The student describes the data collection process, including the use of questionnaires and simple random sampling, and explains how to interpret the results and communicate them to management. The post emphasizes the importance of hypothesis testing in making informed decisions regarding salary adjustments or other HR-related issues. The student also provides relevant references to support the analysis.

Running head: BUSINESS ANALYTICS DISCUSSION 1
Business Analytics Discussion: Hypothesis Testing
By
(Name of Student)
(Institutional Affiliation)
(Date of Submission)

BUSINESS ANALYTICS DISCUSSION 2
a. Describe the key question
Over the past six months within this reputable company, there has been great concern about the
salary increment based on the workload performed by the employees. There has been claims and
disputes by the female employees about their salary adjustment and as a result of this, the Human
Resource manager has raised a ticket to address this claim. Most the female employees over the
last six months have been receiving fewer amounts in their salaries as compared to their male
counterparts. But the main cause of this according to manager is based upon the commission and
bonus program that the workers are granted due to overtime work within the company. The
manager is also not sure whether this may be due to the silent discrimination taking place within
the company and that the solution may be resolved through the adjustment of the whole pay scale
for all the employees. Therefore, to verify this claim it would be important to conduct hypothesis
testing to verify whether there exist significant differences between the salaries of male and
female employees within the company.
b. How would you set up the hypothesis test?
In regard to the problem under consideration, the HR is concerned about where there exists a
significant difference between the salaries of female and male employees within the company.
Thus to determine the truth in this case, hypothesis testing is conducted about the means of the
employees’ salaries. The null and alternative hypotheses for this test are formulated as follows;
Null Hypothesis: There are no significant differences in salaries of female and male
workers within the company.
Alternate Hypothesis: There is a significant difference in the salaries of female and male
workers within the company.

BUSINESS ANALYTICS DISCUSSION 3
That is;
H0: μ1 = μ2
H1: μ1 ≠ μ2
In this test, two-sample t-test would be appropriate in performing the hypothesis test since
the assumption is that the sampling method for each sample obtained is simple random in
nature and that the samples obtained are independent (Lehmann & Romano, 2016). The two-
sample test will help us to determine whether the difference between the means found in the
sample collected from the participants (employees) is significantly different from the
hypothesized difference between the means (Shaffer, 2015).
To test this, the level of significance chosen would be 0.05. This would yield better results as
far as hypothesis testing is concerned sine the sample size to be chosen would be medium-
sized. i.e. n = 100. (Wellek, 2017)
c. Describe the data you would use
To obtain the data, simple random sampling would be applied to collect the data from a sample
of 100 employees within the company. The data would be collected by the use of open-ended
questionnaires issued to the participants with questions regarding their pay and commissions.
The employees’ responses would then be recorded and subjected to analysis. The mean averages
of both the male and female employees we separately are calculated for the purposes of
hypothesis testing.
d. How would you explain to the management the significance of this test?

BUSINESS ANALYTICS DISCUSSION 4
After the hypothesis testing is conducted, either the null or the alternative hypothesis would be
rejected based on the calculated t-value and critical p-value comparison about the test statistics
used (Wilcox, 2018). The management would then made a decision to adjust or fail to adjust the
salary scale of all the workers within the company regardless of their gender differences. The
management will further be moved with the right decisions to perform many other problems
arising within the company using hypothesis testing.

BUSINESS ANALYTICS DISCUSSION 5
References
Lehmann, E. L., & Romano, J. P. (2016). Testing statistical hypotheses. Springer Science &
Business Media.
Shaffer, J. P. (2015). Multiple hypothesis testing. Annual review of psychology, 46(1), 561-584.
Wellek, S. (2017). Testing statistical hypotheses of equivalence and noninferiority. Chapman and
Hall/CRC.
Wilcox, R. R. (2018). Introduction to robust estimation and hypothesis testing. Academic Press.