a) A recruiter for a computer company would like to determine whether there are differences in sales abilities between business, arts, and science graduates. She has an initial plan to select samples of employees who are graduates from these three areas, but have otherwise similar backgrounds in education and work experience. She will consider only employees who have been with the company for at least two years. The data she will collect is the commission earned by each selected person over the previous year.
• What are the factors under study?
Factors, also known as independent variables, are the data that is assumed to be impacted in an experiment or study. In the scenario involving a study of employees for a computer company and how their different majors impacted their performance, the different fields of study by the employees would the be relevant factors being tested. The study is being conducted how the factors of differing college experiences might have created differences in how the staff performs.
• What are the relevant hypotheses?
There are two hypothesis being tested here. The first, or the null-hypothesis (H-0) would be to state that there is no difference between staff members based on their college major. In this case, H¬0: µ1 = µ2 = µ3, with each µ representing a different major. This hypothesis states that the factors of major show no meaningful difference on an employee’s performance,
The alternate hypothesis would be that there are meaningful differences between staff performance based on what their college major is. This hypothesis would be written as: “H¬1: At least one of the means is different from the others.” This states that the data on staff and performance as impacted by factors of college major have a meaningful difference, and therefore the null hypothesis has adequate information to be rejected.
• If the null hypothesis were rejected, why would a post-hoc analysis be important? What information would it give?
A post hoc test is done to ensure that the results showing that differences occurred between groups is accurate. Therefore, this should only be done when the group means have been shown to be statistically significantly different. In this situation, one is trying to determine if staff with a particular college major performs better than others. In this case, a post-hoc would be very prudent. Since such results could have a major impact in staffing and recruiting, if the results showed that one group performs better, one would want to ensure that the study is indeed correct.
Sample Solution
