Description

There are many studies that show women earn less than men. This, however, remains a very hot topic across industries with many people criticizing each study for ignoring factors that could be the reason behind gender pay gap (e.g. women working part time, women being less represented in high-paying jobs in the tech industry, etc.). 

A large research university wants to make sure that their graduate students who have “assistant” positions are treated fairly without a gender bias. The stipend (known as “assistantship”) the teaching and research assistants receive consists of a tuition waiver plus some cash amount, available to the graduate student for up to six years. 

A study has been initiated and data on 1,000 graduate student assistants have been compiled. For each student, the following information was recorded:  

·         Assistantship (Stipend): Annual stipend in dollars; 

·         Gender: F, female; or M, male; 

·         Field of Study: S, science, including the natural sciences and engineering; A, arts, including both liberal and fine arts, and humanities; 

·         Year of Study: year of the student’s graduate studies at the beginning of their contract period;  

·         Marital Status: M, married; or S, single 

At first, an analysis was done to see if the there was a gender gap in pay. The following table is the result of this analysis: 

Some have also expressed a concern that the first analysis is too simplistic since a pay gap could be due to the field of study. So, an analysis on the field of study was also conducted which has resulted in the following:

However, there is still much uncertainty about any conclusion one can make about the issue of gender gap in student compensation based on the results of studies that only focus on one of the characteristics of the student body at a time and do not consider the impact of other variables.  

You have been asked to help in bringing more clarity to this question.  

·         Use data in the Excel file to analyze the impact of the explanatory variables on the Assistantship (Stipend).  

·         As you move forward with your analysis, you should include the Excel output obtained at each step and refer to it in your narrative explanation.  

·         If you eliminate any of the variables from consideration, you should clearly explain why you have done so.  

·         In short, you should back all your findings with the analysis that has been conducted to reach each conclusion.  

Does your analysis contradict or confirm the original belief about role of gender and pay? Why? What is the advantage (if any) of using regression analysis compared to the two-sample t-tests (shown above)? 

Should the University be concerned about gender pay gap in compensation of its graduate assistants? What avenues do you recommend it should explore to further study and/or address this issue? 

Insert your response to the questions posed in the below, including output from your Excel-based analysis, here: