Final Project

To: Salvador Apari, Director of Maintenance, Montclair State University Comment by William Colucci: Standard memo heading. You are NOT writing to the instructor. You are writing to a decision maker who has requested that you gather and analyze data to inform a decision.

From: Luke Skywalker, Institutional Research senior analyst

CC: Susan Cole, President, Montclair State University

Subject: Evidence of gender bias? Analysis of salary/gender data

Date: April 30, 2019

During our April 26th meeting, it was determined that an analysis of salary differences by gender in our department was needed to determine if there is any evidence of gender bias. The concern is raised by the opinion of some supervisors that women feel they are unfairly treated. This might mean that the department will not be able to retain or attract quality female talent, and that the University could be at risk of lawsuits for discrimination. Comment by William Colucci [2]: Why are we doing the analysis? Typically, in business, the analyst will be briefed on the decision in question and asked to gather/analyze relevant data to help inform the decision.

The data set contains salary, gender, level, and length of employment gathered from HR records in this past month. This includes all employees, 67 men and 53 women. Since the data includes all employees, is recent and derived from reliable HR records, it is relevant and valid. Comment by William Colucci [2]: Describe the data in detail – what data was gathered, when, and how? Why is it reliable and valid (relevant)?

In order to evaluate the data, we discussed ways of evaluating any differences in salary that might be found. In looking at data available on departments at MSU, or comparable universities, as well as comparable industry wage gaps, it was determined that a wage gap of 8% was common. It was determined that anything above that would be cause for concern, and a gap greater than 25% of that (10% wage gap) would lead to investigation. Men and women should have similar salary variance, which indicates a wider or lesser range of opportunities in the organization. Research on the wage gap issue suggests that explanations other than gender bias include differences in choices men and women make in terms of jobs, number of hours worked, length of employment. Men and women may tend to make different life choices which impact overall career success. Thus, differences between men and women in terms of level and number of years of work experience were looked at. Comment by William Colucci [2]: Your case study provides you with information about what empirical benchmarks, i.e., how to evaluate empirical findings (the numbers, differences in average, standard deviation, as well as patterns in distribution. This is done through investigation of comparisons that can be used to judge the meaning of outcomes, i.e., how the results will be interpreted in terms of making the decision.

The first analysis was to determine whether there was an overall wage gap. Table 1 (reported in the appendix) reports the average salary by gender. Men on average earn $5785 more than women. Thus, on average, women earn 90% of men. The percent difference is 11%, which, given the benchmarks discussed above, warrants the investigation reported below. Comment by William Colucci [2]: Compare measures of central tendency (average) – what is the raw figure, what is the proportion? Speculate on what this means in terms of the benchmarks discussed above and in real world terms.

Variance or the range of salaries was examined, as this might provide insight into the differences in average salary. Table 1 data (standard deviation, reported in the appendix) reports that women have a lower minimum, and men a greater maximum salary. Men have a greater range of salaries, an 18% difference. The variance in salaries is greater for men, a 9% difference. Men clearly have access to a greater range of positions in the company, especially towards the higher salary levels. Comment by William Colucci [2]: Compare measures of variance (standard deviation) – what is the raw difference? What is the proportion of the difference, i.e., the percent difference? Discuss what this means in terms of the benchmarks laid out in the case study. Exactly how does this inform the reader on how to make his or her decision? Use practical language: What should the decision maker think, feel, do?

Chart 1 in the appendix addresses an explanation for the wage gap in terms of levels of employment, i.e., that males might have a higher salary because they are more often employed at higher paying levels in the organization. It is clear that women dominate at lower ends of the hierarchical ladder, and males towards the higher end, suggesting that the gap might be due simply to which gender holds which position. Chart 2 looks directly at average salaries within levels, and shows very similar salary levels. It also shows that at the highest level there are no female employees. This also might provide an explanation for the wage gap, i.e., that the highest paid positions are held only by men. Comment by William Colucci [2]: Introduce charts or tables before the reader sees them, or reference where they are located. Explain to the reader what data the figure summarizes and how to read the chart/table. Discuss distribution and how that might inform the decision at hand. Reference the benchmarks provided in the case study description. Make any other comments about insights you can provide or speculations about the meaning of the data for the practical question at hand. Tell the reader exactly what to think, feel or do in practical terms.

While there is a substantial pay gap in raw average salary, and in terms of the benchmarks and criterion agreed upon, yet additional analysis suggests this gap is not due to a gender bias as much as simply greater male presence in higher levels in the organization. While that fact in turn might be highlighted as evidence of bias, this analysis does not provide any evidence on that matter. It seems clear there is clear empirical evidence to explain differences in average salary in terms other than mere gender prejudice/bias in this case. Comment by William Colucci [2]: Conclusion – review the basic findings; suggest exactly what the reader should think, fell, do.

Please do not hesitate to contact me with any comments or questions regarding the above information and analysis. I would also be happy to conduct additional research upon your request. Comment by William Colucci: Closing making yourself available for questions and suggesting further analyses. What other variables or factors might provide further insight?

Appendix

Table 1

Chart 1

Chart 2

Page 2 of 2