Available Studies
- Ornstein, Michael. Salary Anomalies Among Faculty Members and Librarians at the University of Windsor.June 2002: Windsor Anomalies Report 2002 03aug04.pdf (uwindsor.ca).
The 2002 report is a two part report in which the first part is a regression analysis of gender pay equity and the second part is a survey of staff to obtain other variables that may influence pay equity, such as aboriginal status, disability, etc. The regression analysis presented is a follow up from a previous study conducted by the same author one year prior. This summary will focus on the gender equity portion of the report. In addition to identifying individual salary anomalies, the analysis investigated the impact on salaries, of three structural factors affecting groups of faculty members: unusual variation within some faculties or groups, whether a government budget constraints played a role, whether older women in faculty were more likely to be disadvantaged.
The data is not described in the report; however, the summary tables indicate that the sample size of the faculty studied was 433. There are mentions of librarians as well, which have a population of 22.
A multiple regression analysis was used to estimate a predicted salary by which staff would be individually compared. Anomalies were defined as the difference between the predicted and actual salaries. Statistically insignificant variables were dropped from the equation. Of interest, the gender variable was found to be statistically insignificant and was not included in the analysis. The regression of salary was calculated using rank, age when hired, experience, merit pay, education, rank at hire, proxies for career progression, whether administrative positions were held, and faculty.
With respect to gender equity, the gender variable was omitted due to lack of statistical significance. This indicates that no gender bias was detected in the average pay. Further, there was no significant gender variation in salaries anywhere in the age range when analysis was conducted on the potential for bias to be concentrated in older female faculty members.