- Joint Womens’ Pay Equity Committee Report. University of Winnipeg. 1999.
The purpose of the 1999 study was to make recommendations, if any, as to the adjustment of faculty members’ base salary. Gender equity issues were called for by a collective agreement between the UWFA (the faculty association) and the Board of Regents.
The study used an administrative database of members of the UWFA. There were 264 total observations, 89 of which corresponded to female faculty. The database was compiled from HR files, and each faculty member was asked to verify his or her information.
A primary concern of the study was whether certain variables would taint the multiple regression analysis of salary. The first stage of the study was to determine the extent by which potential explanatory variables exhibited a gender bias. A t-test based on multi-logit regression is used in this analysis. Then multiple regression is used on salary to determine gender-based differences in pay. Two final multiple regressions were used for determining pay differences. One regression estimated salary in levels and the other used the log salary against gender, age, years since highest degree, years since appointment , education, and a series of dummy variables that indicate position, rank, or tenure.
The first set of regressions did not find evidence that women were discriminated against with respect to promotion. However. some evidence of bias against women was discovered in years of experience, which may reflect past discrimination. A result of this finding was that years of experience was substituted with years since appointment in the final regression.
In levels a $1,346.20 (2%) disadvantage in average salary was found for women. However, for the purposes of the pay adjustment, the study imputed the change needed by adjusting female salaries until there was no longer a detectable difference in pay. This method suggested a 1.8% adjustment should be applied to female staff. Further, the study suggested that the non-linear regression (in logs) provided a better fit for the data.
Ultimately, the study recommended a 1.8% base salary adjustment for all women in the faculty. Further, it was recommended that the methods developed in the study be used again in 2004 to determine if pay equity gaps persist.