- The University of Lethbridge Employment Equity Plan. 2005.
- Mellow, Muriel. Salary Equity Committee Report to the University of Lethbridge. 2008.
As part of University of Lethbridge’s strategic plan, and a number of signed commitments with various provincial and national entities, the Employment Equity plan of 2005 laid out the progression of a set of policies regarding workplace equity. Part of this plan involves the collection of individual-level minority status and other identifying information that allow for comparison between internal and external data. This data collection is intended to permit the analysis of whether minority groups and gender are represented proportionally to Canada’s workforce as a whole, or whether some specific imbalance exists at the University of Lethbridge. The 2008 study is more similar to reports in other universities, in that it is more focused on the statistical measurement of equity between genders.
The 2005 plan worked to compile data on staff in a comprehensive fashion. A survey was sent to allow staff to self identify, followed up by interviews of non-respondents. The collected information was used to classify employees into standardized National Occupational Classifications (NOCS). The 2008 study relied on faculty personnel files from the HR department. There were 437 observations in the data set which includes professors, lecturers, academic assistants, librarians, but not sessional lecturers, nor administrators, nor retired staff, nor staff on long term disability.
The Employment Equity Plan compared employees to standard, external occupational measurements. In essence, they compared the types of positions offered at the university to the overall share of certain underrepresented groups who work in similar jobs elsewhere. They looked at cases in which at least 3 positions existed at the University of Lethbridge, and there were at least 10% of an underrepresented groups’ workforce in the same standard category.
The 2008 report used a standard multiple regression analysis to analyse the pay differential between men and women. The study regressed base salary against gender, rank, faculty, education, years of experience, a dummy variable for term contract, years since degree, and merit. The merit variable is a pair of dummy variables to cover situations where people are ineligible or whether they have no merit awards, and one continuous variable that is the total sum of all merit awards earned during an employee’s career.
The results of the 2005 document were that two areas of occupation offered at the university of Lethbridge were not representative of the available workforce. Women in semi-professional or technician positions were found to be underrepresented and a goal was set to increase female staff in these positions by 10%.
The results of the 2008 study are unique due to consideration given by the researchers that population level inference may be conducted if the data is considered a census rather than a sample. The report illustrates a debate regarding the standard tests of significance in sample data vs. population data. They opted to conduct both levels of analysis. They caution the inference derived from the population level analysis, since there remains uncertainty as to whether the statistical model is the true data generating process. Further, since University of Lethbridge might be considered a sample of universities in Canada, and similar regression analysis is applied elsewhere in the country, for comparability the results should be considered a sample of a larger population.
In viewing the population as a sample, no statistical difference in salary was detected. However, as a full census of the population, the results indicate that men earn $98.73 more than women. The sample-view of academic assistants showed no statistical difference in gender pay, while the census-view indicated that women make more than men, by $1,517 on average. Faculty members also showed no statistically significant difference in the sample-view. In the census-view, men earned more in Arts and Science ($1,512.90), Education ($1,601.6), and Fine Arts ($2,162.50). In management, women earned more ($1,486.50).
The conclusion reached by the researchers was that in the sample-view, there is no measurable bias, and hence no needed action for the university. However, in the census-view, there are pay gaps that should be addressed, although they vary across different groups. Future monitoring was recommended.