# Salary Anomaly Review F.A.Q.

New FAQs added October 4, 2016

As part of the salary settlement effective May 1, 2015 (PDF), the Vice President Academic and Provost and the President of the Faculty Association (FAUW) agreed to establish a working group to investigate possible faculty salary inequities that would:

• investigate all cases where faculty salary inequities, including but not limited to gender-based inequities, may exist and recommend how such cases should be resolved using the Faculties' existing anomaly funds;
• review the processes by which salary anomalies are currently identified and resolved in each Faculty; and
• establish a standardized university-wide process for the detection and resolution of all faculty salary anomalies that may arise in future, wherever they may occur.

The Working Group concluded its work as of May, 2016. Notices went out to affected faculty members in early August, 2016.

## Background

### When was the last University-wide faculty salary review conducted?

In 2008, the salary settlement between the University of Waterloo and FAUW resulted in the creation of a working group on salary equity for women, and this working group made several recommended salary adjustments for individual women faculty.

### Who was in the 2016 Working Group?

The Salary Anomaly Working Group’s membership consisted of six members in total, three appointed by FAUW and three appointed by the VPAP.

FAUW’s representatives were Lynne Taylor (History and co-chair of the Salary Anomaly Working Group), Benoit Charbonneau (Pure Mathematics), and Cecilia Cotton (Statistics and Actuarial Science).

The VPAP’s representatives were Jean Andrey (Dean of Environment and co-chair of the Salary Anomaly Working Group), Christiane Lemieux (Statistics and Actuarial Science), and Bill Power (Chemistry).

## Process

### Which faculty members were included in the review?

All members of the bargaining unit as of April 30, 2015.

### Does the 2016 review provide recommendations that pertain to any faculty members who are no longer employed at the University of Waterloo?

The recommendations only pertain to individuals employed before May 1, 2015 and still employed in March 2016 at the time the review was conducted. It is possible that an employee would have left the university between then and the time the corrections will be applied. Such an employee would not receive a retroactive adjustment.

No.

### What approach was taken in order to identify potential salary anomalies?

A statistical approach, regression analysis, was used. The regression model fitted the May 1, 2015 salary as a linear function of 11 factors. Data were obtained from Human Resources with input from Institutional Analysis and Planning. Details about the factors and a summary of the fitted regression model can be found on page 9 of the Salary Anomaly Working Group Final Report (PDF).

### How were salary differences across Faculties and departments considered?

Based on actual salary distribution data for May 1, 2015, the Working Group organized individuals into 14 academic groups. The 14 groups are the Faculty of Applied Health Sciences, the School of Accounting and Finance, the Department of Economics, the Department of Psychology, all other units in the Faculty of Arts, the Department of Chemical Engineering, the Department of Electrical and Computer Engineering, all other units in the Faculty of Engineering, the Faculty of Environment, the School of Computer Science, all other units in the Faculty of Mathematics, the School of Optometry and Vision Science, the School of Pharmacy, and all other units in the Faculty of Science.

## Variables

### How was starting salary incorporated in the analysis?

The starting salary was not included in the analysis. In fact, the starting salary may very well be the cause of some anomalies and including it as a factor in the regression would (wrongly) explain low salaries.

### Did the Working Group consider faculty annual performance evaluations?

Yes, the average score out of 2.0 using available data for up to six years (for the 2009 – 2014 calendar years) was one of the explanatory variables.

### Did the Working Group consider Outstanding Performance Awards (OPA)?

Yes, the number of OPAs was one of the explanatory factors.

### How were career interruptions dealt with?

The model assumes that individuals have had continuous service. Faculty members who have had several or long career interruptions that resulted in no merit scores for those years did not fit the model assumptions and were not considered for adjustments.

### What threshold values were used to identify those with potentially anomalous salaries?

Using the fitted salary from the regression model we considered the following two quantities to assess anomalies: the Absolute Difference between actual salary and the fitted salary = (Actual - Fitted); and the Proportional Difference between actual salary and fitted salary = (Actual - Fitted)/Fitted. The Working Group decided to use the following criterion to identify potential anomalies: Actual <90% Fitted AND Actual >$5000 below Fitted. Although enough factors were included in the regression analysis to have a good fit of the actual salaries, the regression model cannot explain all individual variations nor be expected to match actual salaries perfectly. A difference of ≤ 10% or up to$5000 was deemed within the range of what might be expected given data and model limitations.

## Outcomes

### How many people were found to have anomalously low salaries?

59 individuals were considered by the Working Group to have low salary anomalies and 12 other individuals were identified as meriting further investigation by their respective deans.

### What proportion of cases are Assistant Professors, Associate Professors, Professors, and Lecturers, and what are the corresponding proportions in the overall bargaining unit?

The proportions by rank are:

Rank

Proportion of anomaly cases

Proportion of bargaining unit

Assistant Professor

37%

19%

Associate Professor

32%

33%

Professor

14%

35%

Lecturer

17%

13%

### What proportion of males and females were found to have anomalously low salaries?

A proportion of 9% of female faculty members in the dataset were found to have an anomaly, while for male faculty members this proportion was 5%.

### Did the analysis reveal any systematic biases by gender?

After individual anomalies were identified, the Working Group applied the recommended adjustments to these 59 cases and then re-ran the regression model, this time including male-female as a potential explanatory variable. The value of the regression coefficient for this factor is the expected salary difference between a male and a female faculty member with fixed values for all other regression terms.

The value for this coefficient was found to be $2905 and highly statistically significant. This indicates that, even after taking account of important work-related factors, a male faculty member at University of Waterloo has, on average, a salary that is$2905 higher than a female faculty member.

### Has the Working Group made recommendations for the detection and resolution of all faculty salary anomalies that may arise in future, wherever they may occur?

Yes, the Working Group has made recommendations for the identification and resolution of future salary anomalies at both the Faculty- and University-level. Details are provided in the “Salary Anomaly Working Group Final Report.”

### Did the Working Group consider gender in any other analyses?

Yes, the Working Group took advantage of available merit ratings (for the 2009 to 2014 calendar years). More specifically, four explanatory factors (gender, rank, gender*rank, and gender + rank) were used to determine whether the overall merit ratings were explained by gender and/or rank. It was determined that, overall, neither gender nor the two-way interaction based on gender and rank were statistically significant. The Working Group thus concluded that there is no evidence of a systematic gender bias in annual merit evaluations.

### When will faculty members identified as anomalies (including female faculty) receive their adjustments?

Female faculty who were members of the bargaining unit as of April 30, 2015, received an increase of $2905 to their base salary as of September 1, 2016. This translates to a$242.08 increase to your gross monthly salary, starting with your September paycheque. The same process applies to other anomalies.

Please note the salary increase advice in myHRinfo is updated each year on May 1 to show details of the annual salary increase, so you will not see the increase reflected there until May 1, 2017.

### What about female faculty hired after April 30, 2015, who weren't included in the dataset?

This answer is from the Provost:

Some women who were hired between May 1, 2015, and 30 June, 2016, have asked whether the salary anomaly correction would be paid to them. The deans appreciate that many new faculty members have been hired since May 1, 2015, and these individuals were not included in the review. Therefore the deans are committed to reviewing all such cases during the annual performance reviews with a view toward equity.

FAUW has asked that we be told of numbers (if any) receiving any adjustment and by faculty and we have been advised that we will receive this.