The CBC recently investigated the availability of racial demographic data across Canadian universities. No surprise to anyone, they found that race-based data is severely limited (read almost non-existent) in the post-secondary sector. In the age of Big Data and the Information Economy, it is heartening to see serious discussion and concern about the ability to make evidence-based decisions in public discourse. Coming on the heels of on-going coverage of Toronto’s Children’s Aid Society’s and the Toronto District School Board’s efforts to show the value of race-based data, this investigation suggests at least some interest to expand this work.
As part of our recently revised Comprehensive Access Strategy, we actually call for the collection of race-based data at the provincial-level. Our General Assembly took some time to reach consensus on this recommendation. It’s easy for policy-makers to take an expectation of ethical research for granted, but this is not as easy for research participants. Those who are being asked to share their information rarely assume data collectors will maintain “Respect for Persons” or have a “Concern for Welfare.” So although there was a general understanding that race-based data would largely improve our understanding of university access (and subsequent persistence), the conversations at our General Assembly centred on one concern: informed consent.
Chief Commissioner of the Ontario Human Rights Commission, Renu Mandhane, is quoted by CBC as saying that students want institutions to collect data about race, so they should not worry about a backlash from students of colour. While this may be true for some, students’ desires are not without caveats. There will always be some amount of reluctance or hesitation. As stated in the article, institutions themselves have reputational and legal concerns, some reasoning that collecting racial data could be seen as a form of discrimination in and of itself. This concern was expressed by our members as well.
To understand where some of this trepidation might come from, I think it would be useful to look at the state of existing sources of sensitive demographic data. The number of Indigenous students and students with disabilities are both tracked at high levels. These are two groups for whom this information has been inaccurate.
Universities are only recently becoming confident their ability to monitor Indigenous student access. In 2013, the Council of Ontario Universities (COU) released a report on their project to help universities develop a common Indigenous self-identification tool. Part of the work on this project was to reveal and overcome barriers to the self-identification process. The COU found both historical and social factors influence students’ decisions to self-identify.
Historically, Indigenous identity has been placed in direct opposition to formal education, in that attending and accepting a degree required individuals to renounce their legal Indian status and treaty rights (as per the Gradual Civilisation Act, 1857). In addition to identity legislation, the legacy of residential schools and historical misuse of data (for example, inconsistent government support for status and non-status first nations students and the exclusion of Metis students from the public school system) have resulted in a deep mistrust of government. These historical factors hugely influence the decision to disclose one’s First Nation, Métis, or Inuit ancestry.
On top of these, a lack of clarity about the purposes of data collection creates suspicion of these processes more generally. While the government and Indigenous communities have been working to overcome this suspicion, some other social factors need to be considered to limit their influence in race-based demographic research. The fluid and subjective nature of (ethnic/cultural/racial) identity, imprecise definitions and classifications, pressure for mixed ancestry students to choose just one category, and feeling like a minority within one’s institution may all discourage students from self-reporting their race.
In research speak, reluctance to participate in data collection creates non-response bias in the results. Simply put: if a particular part of a population does not participate in a survey, then the sample will be restricted and the results will not reflect those experiences. When looking for evidence of non-response bias, researchers might look to compare their data to a variety of existing sources. Red flags go up when considerable variability is observed between multiple sources. This is the case in tracking students with disabilities.
Table: Comparison of proportion of self-identified students with disabilities in different post-secondary data sources.
Data Source |
% of Students with Disabilities |
Strategic Mandate Agreements |
7 |
OUSA OPSSS |
18 |
CUSC first-years |
22 |
ACHA-NCHA |
33.1 |
Some of the previously mentioned social factors that complicate self-identification operate in these data sources as well. None of the sources in the table use the same classification, or categorization, for disabilities. So while our Ontario Post-Secondary Student Survey (OPSSS) lists disabilities according to functional impairment (physical, intellectual/learning, psychiatric, neurological, visual or hearing impairment), the National College Health Assessment (ACHA-NCHA or NCHA) specifies certain conditions (namely Attention Deficit and Hyperactivity Disorder (ADHD) and chronic illness). The Canadian University Survey Consortium (CUSC) First-Year University Student Survey is probably not fair to include in this comparison since results are only reported at the national level, but it is another potential source to help universities estimate the demographic makeup of their campuses. This survey also uses functional descriptions of disabilities (mobility, hearing, speech, vision impairment, learning, head injury, language disorder), but again, they are different from the OPSSS and the NCHA.
The main take away from the data table is that students do not consistently disclose their disability status. According to Statistics Canada (CANSIM Table 115-0002), people with disabilities make up about 15% of Ontario’s population. For post-secondary data sources to range so widely, from below parity at 7% to double the parity at 33%, demonstrates that somewhere along the way survey participants are either withholding parts of their identities or misunderstanding the question being asked. Perhaps the work that has been put in to encourage Indigenous students to self-identify has yet to occur among students with disabilities. As a result, these students may be reluctant to disclose because of concern over how information about their disabilities will be used and fear for stigmatization and discrimination.
So, finally, I’d like to come back to the idea of informed consent. More needs to be done to explain to students why they are being asked to disclose potentially sensitive parts of their identities to large, powerful institutions. But these institutions must also work to demonstrate that they empathize with students’ concerns, confusions, and fears about this type of data collection--not simply opt-out of data collection entirely.
It is really great to see the concern for universities’ lack of race-based data getting the attention it deserves. However, once consultation begins it will be important to come to the table informed. Criticism is important, but not so useful without empathy.
Coming up with an accurate instrument to count something as immutable as race will be hard work. Getting ethics board and administrative approval will be hard work. Getting wide enough student buy-in to conduct censuses will be hard work. All anti-racism work is difficult to accomplish. More than this, it won’t be accomplished without the creation of safe spaces for consultation and assurances of good faith.
Many institutions are still not ready to take on these challenges and if too many institutions act individually, we will end up with disparate datasets. Knowing this, the province must offer more stewardship to our sector. It is nice to see commitments to collect race-based data coming from the Premier and the Anti-racism Directorate, but I am keenly interested to see how work will begin. There is a great chasm between counting students of colour and empathizing with them. I just hope that the work will go into creating cross-cultural competency in favour of expedient data collection.