Skip to main content
This website uses cookies. By continuing to use this website you are agreeing to our use of cookies as described in our privacy policy.
Bulletin | The HR Space

Statistical Correlation Not Enough to Sustain a Human Rights Complaint

Reading Time 4 minute read

Labour, Employment and Human Rights Bulletin | The HR Space

To establish prima facie discrimination, human rights complainants must show that they are a member of a protected group, experienced adverse treatment, and link the adverse treatment to their membership in a protected group.

In April of 2018, the BC Court of Appeal (BCCA) in Vancouver Area Network of Drug Users v. Downtown Vancouver Business Improvement Association, 2018 BCCA 132, (PDF) confirmed that such a link is required, and addressed the type of evidence needed to establish such link. This case is of importance to employers as it limits possible human rights complaints to those with strong fact-based evidence, rather than simply statistical correlation evidence.

By way of background, the Downtown Vancouver Business Improvement Association (DVBIA) had set up a program of Downtown Ambassadors to approach people who were loitering or sleeping in front of businesses, in alcoves of buildings and in a city park to press them to move along. The complainants (a representative complaint of the Vancouver Area Network of Drug Users (VANDU) asserted that the homeless population in the downtown core was disproportionately represented by aboriginal persons and persons with disabilities. As such, VANDU argued that the Downtown Ambassador program was discriminatory.

The initial decision of the BC Human Rights Tribunal (the Tribunal) found that there were significant numbers of protected classes in the group represented by VANDU. The Tribunal also found that they experienced adverse treatment with regard to a service, facility or accommodation customarily available to the public- i.e. access to sidewalks and city parks. However, the Tribunal also found that the evidence called by VANDU, largely statistical and opinion evidence from one expert, did not establish a link between the membership in a protected group and the adverse treatment experienced. Therefore, the complaint alleging discrimination on the basis of race, ancestry, colour and physical and mental disability was dismissed.

The appeal of VANDU to the BC Supreme Court (BCSC) was successful. That court found that the Tribunal erred in looking for a "connection or link". Rather, the BCSC considered that the Tribunal should have simply considered whether membership in a protected group was a "factor" in the adverse treatment. The BCSC found that the Tribunal erred by requiring the complainants to provide "something more", beyond the statistical and expert evidence it presented.

For employers, the concern with the BCSC decision is the "slippery slope" of claiming/proving prima facie discrimination based, in effect, on only two of the three required parts of a complaint. In this case, the statistics showed that "homeless people were significantly overrepresented with mentally and physically disabled and aboriginal people". Essentially, the BCSC said that this statistical evidence, along with denial to a public facility, was strong enough to establish a prima facie discriminatory breach. With this BCSC finding, the DVBIA would then bear the onus of demonstrating that there is a bona fide justification for the complained of actions.

The BCSC decision was appealed to the BCCA, which was clear that in order for a protected group to establish prima facie discrimination, it must show a connection or link between the adverse treatment and membership in the protected group. The BCCA decided that the BCSC decision, which draws a distinction between a "factor" and a "connection or link," was not correct The BCCA acknowledged that systemic discrimination is typically established by showing that apparently neutral criteria in laws or practice can serve as proxies for discriminatory impact. However, the BCCA also found that statistical correlation itself is not sufficient: "while often indicative of the sort of link that must be shown to permit an inference of prima facie discrimination, a statistical correlation is not, itself, a link." As an example, the BCCA pointed out that persons convicted of violent crimes in Canada are overwhelmingly male: that correlation, however, does not establish that statutes that prohibit acts of violence discriminate against men." The BCCA acknowledged that certain protected groups are overrepresented among the street homeless population; however the Court also says that the "root causes of homelessness are complex and multi-dimensional". More and better evidence would be needed to establish the link between the protected group and the adverse actions.

This case confirms that the onus is squarely on the Complainant to provide a clear and compelling explanation, supported by cogent evidence, to bridge the gap between a protected group and adverse treatment. Had the BCSC decision not been overturned, a much wider scope for claims prima facie of employment-related breaches would have been available to protected groups who would claim an adverse impact based, by and large, on their group status alone. Employers would then be required to lead evidence to justify their actions and show they had accommodated to the point of unude hardship. Fortunately, the BCCA limited the ease of establishing a prima facie case.

It is important to note that VANDU has applied for leave to appeal to the Supreme Court of Canada. It remains to be seen whether this matter will be heard by the Supreme Court.


    Sign up for updates from this team

    Receive email updates from our team