Assessing credibility and expert reports

The challenges in credibility assessment and the right approach to expert evidence

Before we dive into the substance of this newsletter, a quick update on my upcoming appearances. November seems faraway - but planning is already underway for my webinar on AI and decision-making and my work on a panel at the SOAR annual conference on November 13, talking about “AI and Administrative Justice”. I’m sure I will have more to say about these events in the months ahead.

Credibility Assessments

Assessing credibility is one of the most challenging aspects of adjudicating. It is especially challenging when faced with just the word of one witness against another (often referred to as “he said/she said”). While any assessment of credibility is important - it takes on heightened importance in professional regulatory settings where someone’s livelihood is at stake.

In Outram v. College of Massage Therapists of Ontario, 2025 ONSC 4201, a massage therapist was alleged to have sexually assaulted a client. The discipline committee of the College of Massage Therapists (the “College”) found that the massage therapist had engaged in non-consensual touching, leading to a finding of sexual abuse and physical abuse. The discipline committee suspended his license to practice massage therapy.

The court started its analysis with a summary of the credibility assessment process:

[25] … it is crucial for fact finders to assess the credibility of witnesses in the context of the evidence as a whole. Importantly, a finding of a lack of credibility in one aspect of a witness’s testimony does not automatically render the entirety of their evidence as non-credible. However, where a witness’ testimony is found to contain significant negative credibility markers, that evidence, assessed in light of all the other evidence in the case, may not render clear, convincing, and cogent evidence -enough to satisfy proof on a balance of probabilities.

In most tribunal settings, the standard of proof is the balance of probabilities, which means that when making a factual finding, the decision maker must “decide whether it is more likely than not that the event occurred” (F.H. v. McDougall, 2008 SCC 53, at para. 40. The Supreme Court concluded (at para. 46) that “… evidence must always be sufficiently clear, convincing and cogent to satisfy the balance of probabilities test”.

The requirement of “clear, convincing, and cogent” evidence is particularly important in cases alleging sexual abuse, because a finding of sexual abuse is “one of the most significant and serious findings that a discipline body can make against a member” (Stefanov v College of Massage Therapists of Ontario, 2016 ONSC 848 (Div. Ct.)):

… Given the consequences of such a finding, the Panel is required to act with care and caution in assessing and weighing all the evidence. In doing so, the Panel must ensure that the evidence is of such a quality and quantity to justify a finding of sexual abuse…

In this case, the client had made additional complaints to the College’s investigator after the police declined to charge the massage therapist with her original allegations. The discipline committee reached significant negative credibility findings on these additional allegations. On the additional allegations, the committee found that her evidence was not of a “quality and quantity to justify a finding of sexual abuse” but, the court noted, failed to go further and assess her credibility as a whole in relation to all the allegations against the appellant.

Credibility assessments are not conducted in silos - the totality of the evidence must be assessed in light of inconsistencies in the testimony. When a decision maker finds significant inconsistencies in the evidence of a witness and rejects evidence on an issue, that decision-maker must then consider how the inconsistencies and rejection affect the overall credibility and reliability of that witness (Stefanov at para. 67).

The court also relied on Karkanis v. The College of Physicians and Surgeons of Ontario, 2014 ONSC 7018 (CanLII), 2014 ONSC7018 (Div, Ct.) at para 57:

It is a matter of common sense that, when a witness says one thing at a hearing, but has said something quite different on an earlier occasion, the fact that the witness has given different versions may be important in deciding whether, or how much, to believe of, and rely upon, the witness’ evidence.

The court concluded that although the discipline committee was aware that the central issue in the case was the credibility and reliability of the witnesses, it accepted the flawed evidence of the complainant: “It failed to address how a major inconsistency and implausibility affected her overall credibility and reliability” (para. 48). The court found that this was a “palpable and overriding error” and that a “correct analysis of the complainant’s credibility in this matter would have shown her to be significantly less credible and far from giving clear, convincing and cogent evidence”.

The court was convinced that the client was not credible and, rather than sending the matter back to the discipline committee for a proper credibility assessment, dismissed the case against the massage therapist “especially since the conclusion of any re-hearing should be the same disposition as set out here”.

Lessons for decision-makers