No matter how fair-minded an investigator may be, the inescapable reality is that we all have inherent biases. The situations we investigate are viewed through our own lens, and sometimes our past experiences and our perceptions can interfere with a fair and neutral information gathering process unless steps are taken to mitigate the risk. The important first step is to be aware of the different kinds of bias, and how they can operate to influence our decision making.

1. Confirmation Bias

This is the type of bias most people are familiar with. My colleague, Cory Boyd, provided a helpful overview of case law regarding confirmation bias in a previous blog.

Confirmation bias occurs when a person searches for, interprets, and recalls information in a way that substantiates their pre-existing beliefs. It’s the kind of bias that is at play when psychics and tarot-card readers do a “cold reading”. If the subject already believes that the psychic can tell the future, they are likely to remember the statements that are true, forget the statements that are false, and interpret ambiguous comments in a favorable manner. Similarly, if you have ever watched a reality show featuring ghost-hunters, you might have seen confirmation bias in action. Every creaky noise becomes a footstep and every slight breeze a ghostly apparition when the person involved is already convinced the house they are exploring is haunted.

While confirmation bias may not be particularly harmful in the scenarios noted above, it can be extremely detrimental in a workplace investigation. If an investigator makes up their mind at an early stage that, for example, the complainant is lying, confirmation bias can cause them to actively seek out documents and witness that support this position, while failing to search for evidence to the contrary. It can even cause an investigator to over-rely on evidence that backs up their belief and forget hearing anything that contradicts it.

While it is natural to try to make sense of the information being gathered while an investigation is ongoing, it is crucial that investigators keep an open mind until the final report has been written. To keep confirmation bias in check, ask yourself:

• What assumptions am I making, and what are they based on?
• Am I asking the right questions during interviews, and am I making good use of open-ended questions?
• Is there any important information or witnesses I might have missed?

2. Primacy effect

A type of bias that is related to confirmation bias is the preference for early information, also known as the “primacy effect”. Studies[1] have shown that people have an innate preference for information they hear first. For example, people form a more positive impression of someone described as “intelligent, industrious, impulsive, critical, stubborn, and envious”, than if the order of adjectives is reversed. Having heard the positive traits first, the listener forms a hypothesis of what this individual is like, and gives less weight to the negative traits that come after.

In a workplace investigation, one version of the facts is always going to be presented first. Usually, this version comes from the complainant. An investigator must work to avoid jumping to conclusions based on this first telling of the events, and to keep an open mind to the respondent’s version as well. To avoid the primacy effect, consider:

• When did you come up with a theory of the case? Was it too early to do so, and how willing are you to waiver from that theory?
• If you had to support the opposite conclusion to the one that you have reached, what evidence would you use?

3. Defensive attribution bias

If you have ever read an online news article about a person suffering a misfortune, and instantly dozens of commenters chime in that it was the victim’s own fault (“He wouldn’t have gotten mugged if he didn’t walk in a bad neighbourhood at night! I would never do that!”), you have seen defensive attribution in action. Psychologists theorize that people do this in order to convince themselves that they are in control of the world around them, and to minimize fear that they will become a victim of a similar event.

Studies show that defensive attribution bias operates more strongly when we are considering the actions of a person who we see as dissimilar to ourselves[2]. Defense attorneys have argued that this type of bias can be seen in trials involving black defendants and all-white or mostly-white juries, causing jurors to have less sympathy for the accused and accordingly be less likely to consider mitigating factors in sentencing[3]. Conversely, studies have shown that male jurors are less likely to sympathize with female sexual assault victims, and are more likely to place blame on the victim rather than on the perpetrator[4].

When it comes to workplace investigations, an investigator is going to encounter complainants who are alleging that they have been victims of harassment and bullying. In many cases one of the parties will have more in common with the investigator than the other, whether it be gender, race or religion. When we are evaluating the truthfulness of what an individual is saying, and the impact that bullying or harassment might have had on a complainant, it is important to consider whether the defensive attribution bias is coming in to play. To see whether you are being impacted by defensive attribution bias, be honest with yourself about the following:

• What was your impression of each of the parties within the first few minutes of the interview? What was that impression based on?
• Do you find it particularly hard to believe the account of one of the parties and if so, why?

4. Courtesy bias

Courtesy bias is the tendency for people to respond in a way that is in keeping with what they think someone wants to hear. This type of bias often comes up in market research situations. For example, someone participating in a focus group might say they like a new soft drink, when in fact they don’t, because they do not want to seem impolite. It can impede the ability for the researcher to get honest and useful feedback.

Although investigations are very different from market research, investigators can also suffer from being too polite, or too eager to please their client. In most cases a workplace investigator is hired by an employer, and that employer might have their own ideas about the guilt or innocence of the respondent. Unfortunately, the result of some workplace investigations is a termination, and when the job of a valued employee is on the line a client might be a bit too eager to share their preference regarding the outcome of an investigation.

Another difficulty arises when the information gathered may appear to reflect poorly not only on the respondent, but on the workplace as a whole. An investigator might have to tell an employer that one of their employees has been harassed, and also that their own deficient policies and workplace culture contributed to the harassment.

Investigative findings can never be tailored to the preferences of the client, and an important part of an investigator’s job is to tell their client the hard truths.  Accordingly, it is vital for an investigator to be aware of the potential impact of courtesy bias in cases where they know a client may not be happy with the outcome of an investigation. Protect your investigations from courtesy bias by doing the following:

• From the first contact with the client, be clear and open about your role as a neutral, objective investigator.
• Consider what information you would need to address the problem if you were in the client’s shoes, and be honest with yourself about whether you have provided all that information.

5. Likeability bias

It’s not surprising that we tend to form a more positive general impression of those who are likeable. Studies have shown that when it comes to hiring decisions, whether someone is likeable is even more important than whether they are competent to do the job.

While it makes some sense that you would want to hire a person you like, it shouldn’t matter whether an investigator likes the person they are interviewing. The likeability of the complainant or respondent should never influence the investigative findings. In practice, however, it’s a common pitfall to think that a person who is nice to us during the investigation process is also more truthful and trustworthy. If one party is kind, friendly and polite and the other is surly, rude and insulting, it’s hard not to let that impression spill over into our findings on their credibility.

Investigators must keep in mind that just because someone is nice, it doesn’t mean they are honest. You can curb the effects of likeability bias by doing the following:

• Consider whether one party is particularly kind or rude to you, and how that made you feel during the interview.
• Review your notes a day or two after the interview, and ask yourself what someone who knows nothing about the parties would think of the evidence provided.

6. Empathy gap

The idea behind “empathy gap” is that it’s hard to imagine what it’s like to be in a state other than the one you’re in. If you’re stuffed from a full meal, it’s hard to imagine what someone who is hungry feels like. In hospital settings, studies[5] have shown that empathy gaps cause staff to routinely under-medicate patients who are in a great deal of pain, because it is hard to understand pain that you are not experiencing yourself.

This also applies to bullying. People have difficulty understanding the severity of the trauma experienced by bullying victims if they are not also being bullied themselves[6].

This is important for workplace investigators, since part of our job is to understand the subjective impact of workplace harassment or bullying on the complainant. It can be easy to think that bullying is less serious if you have never experienced it yourself. Investigators should ensure that they have explored with the complainant the subjective impact that the respondent’s behaviour had on them.

You can limit the effects of empathy gap on your investigations by:

• Asking the complainant how they felt during their interactions with the respondent, and why
• Continually evaluating what assumptions you are making about the complainant’s experiences

7. The Illusory Truth effect

The foundation of the illusory truth effect, also known as the reiteration effect, is that repetition makes things seem more plausible. This is the reason you hear talking points being circulated during elections – one candidate is “tough on crime”, while another is a “flip-flopper”. The more people hear these sayings repeated, the more likely they are to believe they are true.

All investigators have encountered a complainant or respondent who is stuck on repeat; you might hear “Jane is a bully” or “John is a liar” 20 or 30 times during the course of an interview. It is crucial for any investigator to examine the findings and assumptions they are making and to consider what evidence backs them up. If you believe Jane is a bully, is that because the evidence points to that conclusion, or because you heard it so many times during an interview that it became the “truth”?

Prevent the illusionary truth effect from de-railing your investigations by:

• Being alert for statements that are repeated without supporting evidence
• Actively considering what objective facts corroborate each version of the events

8. Information bias

This type of bias relates to the way an investigation is conducted, rather than the conclusions eventually reached. It causes a person to believe that more information is always better. In reality, sometimes extra information adds nothing to an investigation, and can only serve to confuse the issue.

A 1988 study[7] asked subjects to pretend to be a doctor treating a hypothetical patient who had one of three possible fictitious diseases. A test could be run that would determine if the patient had one of the diseases, but would not impact the course of treatment. Most subjects decided that the test should be run, even if it were costly. This is an example of thinking information is critical, even when, in a practical sense, it is unimportant.

How does this apply to workplace investigations? Consider the following scenario:

John says that his manager, Mark, called him an “ugly jerk” and shoved him. Mark admits that he called John a “jerk” and shoved him, but denies that he called him ugly. The harassment policy is clear that a manager calling someone that reports to them a derogatory name and shoving them would constitute harassment. What more information does a workplace investigator need?

Someone who is succumbing to information bias might go to great lengths to sort out whether Mark really did call John “ugly”. An important question to consider is: what difference does it make? In some cases the exact wording of the insult might be relevant, but in most the respondent’s own admission is enough for the investigator to conclude that the behaviour was in violation of the harassment policy.

Investigators tend to be naturally curious people, and accordingly it is easy for information bias to take hold. We don’t only want to understand the relevant facts; we want to understand everything.

Investigators can avoid information bias by:

• Continually evaluating their investigation plan and revising as necessary
• Determining whether the steps they are taking during an investigation are necessary for them to reach their findings

While we may all fall victim to one or more of these biases on occasion, it’s not inevitable that they will impact our work. There are some ways investigators can help avoid the impact of bias:

• Be honest with yourself about your own biases;
• Admit to yourself that you will not know everything at the beginning of an investigation, and that this is okay;
• Relentlessly challenge your own assumptions, and encourage your colleagues to do the same. Talking out a case with a co-worker can provide valuable third-party input that will let you know if you’re heading in the wrong direction;
• If more information is needed, seek out professional training on how to deal with bias.

Michelle Bird

About the Author: Michelle Bird conducts workplace investigations into allegations of harassment, bullying, poisoned work environments, and other problematic workplace behaviour. Michelle also provides workplace investigation and human rights training to staff at all levels.

[1] Asch, S. E. (1946) Forming impressions of personality, Journal of Abnormal and Social Psychology, 41, 258-290

[2] Griffitt, W., & Jackson, T. (1973). Simulated jury decisions: The influence of jury-defendant attitude similarity-dissimilarity. Social Behavior and Personality, 1, 1–7.

[3] Bowers WJ,  Steiner BD,  Sandys M. “Death Sentencing in Black and White: An Empirical Analysis of Jurors’ Race and Jury Racial Composition,”, University of Pennsylvania Journal of Constitutional Law , 2001, vol. 3 (pg. 171-275)

[4] Deitz, S. R., Blackwell, K. T., Daley, P. C., & Bentley, B. J. (1982). Measurement of empathy toward rape victims and rapists. Journal of Personality and Social Psychology, 43(2), 372-384.

[5] Twycross, A., & Powls, L. (2006). How do children’s nurses make clinical decisions? Two preliminary studies. Journal of Clinical Nursing, 15(10), 1324-1335.

[6] Nordgren, Loran, Kasia Banas and Geoff MacDonald. 2011. Empathy Gaps for Social Pain: Why People Underestimate the Pain of Social Suffering. Journal of Personality and Social Psychology. 100: 120-128.

[7] Baron, Jonathan; Beattie, Jane; Hershey, John C (1988). “Heuristics and biases in diagnostic reasoning” (PDF). Organizational Behavior and Human Decision Processes. 42 (1): 88–110