Purpose
Your participation in the IEEE Awards program, by encouraging and recognizing outstanding contributions to our profession, is valuable to the careers of engineers. The IEEE Awards Board is working to ensure that such contributions are equitably recognized without regard to gender, ethnicity, region, or technical specialization. This briefing summarizes a body of social science research on implicit bias, hopes to convince you that implicit bias can be a real impediment to fair decision-making in the awards process, and suggests some concrete steps that can be taken in individual and committee deliberative processes to mitigate its undesirable effects.
What is implicit bias?
Implicit bias occurs when we make judgments or decisions on the basis of personal deep-seated and thought patterns, assumptions or interpretations that subconsciously and unintentionally affect the outcome. The irony is that the allowance of factors not based on pure facts or merit are the inevitable by-products of the efficiency of human cognition.
Most of us believe that we are more fair and less prejudiced than the average or typical person. Research has shown that this is one of many implicit biases that we draw on in order to make fast decisions. Unfortunately, fast decisions can be the enemy of good decisions (perhaps not in avoiding danger, but at least in award deliberations).
Research strongly suggests that we have simultaneously a positive bias towards our in-group, and a negative bias towards an out-group. In the context of IEEE awards, we are each members of several relevant in-groups, including gender, ethnicity, region and language, and technical specialization. We are familiar with members of our in-group and feel on firm ground when judging their excellence and trustworthiness. We perceive a pleasant fluency of action when we experience familiarity, and this makes us feel confident and in control of our decisions. With unfamiliar members of other groups we are on less sure ground. It may seem like selecting a candidate from an out-group involves taking a higher risk. Actually, in the case of both familiar and unfamiliar candidates, it is very difficult to shut out implicit preferences and fears. We may actually (subconsciously and unintentionally) tend to redefine merit for members of out-groups, causing us to believe that men or women or people from different cultures or regions or even different technical disciplines might not be quite such excellent engineers. But in order to make decisions on the basis of merit and excellence alone we must overcome these unavoidable tendencies.
How can we recognize implicit bias?
The very act of realizing that we have implicit biases can enable us to mentally monitor and attempt to ameliorate any hidden attitudes before they are expressed in our deliberative or decision processes. Once we accept that we will all quite naturally use subconscious mental shortcuts, then we can take the time to reconsider our judgements and decisions, asking the tough and uncomfortable questions about whether they are based entirely on facts and merit.
Measuring your own implicit bias
One of the tools for measuring implicit bias is the Implicit Association Test (IAT). Specifically, it measures “the strength of a person's automatic association between mental representations of objects (concepts) in memory”. If you are curious about your own subconscious mental associations, you can take this test online at https://implicit.harvard.edu/implicit/takeatest.html. Most subjects are surprised at the strength of these automatic associations.
What can we do about implicit bias?
None of us wants to manifest any form of bias, but the reality is that we are incapable of either accessing or modifying our own subconscious cognitive processes. We can nevertheless achieve more fairness and improve the quality of our decision-making if we make a commitment to recognizing any stereotypes we may implicitly harbor. This requires slowing down our decision process, while being more contemplative about and vigilant of any possible sources of bias.
Research has found that we are able to see the operation of bias in others more effectively than in ourselves. Increasing the diversity of awards committee membership in all dimensions (gender, geography and language, and the in-groups represented) is an important aspect of addressing implicit bias in awards deliberations. Doing so brings a similar diversity in implicit bias among committee members, rendering the committee as a whole more effective at recognizing and overcoming these biases. However, this works effectively only if committee members openly discuss forms of bias and help one another to recognize instances of implicit bias.
Committee action points
Here are some specific suggestions for how IEEE award selection committees might actively minimize implicit bias:
-
Before meetings, committee members are encouraged to slow down their speed of opinion formation, and during meetings, the committee as a whole is encouraged to do the same for decision-making. This helps to avoid mental shortcuts, which are a primary source of implicit bias.
-
The committee is encouraged to strive for a friendly and collegial environment for deliberations in which members feel empowered to help one another identify and avoid implicit bias. Committee members are encouraged to actively question and discuss the possible contributions of implicit bias in opinions expressed by other committee members, and actively solicit help from others in identifying possible instances of implicit bias in their own thinking.
-
Each committee member is encouraged to consciously consider the justifications for their opinions and decisions, keeping in mind the possibility that these may be in part based on implicit biases or stereotypes.
Attribution
Prepared in September 2016 by David Messerschmitt, Chair of the IEEE Medals Council, and Andrea Goldsmith, Chair of the IEEE Alexander Bell Medal Committee. Updated in 2024