That other four-letter word – Bias

Recognizing, challenging and overcoming it

2019 May

Examples of bias are everywhere: political arguments, medical decisions, jury verdicts – the list is endless. So what is bias, and what is the impact? Bias is prejudice, for or against someone or something, which inhibits impartiality.

Background

There are many adjectives associated with the word: psychological, cognitive, overconfidence, anchoring and confirmation, to name a few. Although bias is receiving a lot of attention lately, it has been studied and written about, since the early 1600s, when Francis Bacon first used the term “confirmation bias,” and psychologists, like the late Amos Tversky, popularized the term “psychological” bias in the 1970s.

If you ask the average person to list their biases, they will say, “I am not biased.” That response is typical, since people are not aware they have a bias. Biases are subconscious – people do not intentionally try to develop a bias. Bias is learned, and as the song from South Pacific laments,

You’ve got to be taught
To be Afraid
Of people whose eyes
are oddly made
And people whose skin
Is a different shade
You’ve got to be carefully taught¹

According to Kendra Cherry, confirmation bias “influences how we interpret and recall information.” She goes on to say that, “People who support or oppose a particular issue will not only seek information to support it, they will also interpret news stories in a way that upholds their existing ideas.”²

The brain processes information in two different ways: quickly interpreting information and relying on intuition, versus slower, deliberative, analytical interpretation. The former is what often produces biased perceptions, which are treated as facts or absolutes. Research has shown that the conscious brain can process less than 50 pieces of information per second, and yet the unconscious brain can process thousands of bits of information in the same timeframe.

There is extensive research that demonstrates how unconscious bias influences decisions. Professor Marianne Bertrand conducted an experiment by sending out over 1000 identical job applications with just one variation – the name. She used names that might indicate the color of the applicant. Applicants that had white-sounding names got more interview calls than those that had names that might indicate that they were black.³ Other experiments have shown that if wine stores play French music, then sales for French wine goes up, and if they play German music, then German wine sales increase. Prior to 1970, most orchestra members were men. When auditions were done with the evaluators unable to see who was playing, many more women were hired.

Indicators of bias may include, but are not limited to placing blame, manipulating statistics or facts to support a point, and overconfidence. Key words associated with bias are “all” and “every.” For example, “All men...” “Every defense attorney...” “All insurance companies...” Telltale signs relating to fact tampering may include, “There was a study in 1984 that says...” The person citing the study ignores multiple and more recent research that disproves the earlier study. Overconfidence results when a person uses only information that supports their point-of-view, ignoring or discounting all information that contradicts their perspective, such as ignoring jury verdicts related to similar cases.

Suggestions for challenging and overcoming bias

There are a number of skills that can help prevent or overcome a variety of biases. Some techniques that attorneys might consider include: “anchor” awareness and questioning, reality-testing and risk analysis, and decision-making criteria.

Anchors and questions

Anchors are established unconsciously and are very powerful. When reviewing a case, if it sounds like a similar case, whatever the first case settled for can become the “anchor” for the second case’s value. This works with the first number thrown out when people are making initial demands or offers. Make an extremely high demand, and it creates an anchor and all other offers will be compared to it. An anchor can be created using a low-ball offer as well.

An anchor can also be created by a person’s physical placement at a table. If they are at the head or narrow end of the table, the subconscious mind says that is a position of power. If someone is seated along the side of the table and not close to the head of the table, the subconscious mind says, “Well that person is not very powerful.” Once the brain makes a determination, then a person is more likely to listen to the perceived more “powerful” person and is less likely to pay attention to and may even discount information presented by the assumed “less powerful” person.

Just becoming aware of what the subconscious mind is doing, will help a conscientious person to take a closer look to determine the accuracy of the biased assumption. One way to evaluate whether an offer or demand figure is credible or biased is to ask a simple question, “How did you arrive at that figure?”  This type of question forces the person to confront how and where they got the number, and the evidence to justify the number. If the person is unable to provide legitimate facts that give credence to their proposed number, then they can consciously reconsider, and review how they arrived at their conclusion, and then find a factually supportable number, thus avoiding the anchor bias.

The anchor bias is a predominate bias among professionals who must take in data, fairly quickly, and then make a decision. This bias is not one associated only with attorneys – it is also well documented in the medical profession.

Reality testing and risk analysis

There are many techniques that can be used to question the soundness or validity of information. Asking questions is extremely useful. In addition to asking the above question regarding an anchor, other questions may provide a reality check. For example, “In similar situations, what are other options that have worked for your clients?” Looking for examples of what others have done may help challenge the original biased proposal.

Other reality testing questions include: “If we are unable to settle today, what are your chances of doing better at trial?” or “What will it cost to prepare for trial?” These questions should get everyone thinking about the “parade of horribles”– never a pleasant task; however, it often forces a person to reanalyze the case value.

During settlement negotiations the parties need to weigh immediate and future financial needs, cost of litigation, probability of successfully defending their position, and the amount of the award. Building on reality-testing questions, is the use of a simple risk analysis, which is based on statistical principles. The primary statistical element associated with risk analysis is the Bell Curve, which predicts a statistical outcome: the high verdict is obtained 25 percent of the time, the middle is obtained 50 percent of the time, and the low is obtained 25 percent of the time.

Using this risk analysis technique can not only put into question the likelihood that someone will win or lose at trial, it is also very useful in helping to define the high and low value range for a case.

A graphic example of risk analysis

Begin the risk analysis process by asking: “If you tried this case in front of 10 computer-selected juries, how many times would you win?” Convert the number provided into a percentage. The next questions are for the predicted amounts for the high, middle and low verdicts. Once these numbers are identified, use the Bell Curve principle to multiply the three suggested verdict amounts by the appropriate percentage: 25 percent, 50 percent and 25 percent respectively. Add the three new figures, and then multiply that total by the original “win” prediction percentage. That new total is the statistical value of the case.

For example, someone speculates that they will win 75 percent of the time, and that the high verdict will be $500,000, the middle verdict will be $300,000 and the low will be $200,000. The results will look something like the following.

$500,000  X  25%  =  $125,000

$300,000  X  50%  =  $150,000

$200,000  X  25%  =  $ 50,000

Their total is $325,000, which might not sound too bad, although lower than the high, it is more than the speculated middle verdict. The risk analysis is not over, since the $325,000 needs to be multiplied by the projected 75 percent win! Seventy-five percent of $325,000 is $243,750. That is more than the low, and it is below the predicted high and middle verdicts.

It becomes hard for the person to argue with the numbers, since they proposed those numbers. Although they will not normally accept the statistical value, it does help move negotiations and it challenges the biased number they originally demanded. This process can also be used by having defense counsel predict the numbers.

Decision-making criteria

Criteria are the factors or standards that are used to determine what a person will decide or do. Decision-making criteria may be used as the checklist for determining what is actually important to a person, rather than relying on their biased statements and predictions.⁴ It allows a person to better understand some of the assumptions and biases an individual may hold.

As an example, if a person decides to buy a computer, specific criteria, consciously or unconsciously, are used to determine which features best meet their needs. Questions a person asks him/herself may include: “Do I want a laptop or desktop?”  “What size memory does the computer need?”  “What software is included?” “What will it cost?”  The criteria that are implicit in the preceding questions include brand or model, memory and economics.

When trying to understand an individual’s biases, it is important to get behind statements and find out what criteria they will use to make a decision. There are hundreds of factors that influence how someone makes a decision. Finality, time, fairness, customs, prevention, cost containment, confidentiality, emotional gratification, and ethics: the list is extensive.

Some basic examples: Time – “I will take this offer today, but not if I have to wait six months.” When time is the factor that is really important, then rather than arguing over an unrealistic number, focusing on how soon the case can settle will be more productive.

If emotional gratification is what the person wants, then learning what that means will speed up settlement, and again avoid the biased numbers argument. Does the person want a “pound of flesh,” or to have the other side admit it was wrong? If it is the former, then money is the typical solution. If it is an admission of guilt or wrongdoing, then an apology might be the solution.

Some people are absolutely convinced that they are right and everyone else is wrong. They are unable to accept any information or data that would disprove their biased assumption. Blaming is a significant bias indicator. Everyone else is to blame for everything that happens. These people are referred to as “professional victims.” Usually the biases are so entrenched that it is impossible to provide evidence that will dissuade the person who continually blames others. The “Blame Game” has now replaced baseball as the national pastime.

Conclusion

These suggestions and techniques accomplish several things. They require a person presenting a biased perspective or statement to first confront their assumptions. Through questions, reality testing, risk analysis and uncovering decision-making criteria, it is possible to essentially “hold a mirror up” and have the person take a look at their assumptions. Although these techniques do not work 100% of the time, they do provide an opportunity to find some common ground, thus helping to avoid personal attacks, and thus focus the conversation on drawing out credible information.