In our day to day professional life, decision-makers can often succumb to the temptation of making snap judgments that are influenced by the context, or experiences from similar situations in the past. This is commonly known as the 'Correlation Bias'.
In principle, the idea of drawing from past experiences to make future decisions isn’t wrong - in fact, it's an essential requirement. However, if one is not mindful - it’s easy for individuals to succumb to a 'default' position, which can lead to an error in decision making.
To understand the meaning and relevance of de-correlation, let us look at the illustration below from a renowned psychologist, economist, and author; who is also a Nobel Prize winner in the field of Behavioral Economics.
While evaluating student examination papers at a University, an economics Professor would evaluate each exam paper the conventional way: One exam paper at a time, before moving onto the next student. The professor eventually began to notice that his evaluations of the answers in each booklet were strikingly homogenous, and began to suspect that his grading exhibited a 'halo' effect; and that the first question he assessed, had a disproportionate effect on the overall grade. He noticed that If he had given a high score to the first answer, he gave the student the benefit of doubt whenever he encountered a vague and ambiguous response in subsequent questions. But he soon realized that there was a serious problem with this approach. If a student had answered two questions - one strong and one weak, he would end up with a different final grade depending upon which answer the Professor read first. While the two questions were to carry equal weight, it was clear that the first one had a much greater impact on the final grade than the second. This was an unacceptable outcome, and so - he adopted a new method.
Instead of reading the booklets in sequence, he read and scored all the students’ answers to the first question and then went on to the next. He maintained the scoring separately so that his evaluation of the first answer was not visible while evaluating the second answer. He also masked the students’ names so he didn’t know which student was being evaluated. Soon after switching to the new method, the Professor made some striking observations. He found that for over half of the students, his scoring for a students’ second answer was significantly different to the score for the first answer, to a point where he felt tempted to change his evaluation, once he knew the score assigned for the first answer. He was now less happy, and less confident with his grading abilities, but recognized that this was a good sign and an indication that the new method was superior. The consistency and correlation he had enjoyed earlier produced a feeling of ease but was ineffective since it had allowed him to be influenced by his assessment of the first question. The procedure adopted is an example of decorrelation.
Studies have shown that our mind tends to build trends and correlate events to help execute daily tasks efficiently. However, other studies and experiments have also found that when observations and assessments are independent, and the assessor is devoid of the ability to correlate to prior events, the outcome is far more objective and fair.
In organizational decision making, It is common for individuals to be guided and influenced by the views and opinions of others; based on past interactions and an assessment of perceived intelligence. However, this snap judgment may be flawed, as it can correlate to a previous event that may not be relevant in the current situation.
In order to ensure organizational effectiveness, it becomes extremely important that managers and leaders practice de-correlation while making employee and business decisions.
Tips to de-correlate while making business decisions
While being mindful of the correlation bias is key, below are other simple practices that organizations can adopt to help promote and enable objective and unbiased decision making.
Management and Executives spend a great deal of their working day in business meetings. A simple rule to decorrelate in business meetings is that, before an issue is discussed, all members should be asked to write a very brief summary of their position on the matter at hand. This procedure makes good use of the diversity of knowledge and opinion in the group. The standard practice of open discussion gives too much weight to the opinions of those who speak early and assertively. The opinions of others and their position on the matter is greatly influenced by those who are perceived as intellectuals or subject matter experts.
Forming independent committees to review strategic programs is another way to de-correlate since the committee members have little or no prior affiliation with those involved in the formulation of these programs. By ensuring diversity of committee members across seniority levels and functional units, organizations can bring in a fresh unbiased perspective.
Thirdly, when part of a presentation, our evaluation can be influenced by our impression of the presenter; as well as the manner in which the data is presented. We can de-correlate by consciously making independent judgments on different attributes of the presentation. Below is an example of how different attributes can be judged independently. While evaluating a vendor proposal for service offerings, the feedback could be framed as: “The presenter was extremely eloquent and possessed great communication and interpersonal skills. His knowledge of the product, as well as our Industry, was impressive. However, the proposal does not address several requirements we had put forth, and data security could be a concern”. By evaluating the attributes independently, the risk of correlation can be minimized.
Organizations and their leaders make crucial decisions every day that have the potential to alter their future. It is hence of paramount importance, that decision-makers be aware of influences that can lead to potential errors in judgment. By being mindful and making small changes to decision-making practices, organizations can mitigate to a large extent, the associated risks of judgment errors.
*Views are personal