When our beliefs blind us from facts and data

Some examples of how cognitive biases, group thinking or communication can distort objective information and lead us to bad decisions based on our beliefs and not on concrete data.

“Sometimes people don't want to hear the truth because they don't want their illusions destroyed.” ― Friedrich Nietzsche

Abstract:

• There are several types of biases, besides the statistical ones, studied in Lean Six Sigma. Some examples were discussed here;

• Cognitive biases can add subjectivity and alienate us from data and facts;

• The Ladder of Inference brings an interesting view of how we move away from the objective information previously collected;

• We must step out of our comfort zone, question paradigms and consider the possibility that the results are different from what we expected.

If you are certified as a Green or Black Belt in Lean Six Sigma methodology, you have learned that, besides the process variation of the “5M´s” (Man, Machine, Material, Method and Mother Nature), there is a sixth “M” that comprehends variation due to the measurement system. We can use Measurement System Analysis tools to evaluate, among other features, the impact of bias. We say that our system is biased when our measurements are consistently shifting to a different value. Therefore, bias is our accuracy, a metric of how far our measurements are from the “true” values. Examples are: a scale that consistently weights 2kg more than our actual weight, a delayed timer, an auditor that is more demanding than all others or just a dice that returns an uneven probability of its numbers when it's rolled.

Cognitive Bias and Data Analyses

Besides the statistical biases, the not so far, cognitive biases also affect our data analysis. According to Haselton et al., cognitive bias is a systematic pattern of deviation from rationality in judgment [1]. This distorted translation from reality leads us to mistaken conclusions thus bad decisions. There are several types of cognitive biases and many of them can directly jeopardize our interpretation of concrete results.

For example, when we collect items that don't represent our population it may be linked to the Halo Effect (e.g. if we assume first suitable units as our reference) or Confirmation Bias (e.g. when we search for data that confirms our expectations or purge data that don't guarantee these).

Communication Bias and Data Analyses

Communication naturally plays a relevant role in how data is understood and transmitted. As we learn within Lean projects, processes with many interfaces are usually more problematic. Approximately one out of five projects fails due to ineffective communication [2], and PMI (Project Management Institute) suggests a project manager should spend 90 percent of their time communicating [3]. Considering variation among stakeholders and how they understand information plus how prepared they are in properly transmitting it, we can realize how relevant the risk is of committing mistakes in post processing data and information.

Erin Meyer brings good examples and explanations on how our culture changes the way we share and understand information. She explains the differences between low and high-context cultures using the USA and Germany as examples for the low. In these cultures, messages are direct, clear and precise. While in high-context cultures such as Asian countries where a more sophisticated communication is employed and frequently messages are implied but not plainly expressed. Brazilian culture and communication is much more high-context based than German culture is [4]. In his speech, Ariano Suassuna, a famous Brazilian writer, once stated that he had heard before (not detailing the source), that when a Brazilian is invited to a party and responds “I will go”, this person might or not be going but, if the response is “I will do my best to be there”, you can be certain this person will not go.

GroupThink effect

Groupthink occurs when the desire to maintain group loyalty becomes more important than making the best choices [5]. It is easy to imagine how its symptoms (illusion of invulnerability and unanimity, belief in group’s competence and morality, self censorship, mind protection against negative information, distorted view of “enemies”, direct pressure over group members..) affect how we interpret data. Lack of discussions and group isolation may lead to a distorted view and increasing bias.

The Ladder of inference

An interesting concept to illustrate how our beliefs and expectations interfere in our conclusions even if we started from facts and data, is the Ladder of Inference. It was initially developed by Dr. Chris Argyris and later presented by Peter Senge’s book The Fifth Discipline Fieldbook. Our capability of meeting our goals is harmed by our confidence in assumptions such as: our beliefs are the true ones and are based on actual data, truth is clearly stated and/or the data I selected are the true ones.

You can find below the illustration of this concept, extracted from Peter Sange's book The Fifth Discipline Fieldbook.

Conclusion

All these and other human effects put together are triggers to adding subjectivity in something so objective as data analysis. As we step up the Ladder of Inference, these biases are generated and, many times lead us to bad decisions, based more in our beliefs than in what actual data is showing.

We must be open to different information and always consider the risk of falling into the trap of biased approaches in order to make better decisions. Improvement only comes if we step out of our comfort zone and consider that results might really be different from our expectations.

References:

1 - Haselton MG, Nettle D, Andrews PW (2005). "The evolution of cognitive bias.". In Buss DM (ed.). The Handbook of Evolutionary Psychology (PDF). Hoboken, NJ, US: John Wiley & Sons Inc. pp. 724–746.

2 - ©2013 Project Management Institute, Inc. Pulse of the Profession In-Depth Report: The High Cost of Low Performance: The Essential Role of Communications, May 2013. PMI.org/ Pulse

3- https://2020projectmanagement.com/resources/communication-management/communication-the-key-to-successful-project-management

4 - Meyer, E., “The Culture Map – Decoding how people think, lead, and get things done across cultures” pp. 37-41

5 - https://ethicsunwrapped.utexas.edu/glossary/groupthink

6 - Senge, P., The Fifth Discipline Fieldbook. Doubleday, New York, 1994