“Noise” delves into the hidden flaws of human judgment, offering a groundbreaking perspective on decision-making. Kahneman, Sibony, and Sunstein’s powerful insights reveal the surprising impact of random variability on our choices, from courtrooms to boardrooms.
Dive into this eye-opening exploration of human judgment and discover how to make better decisions in your personal and professional life.
Table of Contents
Genres
Psychology, Behavioral Economics, Decision Making, Cognitive Science, Business, Social Science, Self-Help, Management, Leadership, Critical Thinking
“Noise” explores the concept of unwanted variability in human judgments. The authors argue that noise, distinct from bias, significantly affects decision-making across various fields. They present compelling evidence from studies in medicine, law, and business, demonstrating how professionals often reach different conclusions when presented with identical information.
The book introduces the concept of “decision hygiene” as a method to reduce noise. This involves strategies such as breaking complex judgments into smaller tasks, aggregating independent judgments, and using algorithms where appropriate. The authors also discuss the challenges of implementing these strategies in real-world settings.
A key insight is the distinction between “system noise” (variability between different judges) and “occasion noise” (variability in an individual’s judgments over time). The book provides practical tools for identifying and measuring noise in organizations, including the “noise audit.”
The authors emphasize that reducing noise can lead to substantial improvements in decision quality, often more so than addressing bias. They argue for a balanced approach that considers both noise and bias in efforts to improve judgment and decision-making.
Review
“Noise” offers a fresh perspective on decision-making, highlighting an often-overlooked aspect of human judgment. The authors’ expertise shines through in their clear explanations of complex concepts and their use of diverse, engaging examples.
The book’s strength lies in its practical approach. It not only identifies problems but also provides actionable strategies for improvement. The concept of “decision hygiene” is particularly valuable for professionals and organizations looking to enhance their decision-making processes.
While the book is thoroughly researched and well-argued, some readers might find certain sections repetitive. The technical nature of some discussions may also be challenging for casual readers.
One of the book’s most thought-provoking aspects is its exploration of the tension between human judgment and algorithmic decision-making. This raises important questions about the future of professional judgment in various fields.
Overall, “Noise” is a significant contribution to the understanding of human decision-making. It’s a must-read for anyone interested in improving their judgment or organizational decision processes. The insights provided are applicable across various domains, making it valuable for a wide range of professionals and decision-makers.
If your judgment is inconsistent, you’ll frustrate the people around you.
- If you’re a manager and your promotions seem random, team members will get frustrated and quit.
- If you’re frequently asked to give estimates at work and they vary greatly from week to week, your coworkers will stop asking for your input.
Author Daniel Kahneman calls variable and seemingly random judgment, noisy judgment. Noisy judgment differs from bias judgment, and it’s harder to correct. To understand the difference, picture two targets at a shooting range:
- The first target has five shots clustered in the bottom right corner, making it easy to spot and correct the bias (simply get the shooter to adjust the scope on the rifle).
- The second target has five shots scattered throughout the target. It’s impossible to predict the location of the sixth shot, and difficult to quickly correct the shooter’s inaccuracy.
The same is true for noisy judgment. If your estimates are consistently 30% to 50% lower than they should be, you factor that into your next estimate. But if your estimates range between 50% lower and 50% higher, you have a more complex problem to solve.
Noisy judgment exists for a variety of reasons:
- Changes in one’s mood create significant noise.
- The sequence in which judgments are made produces noise. Studies show that the chance of being admitted to an asylum drops by 19% if a judge has just admitted two people to an asylum.
- Within groups, individualism generates noise. We all have a desire to be seen and heard, which leads some people to exaggerate their viewpoints and make extreme predictions.
Practice Good Decision Hygiene
The best way to reduce noise is to practice good decision hygiene. Just as washing your hands after going to the bathroom reduces the odds of getting sick and brushing your teeth reduces the odds of getting cavities, there are two things you can do to practice good decision hygiene and save yourself from the eventual pain from noisy judgments:
Establish a “Mean” Anchor
Establishing a mean anchor ensures you don’t stray too far from a statistical average. Start your search for a mean anchor by looking for the statistical averages in the category you’re working in. If you’re determining the likelihood of someone passing an exam, look at historic pass-fail rates. You could now use large language models by asking, “What’s the average rate of X?” or “What is the probability of Y happening?”
Sometimes, you don’t have public data for a judgment. In these instances, leverage the wisdom of a crowd. The key to leveraging the wisdom of crowds is to get enough people so that they start to cancel out each other’s biases and ensure that those people don’t know what each other is saying or which way you are leaning. This could be as simple as emailing five coworkers to get their estimates (without sharing your estimate), and then calculating the mean average value of the estimates you receive.
After you acquire a statistical mean average or “wisdom of the crowd” mean average:
- Make a judgment either higher or lower than the average.
- List the information that supports your judgment being higher or lower.
- Rate the predictive power of the information from 0-100%.
- Move yourself from the mean anchor out to your judgment based on how predictive your information is.
Suppose you were trying to predict the GPA of a girl named Amy, but all you had was Amy’s photo. Your information has virtually zero predictive power, and you should stick to the mean. But if you know Amy’s grades for every subject that year, you have perfectly predictive information and can confidently go with your judgment.
Aggregate Your Judgments
If you truly want to optimize your judgment, use your time before announcing your decision to tap into the wisdom of your “inner crowd.” Since mood, fatigue, stress levels, exposure to recent events, and even the weather influence your judgment, you should check in with yourself over the next few hours or days by setting at least three “judgment revision reminders” on your phone. When you receive the “judgment revision reminder” notification, assume that your previous judgment was off the mark and push yourself as far from your last judgment as possible. Then, write down this new alternative judgment in a note on your phone, and then return to what you were doing. When it’s time to announce your final decision, take the average of all the estimates you made.
“Wherever there’s judgment, there’s noise, and more of it than you think.” – Daniel Kahneman