Decision Toolkits
Student edition
You may already know the research on how judgement goes wrong, but knowing it and using it on your own decisions are two different things. The Student Edition of the decision protocols turns that research into something you can work through when a real choice is in front of you, whether it’s about your studies, your career or how to divide your time. This page is a first look at it, with three pieces to try: a tool for reframing a problem before you try to solve it, a protocol for working out how fast a decision should be made, and a checklist for evaluating the decision you reach.
TOOL: Reframing
Reframing the decision
Premise: How a problem is presented to us is only one form of many.
Every decision arrives inside a frame: a set of assumptions about what the problem is, what is in scope, and from whose perspective we are looking. The frame is usually invisible. We do not experience it as a frame; you experience it as "the situation." But the frame is not the situation. It is one version of the situation, and it determines which solutions are even conceivable.
There are at least seven distinct moves available for reframing. This tool walks you through them.
PROTOCOL: How fast to decide
How much thinking does this decision deserve?
A short protocol for one decision you are facing. Up to eight questions tell you whether to decide now, find out one thing first, or slow down and think it through. This edition uses examples from life as a master’s student.
Use this protocol when
You have one specific decision to make, and you are not sure how much time to give it: an elective to choose, a dissertation topic, an internship offer, a flat to rent. If your problem is too many tasks competing for your time, rather than one decision, this is not the right protocol.
Why this matters
We get this wrong in both directions. Some decisions get far more thought than they deserve: a small choice discussed with friends for a week while the deadline gets closer. Others get far less: a big choice that is hard to undo, made on instinct late at night. Being fast is not always good, and being careful is not always good. What matters is matching the amount of thinking to the decision.
This protocol draws from two key research findings. First, fast and good are not opposites. In a well-known study of fast-moving technology companies, the executives who decided fastest did not use less information than the slow ones. They used more up-to-date information, and they compared several options at the same time. Second, instinct can be trusted only under certain conditions: when you have made many similar decisions, in a setting regular enough to learn from, and you found out quickly and clearly whether you were right. Where those conditions are missing, a strong gut feeling is not evidence that you are right, however certain it feels.
Before you start
Write the decision down in one sentence, as a choice. For example: “Should I accept the internship offer I have now, or wait to hear from the company I prefer?” If you cannot write it as one sentence, you may be facing several decisions at once. Take them one at a time. Most people answer between two and eight questions.
Quick check: Am I making a good decision?
Am I making a good decision?
This check looks at a decision before you make it, when nobody yet knows how it will turn out. It asks whether you are deciding well with what you can know today, which is the part of a decision you control.
Use this checklist when the decision is yours to make, fairly easy to reverse, and cheap to get wrong. A big decision, or one you make with others, needs a fuller review.
What you get. A decision is only as strong as its weakest link, so your lowest answer sets the result. It also shows where your next step should be.
Before you start. We tend to rate our own reasoning more kindly than someone else would. If possible, show your answers to a person who is not involved, such as the person you explain your decision to.
What it cannot do. Passing this check raises the odds of a good outcome but does not promise one. Outcomes also depend on luck and on events nobody controls, so a well-made decision can turn out badly and a poor one can turn out well. Over many decisions, though, a good process brings more good results and fewer bad ones than a poor process does.
It also cannot protect you from hindsight. Studies find that people rate the same decision as worse when told it turned out badly, even when they had exactly the same information as the person who decided. That is why the last question asks you to write down what you expect: later, the record shows what you knew when you decided.