In A Tutorial Introduction to Decision Theory, D. Warner North succinctly describes decision theory as "a rational framework for choosing between alternative courses of action when the consequences resulting from this choice are imperfectly known." In this letter I'm using the term decision theory a little more loosely, to refer to almost any kind of math or logic or reasoning that you can apply to real-life situations to make better decisions. In reality there are a number of related fields, with a number of related names: decision theory, operations research, decision analysis, systems science, Bayesian analysis, management science, systems analysis, game theory, etc.
Practicing full-blown, formal Decision Analysis requires a lot of probability and mathematics, but fortunately, the complete rigorous treatment of the discipline is rarely necessary. The simplest precepts of Decision Analysis are also its most useful. A couple of basic skills can sharpen anyone's decision making.
Here are some of the Decision Analysis strategies a person can use when thinking about a problem: Ask what the alternative courses of action are. Ask what the likely outcomes are. Decide what your relative preferences among the outcomes are. Concretely list alternatives and outcomes. Use numbers to represent values. Resist the temptation to just juggle and compare perceived values in your head. Work in numbers, even if they are just estimates. Estimates, guesses, and probabilities are okay, but you need to include all the cases -- even the ones which seem unthinkable. The only answer which is not allowed is "I have no idea." That is almost never the case. You can always make an estimate. This analysis will not tell you what is going to happen, but it can give you a good idea of what is likely and what impact your decision may have. One of the main results of Decision Analysis research has been to show that a decision patched together with guesses and estimates is far better than no decision.
The basic strategies outlined above form the foundation of a good decision making technique. Good decision making should also account for important factors like personal priorities, short and long term effects, and interrelationships between different problems.
As I pointed out in the first section of this letter, people's values and priorities vary widely. For instance, many people have a preference for preventing tragedies in their own country before helping people overseas. Our decision making process should be sensitive to personal values and priorities, and similarly, should account for both short-term and long-term effects. Take famine as an example. We can prevent famine tragedies either by sending food today to the current famine hot spot or by working to prevent famine a decade from now by funding family planning, irrigation, education, etc. Both methods will work, but some people have a preference for preventing the tragedies that will happen next year before preventing tragedies that will happen later, so our decision making process should account for both the short and long term effects, and weigh them against our priorities.
Additionally, our decision making process must not treat problems as being independent of each other, and the solutions as having only one effect. Actually, all the problems are tied up with each other, and a "solution" often alleviates a number of different problems, while contributing to other problems. A common mistake we make is to think an action can have only one effect. In reality, you can never do just one thing. Our culture has learned to think in terms of effects and side-effects, but side-effects are really first-class effects that we've simply chosen to downplay. Our decision making process should account for interdependencies, and should help us identify solutions that leverage off interrelationships and common problem roots.
"Remembering that one of the functions of language is to prevent thought, we easily recognize the purpose of the term "side effects": it is to discourage thinking about the total effects of a new medicine, a new pesticide, or a new public works project when some of the consequences prove embarrassing to the promoters. ...an effect is an effect is an effect. The adjective "side" is added to coerce thinking -- to restrict questions to safe channels (safe for the promoter's enterprise)."
-- Garrett Hardin, Filters Against Folly
The Bottom Line
We need to use decision making techniques that account for
personal values and priorities, short and long term effects, and
interdependencies between problems.
