Class Notes (1,100,000)
US (450,000)
UMD (8,000)
INST (100)
Anton (10)
Lecture 11

INST 354 Lecture 11: INST354 Lecture 11: Picturing Decisions


Department
Information Studies
Course Code
INST 354
Professor
Anton
Lecture
11

This preview shows half of the first page. to view the full 3 pages of the document.
INST354 Lecture 11: Picturing Decisions
We will use schematic decision treediagrams to describe decision situations throughout this
book. One of the major uses of these diagrams is to summarize the essential structures of
personal or public decision situations in order to apply the principles of scientific decision theory
to choose the best course of action. We will introduce this applied “decision analysis” approach
in Chapter 11. But for the moment, we want to explain the method of constructing the diagrams
so that we can use them to describe the tasks and situations that are important in research on
decision-making behavior.
The conventions of the decision tree diagram are that the situation is represented as a
hypothetical map of choice points and outcomes that lead to experienced consequences, like a
roadmap representing forks in a road and the objects that are located along the road. For
example, we might summarize a medical situation concerning a knee injury, as in Figure 2.1. On
the left is a choice point—we use squares, to indicate “choice points” at which the decision
maker chooses a course of action; the lines represent choices that lead to the outcomes that
follow from choosing each course of action. In the medical example, we imagine two possible
courses of action: have a knee operation or do not have the operation. Events that are out of the
decision maker’s control are indicated by circles () representing uncertain outcomes, sometimes
the actions of a competitor or just another less-than-perfectly predictable human agent; we don’t
know for sure, nor can we control which path we will take out of a circle. In the medical example,
the upper path (“do not operate”) is associated with two possible outcomes: The knee improves
on its own (it was “normal” in the first place) or the knee remains in bad shape (it was truly
injured). The lower path, representing the “have the operation” course of action, is also
associated with two outcome paths: The operation is successful (maybe the operation was
necessary and fixed the problem, or maybe the operation was unnecessary) or the operation is
a failure.
On the far right-hand side of the diagram, we list the consequences that are associated with
choice points and events in the decision tree. We will often summarize the decision maker’s
evaluations of those outcomes (traditionally called “utilities,” but we prefer to call them “personal
values”) with numbers. Sometimes a decision problem is stated with numbers associated with
the consequences (e.g., money payoff gambles; life-and-death medical and policy problems,
with “lives saved–lives lost” tabulations). In these problems, we may use the numbers in the
You're Reading a Preview

Unlock to view full version