Overview Judgment and Decision Making
Lecture 1: Decision making task
Bad decisions:
o Errors in strategic decision making are not exceptional
o The majority of a survey said bad decisions are just as frequent as good ones.
o When it comes to certain types of decisions, failures are much more frequent
than successes.
o Every successful l strategy is successful in its own way. But all strategic
failures are alike.
o Bad decisions are not made by bad leaders. These are good, even great,
leaders who make predictable bad decisions.
Humans are not perfect decision makers. Not only are we not perfect, but we depart
from perfection or rationality in systematic and predictable ways. The understanding
of these systematic and predictable departures is core to the field of judgment and
decision making. By understanding these limitations, we can also identify strategies
for making better and more effective decisions.
Lecture 2: Loss aversion & framing of decisions
There are different types of decisions:
Major reflective decisions:
o Invest in a stock, buy a house, end a relationship, etc.
Low level decisions:
o Shopping in a supermarket, which way to go, what to wear, etc.
Factors that should affect a decision:
What you want and what you want to avoid
Strength of preferences (and dislikes) for particular outcomes
What factors or events will affect whether the outcome will be good, mediocre or bad,
and to what degree
How likely the different possibilities are
How should decisions be made:
Define the problem
Identify the decision criteria
Weight the identified decision making criteria
Generate possible alternatives
Rate each alternative against the decision maker’s criteria
Compute the optimal decision
Assumptions:
Assumes the decision maker is rational
Assumes the problem is clear and unambiguous
Assumes the decision maker has complete information
No time or cost
Choice will be one with the maximum payoff
Normative decision analysis:
Enumerate options
Enumerate outcomes
Construct a decision analysis for the decision
Evaluate the probabilities of different possible outcomes
, Determine which option has the greatest “expected utility”
Decision making research before 1970s:
normative theories that prescribe how people “ought” to make decisions in a perfectly
rational way, and many implicitly assumed that most people, in daily lives, followed
these normative rules.
Reference point: When we respond to attributes such as brightness, loudness, or
temperature, the past and present context of experience defines an adaptation level, or
reference point, and stimuli are perceived in relation to this reference point. Thus, an object
at a given temperature may be experienced as hot or cold to the touch depending on the
temperature to which one has adapted.
DMT:
You are given Additional €500 for 50% chance for
€1000,- for sure sure getting another
€1000,-
You are given
€2000,- for sure
Lose €500,- for sure 51 23 74
50% chance for 31 21 52
losing €1000,-
82 44 126
51 kiezen voor €1500. Maar meer mensen kiezen voor 50% kans op verliezen met
begin €2000,- dan 50% kans op winnen met begin €1000.
The Asian disease problem shows that people do not act normally.
DMT:
There are different programs in this situation:
A. 200 people will be saved
B. 33% change that all 600 will be saved;
67% change that none will be saved
C. 400 people will die
D. 33% change that none will die;
67% change that all 600 will die
Expected utility theory:
People think about uncertain events in terms of gambles
Gambles have two components
o Probability, p
o Value, v
The expected value of a gamble (EV) = p x v
The expected value of a gamble that offers a 25% chance of winning €100,- = €25,-
Dominance principle alternative gambles can be ranked from best to worst in terms
of expected value
Cancellation a choice between gambles should depend only on those outcomes that
differ
Transitivity i.e., if you prefer A to B and B to C, then you must prefer A to C
Invariance preference should remain invariant or stable, no matter how choices are
described
The framing effect Asian disease problem: