Chapter
5: The Mighty Decision Table: How do I choose
The
Decision Latter—start from the bottom of it.
Start with the most-simplest method and move up only as necessary. (For
this class I will ask you to work your way thought several steps).
Start
with the important considerations, bringing in details only when and if needed.
1. Prioritized List: Rank order your alternatives
a. You can do this by either
Intuition or actual analysis
b. Analysis breaks the problem
into sub problems while intuition takes into account everything at once.
c. Analysis takes longer,
requires training…but one isn’t necessarily as good as another.
d. General rule…if decision is
important you should do both…. (put intuition away, take a break, come back and
do and analysis).
e. If they are not in
agreement, either one or both could be wrong…this means that there is an error.
f.
Intuition is important, do not confuse it with biased feelings
however (flat earth)
2. Reasons versus outcomes.
a. You can choose among a set
of alternatives for an alternative that has good consequences or good reasons
for being chosen….to justify it to other. (example voting)
Screening:
Screening
is a process that you can use to reduce the number of alternatives that you
have by examining each alternative one value at a time.
1. Doesn’t take into account
trade offs….so it’s limited in that regard.
2. Rejects alternatives that
are clearly unacceptable *BMW Z3 for under 10K*
3. This process is referred too
as a non-compensatory decision…as you are not trading off on values.
Screen
with the most important value first, then the next…and so forth. If in doubt retain it. When only one alternative remains accept
it. If you have more than one
alternative, move up the decision ladder.
This
is a decision method of choice when your values have overriding importance
(money is one of them, book talks about ethical decisions). Also in decisions of low importance, and the
other in high cognitive processing cost.
Decision
Tables: The idea of a controlled
comparison
Describe
how one is set up.
Allows
for attention to be drawn to cells that are overlooked
Acts
as a form of stimulus variation: seeking of dominance of an alternative
Fact
table: getting facts is crucial to
decision making….
You
must be willing to put your values on your alternatives to the test, or this
will end up being a self enhancing belief….Real units are superior to
percentages.
Plusses
and Minuses Value Table
This
is the simpliest way to enter values into a value table.
Decision
Table. Plus Zero Minus…plus plus
You
can put these in in terms of absolute values…a plus is a positive
experience..zero neutral experience, negative…negative experience…
Or
in relative terms… plus represents an improvement… zero status quo and a minus
a worsening from status quoe…
Or
highest and lowest value.
This
is all we need to screen out irrelevant values and dominated alternatives.
1. attribute is all plusses…all
zeros or all minuses…it may be not worth considering…
2. same with too many minuses
3. specifically what we are
looking for is dominance, which helps us reject the alternative
Dominance: Alternative B is dominated by Alternative A
if there is at least one difference in A’s favor and no difference in B’s
Favor.
See
example in book for both alternatives and values…
Dominance
indicates a superior alternative.
1. If you are unable to
identify an alternative that dominates all others, try to create one.
1
to 10 Tables
If
it isn’t clear from the pluses and minuses table…then we may need to go this
far (which you will have to do for your project)
There
are four steps.
1.
Rate
(usually based upon putting a value upon the fact table, 1 is assumed the
worst)
We will talk about local ranges, where 1 is the
worst case, and 10 is the best case.
However if needed use a Global judgment *worst case situation is zero,
where best is 10*
2.
Weight
the value (The book uses percentages of 1—and a range table)
a. first step is to rank order
the attributes
b. try using the magick wishes
technique in the book…which attribute would result in the greatest improvement
c. Think about magnitude of
impact, duration of impact, probability of impact, and number of people
impacted.
d. Distribute a 100 points
across the values in a way that it seems to reflect your weights.
e. These weights reflect the
full range changes on the attributes, rather than the attribute itself. Is the difference of 10k a year in
salary….how important is that range.
Ask
our intuition if this ‘feels right’?
3.
Multiply
down and add across