1/52. With a six-sided die, the probability of throwing a specific number, for
example a “3,” is 1/6.
Truth of a Causal Law
A legislature or court may invoke a causal or behavioural law when making or
interpreting law. A court may invoke such a law when trying to decide the facts
of a case. In the ideal case such laws would be established to the point of
certainty. Unfortunately behavioural science does not normally yield absolute
certainty about the truth of a behavioural law. Instead, to put it broadly, it can
indicate the probability that it is true based on the experimental methods
used.
237
In areas where there is no scientific research at all or insufficient research, the
existence of truth of causal laws is much more speculative. Ideally this
uncertainty would be captured and given a numerical value. Again, behavioural
science generally cannot do this. One can merely find considered assessment
of why a particular behavioural law may or may not be true.
Deriving Probability
Introduction
In the previous section we considered methods of estimating the probability
that something is true. Once we have estimated a probability, by whatever
method, there are a number of rules of probability that we can apply to the
figure so obtained to derive, calculate or compute further information about
the probability of events. Although the logic behind these computations is
sound (being based on deduction), the results are still no better than the
estimates of probability that we use. So, if an estimate is unsound, the result of
a computation based on this estimate will also be unsound. Indeed, the
computation may magnify the extent or effect of the error.
Complementarity Rule
In measuring the probability of an event happening we are also measuring the
probability that it will not happen. The probability of an event not happening is
easily derived from the probability of its happening because an event can only
happen or not happen. Therefore there is absolute certainty, that is, a certainty
of 100%, that one or other outcome will occur. So, if the probability of an
event happening is 0.3 the probability of it not happening is 1 minus 0.3,
namely 0.7. This is the complementarity rule. In formal terms, if A is an event
with probability P(A), then the probability that A will not occur is [1 - P(A)].
___________________
237
There is a brief description of some of the methods for establishing causal
laws in Chapters 13-15.
 |
 |
 |