While this latter part of the discussion has focused on the procedure as the
cure, the continued and invariant success of the procedure also suggests that
the reasoning underlying it is correct. That is, Substance Y is the continuing
cause of Disease X. However, as would be clear from the reasoning so far,
this is not established conclusively although there is a lot or supportive
evidence. Moreover, in real life the hypothesis that Substance Y is the
continuing cause of Disease X will not be idle speculation but usually have
some good sense to it as well.
That said, it is still possible that Substance Y is not the continuing cause of
Disease X. The real truth may be that something else is the cause. What
happens is that the procedure that removes Substance Y at the same time
removes the real cause of Disease X.
Experimental Method
Experiments test proposed causal laws. Researchers set up a situation where
the causal law might operate and see if it does. In ideal circumstances, which
scientists aim for, an experiment is conducted under conditions that are as
controlled as human endeavour can make them. In the experiment, the scientist
want to test a supposed causal law, for example that X causes Y. This is
referred to as
the hypothesis. For example, the hypothesis might be that an
increase in temperature upward from 20 degrees centigrade negatively affects
the capacity of a person to perform mental arithmetic. To test this the core
method might involve testing subjects in
five stages. The first stage tests
subjects on mental arithmetic at a temperature of 20 degrees. Then the
temperature is raised on four occasions by a specific amount, say 3 degrees,
and a fresh test in mental arithmetic is administered at each new level of
temperature.
404
Here X is known as the independent variable. This is because the experimenter
determines the level or degree of X before the experiment begins – therefore
the degree or level of X is independent of anything that happens during the
experiment. Y is known is the dependent variable because the value of Y will
depend on X if there is a causal relationship between X and Y.
___________________
404
For the sake of simplicity in conveying basic notions this account avoids
some complications in and details about experimental design, especiall those
needed to eliminate or reduce confounding variables. (i) It would be necessary to
ensure that the questions in each test were of equal difficulty. If this was not the
case, and the questions became more difficult, the increase in difficulty presents
an alternative explanation for a lowering of performance. (ii) As a subject does
more and more mental arithmetic some other variables may occur. (a) They may
“warm up” as it were and just naturally get better at it. (b) Their performance may
decline as they become bored or tired.
 |
 |
 |