4.6 Critical thinking and science 163
4.6 Critical thinking and science
Science is a highly disciplined form of critical
thinking. This is not surprising, since science is
a methodology that is reliant upon evidence, in
particular the evidence provided by
observation and experiment. Scientists make
observations and use them both to construct
and to test their theories. A scientific theory is
only as good as the evidence on which it is
based and the reasoning by which scientists
proceed in drawing their conclusions. All that
has been said about not leaping to conclusions,
or making unwarranted assumptions, applies
with particular relevance to science.
An observation in scientific terms is any fact
that can be verified by experience: for example,
evidence of the senses. It means more than just
visual data. If I suddenly sense the ground
trembling beneath my feet, or hear a rumbling
sound, or see a cup fall off a shelf, these are all
observations. I may not know what has caused
them: they may be indicative of an earthquake,
or just a heavy vehicle passing on the road, or a
controlled explosion in a nearby quarry.
Without further evidence I have no way of
inferring which, if any of them, is the correct
interpretation. But the experience itself – the
observation or sensation – remains the same
whatever its cause turns out to be.
Of course, people can be mistaken about
what they experience. We sometimes imagine
things, or misremember them. A reliable
scientific observation is therefore one which
cannot be dismissed easily. If many people
describe having had the same experience at the
same time, that is better evidence than one
person’s word. The term we use for this, as
introduced in Chapter 4.3, is ‘corroboration’.
Observations may be even more trustworthy if
they are detected and recorded by instruments
or sensors. Moreover, instruments can often
pick up information that human senses cannot
detect. They can make measurements of things
where humans can only estimate crudely. A
seismometer, for instance, is a device for
measuring earth tremors. It can give accurate
readings of movements far below the ground
that no human would notice or find significant.
Such readings are also ‘observations’.
If they are made accurately, these are facts; but
without accuracy they remain observations.
Their importance, scientifically, lies in the use
they can be put to as evidence for hypotheses
or predictions: for example, the causes of
earthquakes, or the risk of earthquakes in a
given region. For such purposes single
observations are rarely sufficient for
establishing conclusions. A large part of
scientific inquiry therefore involves the
analysis of collections of data to identify
patterns and correlations. Observations on
their own can be thought of as ‘raw’ data. To
function as evidence this raw data generally
has to be collated and interpreted, often in the
form of tables, graphs, reports and so on. A
critical question therefore arises as to whether
the processed data is fair and objective, or
whether it distorts the facts in one direction or
another. For instance, if the observation
concerns a sample of data, is it a representative
sample; or is it selective, exaggerated, biased or
misleading in any way?