|
Correlational
analysis
In the past
year we’ve seen lots of examples of this. For example whenever
I’ve criticised a study because it doesn’t show cause and effect
it’s probably been a correlational study.
An example:
For example we
could look for a correlation between IQ and performance at GCSE
or A-level. Common sense would perhaps tell us that students
that have higher IQs are more likely to perform well at GCSE.
Correlations
do not have an IV and DV! Nothing is manipulated as with an
experiment. A correlational analysis involves the comparison of
two co-variables. As one increases what happens to the
other? Is there some form of association or relationship?
Types of
correlation
Positive:
the most common; as one variable increases so does the other,
e.g. IQ and GCSE score in the example
above.
Negative:
as one variable increases the other decreases, e.g. it might be
fair to assume that the higher your stress levels the lower your
life expectancy. Again we are unable to show cause and effect.
As mentioned frequently in ‘Stress,’ illnesses could be due to
secondary habits such as smoking, poor diet etc.
|
Advantages of correlations |
Disadvantages of correlations |
|
Correlations allow us to study links between variables
that could not be studied in any other way. We could
not inflict so much stress on a person that we endanger
their life. However, we can use a correlational
analysis to show a possible link between the two
occurring naturally. |
Cause
and effect: do I really need to explain this one? A
correlation shows a possible link between 2 variables it
does not prove that one causes the other, e.g. smoking
and heart disease, early deprivation and later
delinquency etc. |
|
If a
correlation is found a possible cause and effect
relationship can be checked using experimental methods.
No correlation would tend to suggest no such
relationship. |
The
findings of correlations are often misinterpreted.
People tend to assume a cause and effect relationship
when one does not exist. For example 44 thieves and
Bowlby’s assumption that early deprivation was causing
delinquency. This can create unnecessary guilt in
parents for example.
|
|
Economical and fast: large amounts of data can be
compared quickly and cheaply, e.g. by using a
questionnaire to collect data. |
|
Correlations (like experiments) should be easy enough to
replicate so findings can be checked for reliability. |
All
too often there are extraneous variables causing the
link. For example smoking and poor diet may explain the
link between stress and CHD |
|
Scattergrams and correlation coefficients
Correlations are best illustrated using scattergrams
The two
co-variables are plotted (one across the x axis and the other up
the y). Any link can immediately be seen.
A perfect
positive correlation has a coefficient of + 1.0, no correlation
has a coefficient of 0.
In the real
world neither of these extremes usually exists. Coefficients
are lie somewhere between 0 and +1.0 for positive correlations
and between 0 and -1.0 for negative. The nearer 1.0 the higher
the correlation.
The board
expect you to be able to guestimate a correlation coefficient.
Below I’ve included a few examples:
Curvilinear correlations
Correlations
don’t always lie in a straight line. For example when we looked
at the effects of anxiety on recall we saw an inverted ‘U’
shaped correlation with low levels and high levels of anxiety
resulting in lower levels of recall (the Yekes-Dodson law)
Beware of spurious correlations
Sales of ice cream are closely
correlated to drownings in swimming pools and to the number of
shark attacks!
According to QI, in the UK, in
the 20th century hair length was closely correlated to
performance on the Stock Market, as were lengths of women's
skirts!
Clearly there is no obvious
causal factors in any of these, rather third variables causing
both.
Even reputable broadcasters
report what at first glance may appear to be bona fide
relationships such as the correlation between clumsiness in
childhood and later, adult obesity.
Next
page
|
