SOCSCI RESEARCH CONCEPTS 3: VARIABLE RELATIONSHIPS
1.
Philosophers
have long debated the concept of causality. For instance, Hume said if one
billiard ball hits another, we cannot prove that the first caused the second to
fall into a pocket. Hume argued that we can only observe sequence of events.
2.
Nevertheless,
we say that X causes Y if:
2.1.
There
is a relationship between X and Y
2.2.
Relationship
is asymmetrical: change in X results in change Y but not vice-versa
2.3.
A
change in X results in a change in Y regardless of the actions of other factors
3.
At
the same time, Bailey says that it possible to define symmetrical
relationships. This is the “so-called mutual causation” in which X causes Y and
simultaneously Y and X so that each is both a cause and effect.
4.
Bailey
also argues that it is useful to think of causality in terms of necessary and sufficient conditions. Cause X
is necessary for effect Y if Y never occurs unless X occurs. X is sufficient
for effect Y if Y occurs every time X occurs. There are 3 combinations: (a) X
is necessary but not a sufficient condition for Y; (b) X is sufficient but not
necessary; and (c) X is both necessary and sufficient for effect Y.
1.
NO
CORRELATION
2.
LINEAR
VERSUS NONLINEAR (CURVILINEAR) RELATIONSHIPS
3.
SPURIOUS
AND INTERVENING RELATIONSHIPS
a)
A
spurious relationship exists when two variables A and B appear to be related
only because both are caused by a third variable C.
b)
Often,
what appears to be a relationship between two variables A and B is caused by an
intervening variable C that is literally between A and B. C is the intervening
variable and the relationship is called an intervening variable relationship.
4.
SUPPRESSOR
AND DISTORTER VARIABLES
a)
Two
variables that are actually related may appear to be unrelated because each is
correlated with a third variable that suppresses the relationship between them.
b)
The
third variable may also have the effect of distorting the relationship between
two variables, making the relationship positive or negative instead of the
other way around.
Prof. Art Boquiren January 2004 based on Bailey
1994: 48-52