Vous êtes sur la page 1sur 2


When you need to display what happens to one variable when another variable changes
in order to test a theory that the two variables are related. The scatter diagram is used
to test the possible cause and effect relationship. Scatter diagram can not prove that
one variable causes the other but it makes it clear whether a relationship between two
variables exists or not and predicts the strength of the relationship.
It is set up between variable number 1 on x axis & another
Typical Scatter variable on y axis. Notice how the plotted points form a
Diagram cluster pattern. The direction and tightness of the cluster
provides the clue about the strength of the relationship.

The more that this cluster resembles a straight line, the

stronger the correlation between the variables. This makes
sense since a straight line would mean that every time one
variable would change the other would change by the same
Some tips for the scatter diagram
1. A negative relationship (as y increases, x decreases) is
as important as positive relationship (as y increases, x
Look for cluster increases).
1. Direction. 2. You can only say that x and y are related but not that
2. Tightness. one causes the other.
3. There are statistical tests available to test the exact
degree of correlation but not covered in this.

Representative diagram Correlation Analysis

An increase in y may depend on
increase in x. If we control x, there is
chance that y can be controlled. e.g.
Education Vs Performance.

If x is increased, y may increase

somewhat but y seems to have
causes other than x.

There may be no correlation. Y is not

related to x.

An increase in x may cause a

Possible tendency for a decrease in y. e.g.
Negative Education Vs Rejects, Quality Vs
Customer complaints.

An increase in x may cause a

Negative decrease in y. If we control y, there is
chance that x can be controlled.