PYB110 Week Three Revision Notes
Correlation is the extent to which two variables are related. It helps us to examine the relationship
between two variables. There are three different types of variables:
Dependent Variable (DV)
o The response, outcome or criterion variable.
o It is the variable in which we expect to observe a change.
o The values of the DV are dependent on other variables.
Independent Variable (IV)
o The variable which can potentially cause a change in the values of the dependent
o The explanatory or predictor variable.
o The values of the IV are independent of other variables.
Major Types of Research Design
Direct and controlled manipulation of the IV.
Makes a stronger case for cause and effect relationships.
Can control other variables.
Naturalistic or Observational Research
Less control but more naturalistic.
Can determine the strength of the relationship.
Cannot determine cause and effect.
Graphing Variables Using a Scatterplot
One way to understand the relationship between the distributions of two variables is to graph them
together. A convenient way of doing so is to use a scatterplot.
How to Draw a Scatterplot:
1. Draw x (horizontal) and y (vertical) axes.
2. Determine which variable is the IV and the DV.
a. The IV goes on the x axis; the DV goes on the y axis.
b. Make sure the scale you’ve used for your axes will fit the values of the variables.
c. Label your axes.
3. For each pair of observations, mark the point at which the IV and DV meet on the plot.
Patterns of Linear Relationships
As scored rise on the IV, so do scores on the DV.
Direction of the data will rise to the right.
E.g. hours of study positively related to success on exams.
As scores on the IV go up, scores on the DV go down.
Direction of the data will rise to the left.
E.g. number of nights spent partying negatively related to success on exams.
No Relationship Scores on the IV demonstrate no relationship to scores on the DV.
The data typically does not conform to any one direction. May be circular or completely
scattered in distribution.
E.g. hat size has no relationship to success on exams.
Patterns of Relationships
If the points in some way fo