Correlation

Correlation

1. Overview

   • Measures the relationship between 2 variables
   • The relationship between variables is shown on a scattergram
   • Statistical tests are carried out to produce a correlation co-efficient.
     • Always has a value between +1 and -1
     • Indicates both strength and direction of the relationship
       • The number = strength. (0 -= weakest relationship).
       • Positive and negative shows direction.

2. Positive correlation

   • When one variable value increases so does the other.
   • Perfect positive correlation would be +1
     • It's when variables increase in exact relationship to one another.

3. Negative correlation.

   • One of the variable increases s the other decreases.
   • Perfect negative correlation co -efficient would be -1.
     • A variable increases at exactly the same rate the other decreases.

4. Zero correlation

   • There's no relationship between variables.
   • Zero correlation has a co efficient of 0.
     • When there's no pattern between variables.

5. Limitations

   • Don't demonstrate cause and effect
   • Hypothesis for a correlation study does not predict will affect one variable or another it predicts their relationship.
   • Lack of control over other variables causes problems in studies.
     • Other factors could account for relationships.
   • Relationship may be due to coincidence.
     • No meaningful relationship.
   • Co efficient of 0 may be due to curvilinear relationship.

6. Strengths

   • On technique to show strength and direction
   • Statistically analyse natural phenomenon using experiments due to ethics and practicality
   • Once relationship is known, other types of studies can investigate further.
   • Study: Ohnishi et al (2000)
     • Positive correlation between impairment in communication and social interaction and the blood flow to frontal cortex.

Images and files in the map