Correlations

Non-Experimental Research Methods

1. Correlations

• Definition: Establishes whether there is a statistical relationship between two variables.

• Example:

  • Research Question: Is there a relationship between the amount of time a student studies and the score received on a quiz?

  • Variables:

    • Number of minutes studied

    • Quiz scores

  • Method: Administer a quiz and ask students to report study duration, then record the scores.

• Types of Correlational Relationships:

  1. Positive Correlation

  2. Negative Correlation

  3. Zero Correlation

A. Positive Correlation

• Definition: An increase in the value of one variable corresponds with an increase in the other variable.

• Visual Representation: Scatterplot

  • Example: As minutes studied increases, quiz scores also increase.

  • Important Note: Causation cannot be assumed. A correlation does not imply that increased study time causes a higher score.

B. Negative Correlation

• Definition: An increase in the value of one variable results in a decrease in the value of the other variable.

• Visual Representation: Scatterplot

  • Example: Variables include 'Number of Hours of Netflix Watched' and 'Test Grades.' More Netflix viewing is correlated with lower test grades.

  • Important Note: Again, causation cannot be determined based on correlation alone.

C. Zero Correlation

• Definition: No relationship exists between the two variables.

• Visual Representation: Scatterplot

  • Example: Comparing 'Number of Pounds of Candy Eaten' and 'Test Grades,' showing no correlation.

When to Use Correlation as Research Methodology

• Appropriate Scenarios: a. In the early stages of research to gather data. b. When variable manipulation is impossible or unethical. c. When relating two naturally occurring variables.

• Additional Insights:

  1. Direction:

     • Positive or Negative indicated by scatterplot.

  2. Magnitude:

     • Strength of relationship indicated by the absolute value of the correlation coefficient.

     • Values range from –1.00 to +1.00.

     • Example Scale:

       • Strong: |0.80| - |1.00|

       • Moderate: |0.40| - |0.60|

       • Weak: |0.10| - |0.20|

Limitations of Correlational Research

• Important Considerations: A. Absence of independent variables:

  • Causal relationships require at least one variable to be manipulated.B. Third Variable Problem:

  • Possibility exists that an unmeasured variable may influence the relationship, leading to incorrect assumptions of causation.

  • The danger is mistaking correlation for causation when a third variable might be influencing the two main variables.