Correlation

• Correlation


  A research method that looks at the relationship or association between two variables rather than a difference between conditions
• Types of correlations


  • Positive correlation
  • Negative correlation
  • No correlation
• Positive correlation


  Both co-variables increase/decrease together
• Negative correlation


  As one co-variable increases, the other decreases
• No correlation


  There is no relationship between the two co-variables
• Investigator effects


  • The way an interviewer asks a question or unconsciously communicates expectations to participants by smiling or frowning at certain points
• Minimising investigator effects


  1. Provide a standardised script for the interviewer to use so that they all ask the same questions in the same way to avoid any bias in participant responses

  2. Interviewers can be trained to greet participants in the same way and ask questions with a neutral tone
• Correlational research


  A research method that looks at the relationship or association between two variables rather than a difference between conditions (experiments)
• Co-variables


  Variables in a correlational study, rather than an independent and dependent variable
• Types of correlations


  • Positive correlation - both co-variables increase/decrease together
  • Negative correlation - as one co-variable increases the other decreases
• Correlation coefficient


  A number calculated to see how strongly related the two variables are, ranging from -1 to +1
• Correlational hypotheses


  • Non-directional - There will be a correlation/relationship between variable 1 and variable 2
  • Directional - There will be a positive/negative correlation/relationship between variable 1 and variable 2
  • Null - There will be no correlation between variable 1 and variable 2
• Strengths of correlational research

  • Can look at situations which cannot be manipulated experimentally due to ethics
• Weaknesses of correlational research

  • It is only a relationship therefore cause and effect cannot be established, there may be other variables involved
  • Correlation does not equal causation