Correlational studies

• Definition

  A non-experimental method that measure the strength and direction of relationship or link between two co-variables, these can be positive or negative.
• Main features of correlation studies

  • Involves measuring two or more co-variables
  • Does not involve an IV or DV  
  • Can be quickly used to analyse relationships in large amounts of data 
  • Identifies the direction of relationship – positive, none or negative
  • Identifies the strength of relationship – weak, moderate or strong.
  • Correlation coefficient can be calculated to identify strength and direction of correlation 
• Correlation coefficient

  • This is a way of measuring how strongly two variables are correlated.
  • Correlation coefficient is a statistic ranging from +1 to -1 which indicates the strength (the number) and direction (+ or -) of a correlation between two co-variables.
  •  – 1 being a perfect negative correlation, 
  • 0 being no correlation and 
  • +1 being a perfect positive correlation.
• Strengths of correlation studies

  • Further research could be triggered - if a relationship is found between the two variables could be suggested that we can predict one variable may impact on the other - therefore this may indicate that further experimental research into the topic is justified.
  • Good for certain kinds of studies - they can be used when it would be unethical /impractical to manipulate variables - can make use of existing data.
  • Can be tested for reliability as they are easily repeatable to check for consistent findings into the relationship between the two variables being tested.
• Limitations of correlation studies

  • Difficult to establish cause+effect between variables as only relationship is found between the two variables - other EV’s responsible - lowers internal validity.
  • Misinterpreted by the media + society when a link found - may assume conclusion can be made about causes for relationship, misused by the public to support/contradict an argument.
  • May lack validity - Internal validity - problems with the method used to measure behavior is legitimate - External validity - sample used may be small + unrepresentative, results can't be generalized to target population.