correlations

• What is a correlation?

  A type of research method which investigates the relationship between two co-variables
• What are three types of correlation?

  Positive, negative and no correlation
• What is a positive correlation?

  As one variable increases, so does the other
• What is a negative correlation?

  As one variable increases, the other decreases
• What is no correlation?

  No relationship between the co-variables
• What is the difference between an experiment & correlation?

  An experiment has an independent variable whereas a correlation only measures the relationship between two co-variables
• What is a correlation co-efficient?

  A single number between +1 & -1 that represents the strength & direction of the correlation. Represented by the letter "r"
• What sort of relationship is suggested by r = -0.4?

  Weak negative
• What sort of relationship is suggested by r = +0.9?

  Very strong positive
• What sort of relationship is suggested by r = +0.13?

  Very weak positive
• What sort of relationship is suggested by r = -0.76?

  Strong negative
• Identify a strength of correlations?

  Quick and cheap
• Why are correlations quick and cheap?

  No manipulation of variables is required & no need for a controlled environment
• Why is it useful the correlations are quick and cheap?

  Cheaper researcher means that more correlations can be conducted for less cost
• Identify a second strength of correlations?

  Correlations are useful as preliminary research
• How can correlations be used as preliminary research?

  A correlation can indicate how two variables are related, this can provide ideas for future research.
• Why is it useful that correlations can be used as preliminary research?

  Because correlations are cheap, they can see if two variables are related before wasting money on an experiment.
• Why can correlations not establish cause and effect?

  It cannot establish cause and effect because a correlation does not manipulate or control any variables.
• Why is correlations not establishing cause and effect a problem?

  Correlations are limited in their usefulness.
• How are correlations open to the "third variable problem"?


  There may be another variable altogether which hasn't been measured that is causing the relationship.
• Why is the "third variable problem" a problem?

  Correlations can be misleading.