Correlations

• Define correlation?


  It is a mathematical technique used to investigate an association between 2 vairables (co-variables)
• how do correlations differ to experiments?

  • variables are measured, not manipulated like in experiments
  • Only an association is found, not a cause and effect relationship (thus why terms DV & IV are not used)
• What are corrleation co efficients?

  They determine the strength and relationship between co- variables
• Describe a negative correlation?


  When one varible increases the other decreases, it has a negative gradient
• What is the corrleation co effiecient of a negaitive correlation?


  Less than 0
• What graph are all correltions presented on?

  A scattergram
• Describe a positive correlation?


  When one variable increases the other also increases, it has a postive gradient
• What corelation co effiecient does a postive correlation have?


  More than 0
• Describe zero correlation?


  Where no relationship is found between the co vairables, no line of best fit can be drawn as the points on the scattergrpah are random
• what is the correlation coeffiecient of a zero correlation?


  Equal to 0
• Explain a curve linear relationship?


  As one variable increases, so dones the other but only up to a certain point after which one variable continues to increase the other beings to decrease
• What shape does a curvelinear relationship form on a graph?

  an inverted U shape
• Give an example of a curevelinear relationship?

  Yerkes-Dodson Law, showing how anxiety affects eye witness testimony
• What are some strengths of correlations?

  • Useful preliminary tool as a starting point to assess patterns between co variables before researchers commit to conducting an experimental study, increasing the reliability of the findings
  • Quick and economical to carry out- no need for a controlled environment & no manipulation of variables is necessary, therefore cost effective so a more appropriate method
• What can be used to make correlations even less time consuming?

  Secondary data (such as government statistics) can be used in the study
• What are some limiations of correlations?


  • Cannot demonstarte a cause and effect relationship between 2 variables, establishing the direction of the effect is an issue as we do not know which co-variable is causing the other to change. Therefore they may be misread reducing the validity of any conclusions drawn
• What is the 3rd variable problem?


  An intervening variable might be responsible for the relationship between the co-variables, in which the researcher is unaware of.
• Why is the 3rd variable an issue for establishing correlations?


  As we cannot be certain that any correlations found are measuring the relationship intending to be measured which furhter increases the likelihood of corrleational data being misinterpreted. Therefore reducing the internal validity of correlations
• What is another limitation of correlations?


  That correlations tend to be misused/interpreted when made public by the media, correlation is aoften presented as a causation