correlations

• a correlation illustrates the strength and direction of an association between 2 or more co - variables
• Correlations cannot infer cause and effect.
• Correlations are shown on scattergrams, which plot one variable against another.

  • Positive correlations suggest that as one variable increases so does the other.
  • Negative correlations suggest that as one variable increases, the other decreases.
• Correlations only show the relationship or association, they do not tell us what is causing it
• • Correlations are measured using Spearman’s Rho or Pearson correlation coefficient analysis.
• examples of correlation - age and memory, no. of hours worked and income etc
• Types of hypothesis when doing correlations

  • directional hypothesis- states the 2 co-variables will be related and the direction of that relationship (previous research)
  • non-directional hypothesis - states that the 2 co-variables will be related but not the direction of that relationship
  • null hypothesis - show no relationship/correlation and that any relationship is due to chance factors/coincidence
• ALWAYS OPERATIONALISE THE CO-VARIABLES
• the difference between correlations and experiments

  • in an experiment the researcher controls or manipulates the IV in order to measure the effect on the DV therefore it is possible to infer that a change in variable is bcs the IV caused any observed changes in the DV
  • in a correlation, there is no manipulation of a variable and therefore is not possible to establish cause and effect btwn one co-variable and another.
  • e.g even if there was a strong correlation btwn caffeine and anxiety levels we cannot assume that caffeine was the cause of the anxiety
• correlation coefficient - a number between -1 and +1 that tells us how closely the co-variables in a correlation analysis are linked

  • can be anywhere on the continuum from -1 to +1. The closer to the value of 1, the stronger the correlation.
  • a weak correlation would have a coefficient closer to the value zero
• • Spearman's Rho is a non-parametric test of correlation that assesses the strength and direction of the relationship between two variables
  • Pearson correlation coefficient is a parametric test.
• Limitations of correlations

  • the inability to establish causation/ can't infer cause and effect (manipulation of IV on the DV)
  • can lack internal/external validity depending on the way variables are measured and whether we can generalise to people etc.
  • the potential for confounding variables
  • the reliance on the strength and direction of the relationship.
• Strengths of correlations

  • secondary data is used so is less time consuming and more economical to carry out
  • they provide a precise and quantifiable measure of how 2 variables are related
  • are more ethical for collecting data
• ways to determine the strength of a correlation

  • calculate the correlation coefficient
  • examine the spread of scores away from the line of best fit on the scattergram
• the type of correlation can be seen on the scattergram by looking at the direction of the line of best fit
• example of + and -

  e.g grade in % against time worked in hours (OPERATIONALISE)

  
• correlation investigates the relationship between 2 co-variables