correlations

• correlations:

  • illustrate the strength + direction of association between two or more covariables - represented on x and y axis
  • plotted on scattergrams
• positive correlation:

  • as covariable increases, other increases
• negative correlation:

  • as one covariable increases, other decreases
• zero correlation:

  • increase in 1 variable is not associated with change in other variable
• linear + curvilinear correlation:

  • positive + negative correlations are considered linear = straight line
  • curvilinear occurs when relationship is still predictable but curved e.g yerkes dodson law
• correlation coefficient:

  • used to measure extent of correlation between covariables
• correlation coefficient pt2:

  • maximum value of 1 - +1 = positive correlation, -1 = perfect negative correlation
  • +/- shows whether positive or negative
  • number tells us how closely covariables are related
• tables of significance + correlation:

  • tells us how big the coefficient needs to be in order for correlation to be significant
• strengths of correlations pt 1:

  • useful for investigating trends in data
  • if correlation is significance , further research = justified
  • often used as starting point to assess patterns before researchers commit to experimental data
• strengths of correlations pt 2:

  • relatively quick + economical to conduct
  • no need for controlled environment, no manipulation of variables
  • secondary data can also be used
• weakness of correlations:

  • cannot demonstrate cause + effect
  • dont know which covariables cause other to change
  • untested third variable (intervening variable) could be causing relationship
  • can be misused or misinterpreted - particularly in media