correlational observation

• what is a correlation?

  a descriptive statistical technique that measures the relationship/association between two+ co-variables to see if a trend/pattern exists between them
• what are co-variables?

  the two measured variables in a correlational analysis. the bariables must be continuous (ordinal/interval data)
• what are the 4 types of correlation?

  positive

  negative

  none

  curvilinear
• positive correlation

  • relationship between 2 variables
  • co-variables move in same direction
  • as the value of one variable increases, the other does too
  • e.g. more hours studying, and higher grades
• negative correlation

  • relationship between 2 variables
  • co-variables move in opposite directions
  • as the value of one variable increases, the other decreases
  • e.g. more hours studying for an exam, and stress levels you feel on the day
• no correlation

  • no relationship between co-variables
  • e.g. a persons eye colour and exam performance
• curvilinear correlation

  • non-linear relationship between co-variables
  • e.g. Yerkes-Dodson law
• what is correlation co-efficient?

  a number between -1 and +1 that tells us how closely related co-variables in a correlational analysis are
• scale of correlation coefficient

  0<r≤0.19 very low correlation

  0.2≤r≤0.39 low correlation

  0.4≤r≤0.59 moderate correlation

  0.6≤r≤0.79 high correlation

  0.8≤r≤1.0 very high correlation
• scatter diagram

  • correlations illustrated using a scatter diagram
  • scatter dots indicate degree of correlation between two co-variables
• strengths

  • procedures usually easily replicated again, meaning findings can be tested for external reliability
  • if correlation signif, further investigation can be justified. if correlation not signif, you can rule out casual relationship
  • can be used to study naturally occurring variables its unethical to manipulate (e.g. relationship between smokers and lung cancer)
• weaknesses

  • commonly misinterpreted. cause & effect relationship CANNOT be shown, we can only say there's relationship between 2 co-variables
  • other variables may be responsible for affect on co-variables (intervening variables)
  • lacks internal & external validity
  • sample used biased & not representative of gen pop = therefore no have pop validity
• what are intervening variables?

  another variable that may be responsible for the affect on co-variables