Correlations

• what is a correlation?

  asseses the relationship between 2 variables.
• what is a positive correlation?

  • the variables change in the same direction.
  • as one increases, the other increases.
  • as one decreases, the other decreases.
• what is a negative correlation?

  • the variables change in opposite directions.
  • as one increases, the other decreases.
• what is no/zero correlation?

  no relationships found.
• what is a coefficient correlation?

  • if you were reporting your research, you couldn't just say "the line slopes to the right a bit" or, "it looks positive".
  • there is a mathematic value for any correlation.
• what are the correlation coefficients? (mathematic values for correlations)

  +/- = positive/negative

  • +/- 0.7-0.9 = strong
  • +/- 0.4-0.6 = moderate
  • +/- 0.1-0.3 = weak

  0 = none
• more info about correlation coefficients:

  • a value of 10 is a perfect, positive correlation.
  • if there is a 100% correspondence between the 2 variables then the correlation would be +10.
  • a value of -10 means a perfect, negative correlation, e.g. the more money you spend, the less there is in the bank. there could be a 100% correlation between spending & saving.
• what is an imperfect positive correlation?

  • when two variables tend to move in the same direction, but the relationship isn't perfectly linear. 
  • this means that as one variable increases, the other tends to increase as well, but the relationship is not a straight line
• what is an imperfect negative correlation?

  describes a relationship between two variables where, as one variable increases, the other tends to decrease, but the relationship isn't perfectly linear or consistent.
• what are some strengths of correlations?

  • can see connections between 2 issues that we will not be able to experiment on.
  • don't require manipulation of variables, more ethical than experimenting on humans.
• what are some weaknesses of correlations?

  • don't prove a casual relationship.
  • don't reflect curvilinear (non-linear) relationships.
  • the relationship between two variables could be caused by a fluke or influence of a 3rd variable.
  • relationship vs causation (is it actually that one variable causes the other) spurious correlation = the two may not be related at all.