correlations

what is a correlation?

measurement of how strongly two or more variables are related

the variables are called co-variables, no cause-and-effect established as there is no IV

types of correlation

positive - both co-variables increase or decrease

negative - one variable increases and the other decreases

no - no relationship between them

hypotheses are the same as experiments

correlation coefficients

shows the strength and direction of a correlation - +1 is perfect positive and -1 is perfect negative

scatter grams are used to display data - one variable on each axis

sometimes there are curvilinear relationships - more complex

pros of correlations

precise, quantifiable measure of relationship

quick and economical to do

suggest patterns that lead to experiments

used to study sensitive topics

allow us to study naturally occurring variables

cons of correlations

cannot demonstrate cause and effect so we don’t know what is changing what - bidirectional ambiguity

another variable may cause the relationship between the two co-variables - intervening variable

correlations can be easily misinterpreted