Consumer Behavior Research Methods

Descriptive Statistics

Bivariate statistical analysis Correlation indexes | Covariance

You can use the covariance to determine the direction of a linear relationship between two variables as follows:
 
  • Covariance
    • Cov (x,y) > 0 --> If x and y tend to increase or decrease together, the coefficient is positive

    • Cov (x,y) < 0 --> If one variable tends to increase as the other decreases or vice versa, the coefficient is negative

    • Cov (x,y) = 0 --> discordance between x and y

It is an absolute index: it indicates the existence and direction of the relation between two variables, but not the strength of their relation
 

Diskussion