Associations

Correlations

Correlation: the degree with which one variable is associated with another variable; interpreted in terms of (a) direction (positive or negative), (b) strength (weak, moderate, strong), and (c) significance (significant or not significant – as indicated by the p value).

• r is the statistical symbol for a correlation.

Partial correlation: the degree with which one variable is associated with another variable over and above these variables’ association with a third variable (i.e., the correlation between two variables minus the variance explained by a third variable).

• r is the statistical symbol for a correlation. In the content of a partial correlation, r is the association between two variables, controlling for the third variable.

Regressions

Regression: this statistical test is used to predict a value of a dependent variable based on the value of at least one independent variable.

• B is the unstandardized regression coefficient; it is interpreted in the units of each independent Variable.
• β is the standardized regression coefficient; it is interpreted in standard deviation units and thus can be used to compare all predictors (independent variables).

Multiple regression: this statistical test is used to predict the value of a dependent variable based on the values of at least 2 independent variables, allowing you to control for additional variables in the model by removing the overlapping variance.

• B is the unstandardized regression coefficient; it is interpreted in the units of the variables.

Interpretation: As the IV increases by one unit, the DV changes by B units.

• β is the standardized regression coefficient; it is interpreted in standard deviation units and thus can be used to compare all predictors (independent variables).

Interpretation: As the IV increases by one SD, the DV changes by b SDs.

• ΔR² is the change in the amount of variance in the dependent variable that is explained by the additional of new variables to the model.
• R² is the amount of variance in the dependent variable that is explained by all variables in a model.