Partial Correlation is the measure of association between two variables, while controlling or adjusting the effect of one or more additional variables. Partial Correlation can be used in many cases, like whether or not the sale value of a particular commodity is strongly related to the expenditure on advertising when the effect of price is controlled. If the partial correlation becomes zero, then it can be inferred that the correlation that was computed before is false.
In SPSS, partial correlation can be computed by selecting “correlate” from the analyze menu, and then selecting “partial” from the correlate.
There are certain terminologies in partial correlation that will help in understanding partial correlation in a much better manner:
The control variables in partial correlation are the variables which extract the variance which is obtained from the initial correlated variables.
The order of correlation in partial correlation refers to the correlation with control variables. For example, first order partial correlation is the one which has a single control variable.
The spurious correlation in partial correlation refers to that type of correlation that is false or the correlation that actually didn’t exist.
Part Correlation in partial correlation is also called semi partial correlation. It is a measure of correlation of the dependent variable (Y) and independent variable (X), when the linear effects of the other independent variables have been removed from X but not from Y.
For small models like models with one control variable or sometimes with two or three, partial correlation is useful and is quite common. Partial correlation is generally helpful in detecting a false relationship model.
Assumptions
Partial Correlation is useful in only small models like the models which involve three or four variables.
Partial Correlation is used in only those models which assume a linear relationship.
For partial correlation, the data is supposed to be interval in nature.
The residual variables or unmeasured variables in partial correlation are not correlated with any of the variables in the model, except for the one for which these residuals have occurred.
Multicollinearity is a statistical phenomenon in which two or more independent variables in a multiple regression model are highly correlated. In partial correlation, multicollinearity should be low; otherwise the calculations regarding individual predictors will be affected.
