An intervening variable in statistics and psychology is a variable that acts as the link between the independent and dependent variables. The independent variable causes a change in the intervening variable, which in turn causes a change in the dependent variable.
The intervening variable is also called the mediating variable. This is because as we saw above it acts as the casual link, that is, it “mediates” between the dependent and the independent variables. The intervening variable is generally hard to quantify and is hence not usually taken to be part of our statistical model.
Examples of intervening variables:
Suppose we are interested in the relationship between the following two variables – “Time spent in studying for the exam” and “Marks obtained in a test”. Here the time spent studying is the independent variable. If more time is spent studying then better marks are obtained. Hence, the mark obtained is the dependent variable.
Here the intervening variable is – “The gain in understanding of the subject”. This is because if we spend more time studying, we gain a better understanding of the subject which in turn causes us to obtain better marks in the exam.
Notice that in the above example the independent and the dependent variables are easy to quantify and measure whereas it is hard to quantify the intervening variable. It is difficult to assign a numerical value to a person’s understanding of a subject.
The above observation is also true in general. The independent and dependent variables are quantifiable whereas the intervening variable is qualitative.
How to identify intervening variables?
There is no systematic method to follow in order to identify intervening variables. The researcher must use his knowledge of the situation under study to try and identify what might be acting as the causal link between the two variables.
Why are intervening variables important?
The intervening variables are important to study because they improve our knowledge of the causes that affect the dependent variables. Hence by controlling the intervening variable we can obtain control over the dependent variable.
The intervening variables are of less significance in statistical models where they are not included because of the difficulty in measuring them.