Dear Manmeet,
When you conduct a regression analysis in compensation, you are trying to establish the correlation or closeness between two variables. For example, age and salary, tenure and salary, job size and salary, etc.
I have attached a regression sample for job size and salary using Excel for your reference. Such analysis is used:
1. To determine internal equity of the company, i.e., the bigger the job, the higher the salary;
2. To determine the salary spread of jobs within the same job points/grade;
3. To identify outliers, i.e., jobs falling outside the two controlling lines (maximum and minimum); and
4. To identify gaps in grade structure.
This graph is also known as scattergrams. The colored points represent the job incumbents. A job evaluation must be conducted first to determine the job size/points, followed by salary inputs before such analysis can be done.
In this case, the 100% line, also known as the mid-point line, represents the company's salary practice line. When this practice line is compared to the market through participation in a salary survey, you can determine how well you are paying your staff and also identify the gap between your pay practice (where you are today) and pay policy (where you want to be).
The least square regression equation is indicated at the bottom right of the graph. When r2 = 1, it represents a "perfect" situation (although in real life, it will never be a "1" situation). The further r2 is from "1", the more "outliers" there are, causing internal inequity. This internal inequity needs urgent attention as it affects staff morale and impacts recruitment and retention.
Hope this is useful.
Regards, Autumn Jane