BCGInstitute: Serving the HR Community

I'm in the process of determining factors that affect compensation at our company. Since our company is not too large (just over 500 employees), our compensation system isn't very sophisticated. Still, I want to ensure I build a statistical model for anlaysis that includes all relevant factors.

Please post factors that could affect compensation so that we can develop a longer list of potential compensation factors. Hopefully the list will help others who are tackling a similar task.

We can start the list with the factors that OFCCP requests for an audit (yes, they also request race and sex but those better not be factors that help determine compensation!):
– Job title
– Location (if different from facility under review)
– Annual(ized) salary
– Grade/Band/SSEG
– Time at the company
– Time in the job
– Exempt/Nonexempt status
– Part-time/full-time status
– Previous experience (age as a proxy)

Other factors that could affect compensation include:
- Business Unit
- Skills/qualifications
- Compensation at previous employer
- Degrees/Certifications
- Compensation Status (hourly vs. salaried)
- Performance
- Time at company (not dependent upon current position)

Please add to the list by posting a reply. Thanks!

PS - I'll create a separate post to ask how others quantify performance so that it can be analyzed. Combining these two topics in one thread might be too confusing.

This is a great question that we field often and my response is, unfortunately, not simply an itemized list of variables to include in your analysis.

All companies differ in their compensation practice. Consequently, analyzing a laundry list of explanatory compensation factors and seeing which sticks, and iteratively refining it down to an acceptable model is not the most appropriate approach. In practice, this is sometimes referred to as a "datamining" approach.

The most valid list of explanatory compensation factors is a reflection of your company's compensation policy and practice. To obtain this list, what we've done is to interview HR compensation dept and ask for the factors that go into determining pay and promotions.

Through this exercise, you get a list of compensation factors that reflects what "should" be done in determining pay. An apriori approach to identifying legitimate compensation factors prior running an analysis is more diagnostic and imminently more scientifically defensible and robust. I strongly recommend this approach.

Datamining and sifting through a myriad of possible explanatory factors in building a compensation model may result in misleading findings. Unfortunately, I see this all too often.

I wish there was an easy answer, but there's no shortcuts at this critical stage of your comp analysis.

Again, great question!

best,
dk

Dan Kuang, Ph.D.
Principal Consultant
Biddle Consulting Group, Inc.
193 Blue Ravine Road, Suite 270
Folsom, California 95630
916.294.4250x145 (w) 916.294.4255 (f)
DKuang@biddle.com
www.biddle.com - www.opac.com - www.criticall911.com