|
My point was that they weren't comparing people based on equivalent experience or equivalent jobs. They were just averaging.
Example:
All people are college grad women:
#1 - graduated 25 years ago and is at top salary - 120K
#2 - graduated last week and is in social work - 30K
#3 - just came back from taking 4 years off to raise kids 40K
#4 - works part time and raises kids 20K
White average is 52.5K
#1 - has two jobs (possible discrimination here - doesn't distinguish in terms of how much each job pays - only lists total amount earned) - 80K
#2 - graduated 25 years ago and is at top salary (discrimination here) - 110K
#3 - just came back from taking 6 months off to raise kids 55K
#4 - graduated last week and is in social work - 30K
Black average 68.75
See my point by averaging this way and not controlling for variables you can "hide" discrimination or get misleading results.
Note: this is an example I made up. Do not jump down my throat about where I got the numbers. It is just an example. Chill.
P.S. If I am wrong and this is not how it was done feel free to correct. It was just my interpretation from reading.
|