I didnt look at the source of the data, nor I am an expert in the field, but I can see many reasons what that can be ill-portrayed. There could be a selection

I didnt look at the source of the data, nor I am an expert in the field, but I can see many reasons what that can be ill-portrayed. There could be a selection bias: does it show how many people switched to a job that turned out to be worse (pay/benefits/quality of work), or people that got fired because they had a much lower tenure than other employees, etc, or the cost of retraining personally for a different position.