This seems like they developed their conclusion then went searching for data to back it up. For example they only "considered only separations that result in at least six months of unemployment or at least a 50 percent drop in earnings from pre-separation levels." So anyone who found a new job before resigning their position isn't counted, anyone who switched jobs to one with more flexible hours but took less than a 50% pay cut isn't counted. They also didn't compare this number for people both over and under 50 to see if age had any relevance at all.
Comments
This seems like they developed their conclusion then went searching for data to back it up. For example they only "considered only separations that result in at least six months of unemployment or at least a 50 percent drop in earnings from pre-separation levels." So anyone who found a new job before resigning their position isn't counted, anyone who switched jobs to one with more flexible hours but took less than a 50% pay cut isn't counted. They also didn't compare this number for people both over and under 50 to see if age had any relevance at all.