They only count those who are signing up for unemployment for the first time. Farm workers, self employed, college graduates, those already unemployed, and those who ran out of unemployment insurance aren't counted.
In other words, things are a LOT worse than 9.1%. (9.1% only represents the new ones.)
Yeah, I think the answer is in there. It still grates me that we need to create something like 350,000 new jobs each month, to take care of the new pool of workers (someone adjust the number if need be), and we’re not even coming close, yet the number remains pegged at 9.1%. This method seems absolutely worthless to me.
As you state, there’s a massive number of people who have dropped out. The number of idle people is grown larger each month.
I would rather see what the pool of working age people is, discount those that are unable to work, and compare how many of them are employed. I’m sure there would be problems with that too, but it would seem far more accurate.