>Current Unemployment Statistics are the percentage of the population
>registered with the unemployment office and actively looking for work.
>These statistics do not include those who have been unemployed for a
>certain length of time. (I dont remember the exact length of time- maybe
>someone else on the list does- I think it was something like 3 months)
>It does not include those who have never worked or are otherwise outside
>the 'workforce'. It also does not include 'illegal' immigrants and
>others who do not work in the 'legal' economy, such as those who had
>been working 'under-the-table' jobs. Another thing that is not counted
>is underemployment.
Underemployment is very difficult to define, much less measure. If a violinist can't find a job as a fiddler, but only as a cab driver, is s/he underemployed? Is the answer the same if s/he is a bad violinist as a good one?
Almost everything else in the paragraph is wrong. The monthly U.S. unemployment numbers are based on a survey of appx 60,000 households. Respondents are asked if they were working during the week containing the 12th of the month. Being employed just one hour counts as a yes. If they say they weren't working, they're asked if they were actively looking for work. If the answer is yes, then they're counted as unemployed. If the answer is no, then they're counted as not in the labor force. There's no time limit. You can be jobless for 10 years, but as long as you're looking for work (e.g., reading the want ads, sending out resumes, etc.), then you're counted as unemployed. Illegals are not officially excluded from the survey - they're just not likely to play along with a survey for obvious reasons. If people are working off the books, and they answer yes to the employment question, then they're counted. The Census/BLS surveyors don't check up on where people work.
The monthly employment survey started in the late 1940s. I don't know how the numbers for earlier years were assembled, and the BLS's website is down.
Doug