pinqy
Gold Member
Sure, but look at it from a data collection viewpoint...Each month approx 60,000 households are surveyed (Census goes to many more, but of course there are refusals, empty houses, no longer residences etc). Each household represents anywhere from a few hundred to a few thousand other households. There are dozens of variables for each individual in the survey: age, race, sex, veteran status, education, marital status, disability, etc, and then work or not work, full time or part time, self employed, industry, multiple jobs, unemployed, job search activities, duration of unemployment, reason for unemployment, former industry, why not looking, etc. Changing just one data point can have a serious and unpredicted effect on the aggreagate. And everything has to feed into state and metro area numbers later.I guess it comes to a point of trust here
Lie may have been the wrong word. I deal in numbers and when numbers start stacking up in a place that there is reason one begins to start looking around
That should give you an idea of the huge amount of data and variable involved. Census has one week to collect all 60,000 households. The results are sent to Washington, DC. Analysts there have 9 work days after collection ends to compile everything, request followups and clarification, calculate the aggregate numbers and sort into tables, apply seasonal adjustment, aggregate and/or disaggregate, and write the news release which has to be prepared, edited, send to publications and then GPO, send to the Associate Commisioner and the Commissioner for final review. I think you can appreciate the difficulties and time crunch of that. The evening before Friday release, the report can be given to the President's Council of Economic Advisors, and that's it, by law. The President doesn't even know the numbers until release on 8:30 EST.
There is literally no time available to go back and change things. With all the tables and all the available data, it would be damned hard in any case to make changes because any one change you make effects a lot of other parts.
That's part of it. Honestly I think it's exaggerated. I haven't done a detailed analysis, but while % of the population 65+ is growing, a larger % of the Labor Force is also 65+ So more people are hitting retirement age, but a lower percent are actually retiring.I have heard the changing demographic argument allot also.
It's a statistical issue. The projections and estimates Census was using were off..undercounting teenagers and old people. Since a larger percent of teenagers and 65+ people than the 25-64 population are not in the labor force, it was the Not in the Labor Force category that was most effected. Here's the Jan 2012 news release that explains the changes. Basically 1.5 million people were added to the population because of changes in the estimates...(there had been an undercount and the undercount was corrected). 1.252 million of that happened to be Not in the Labor Force, 216,000 employed, and 42,000 unemployed. All that on top of the recorded changes from the survey from Dec to Jan.I can use the "not in work force number" every day all day and no matter how any-one tries to justify the 1.6 million person spurt, they cannot explain why the other number was not effected the same