# Where do the government's job numbers come from? **URL:** \ **Category:** Factual Questions\ **Created:** [August 5, 2002, 8:29am UTC](https://boards.straightdope.com/t/where-do-the-governments-job-numbers-come-from/122408 "2002-08-05T08:29:37Z")\ **Posts on this page:** 9\ **Page:** 1
**Author:** ![Stoid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/stoid/32/272_2.png) [@Stoid](https://boards.straightdope.com/u/Stoid)\ **Post date:** [August 5, 2002, 8:29am UTC](https://boards.straightdope.com/t/where-do-the-governments-job-numbers-come-from/122408/1 "2002-08-05T08:29:37Z")
Because it seems to me such a thing is really hard to determine. For instance, this quote: > [@](#): > > * * * > > The worrisome part of the employment report on Friday was that the economy added a paltry 6,000 jobs in July. > > \*\* Huh? How do they know that? They are asserting that the WHOLE ECONOMY added a mere 6000 jobs. They know every job that was added? Is this taken from tax information, or somewhere else? It seems particularly suspect because the info is so fresh. Anyone? ---
**Author:** ![nightshadea](https://avatars.discourse-cdn.com/v4/letter/n/7ba0ec/32.png) [@nightshadea](https://boards.straightdope.com/u/nightshadea)\ **Post date:** [August 5, 2002, 9:29am UTC](https://boards.straightdope.com/t/where-do-the-governments-job-numbers-come-from/122408/2 "2002-08-05T09:29:34Z")
Well I dont know but one assumes all they had to do is check for new w2 form filings for new jobs But theres probably more than one way to do it ---
**Author:** ![Coil](https://avatars.discourse-cdn.com/v4/letter/c/bb73d2/32.png) [@Coil](https://boards.straightdope.com/u/Coil)\ **Post date:** [August 5, 2002, 10:33am UTC](https://boards.straightdope.com/t/where-do-the-governments-job-numbers-come-from/122408/3 "2002-08-05T10:33:37Z")
While I do not know how it is done in the US I know at least something about how it is done in my country (Sweden). Here we basically have two types of statistics on employment, survey based and register based. The statistics based on information from registers is the most accurate but it is significantly slower than the survey based numbers. The employment survey (called AKU by those of us in the know) is conducted every month and the statistics are published the following month. A sample of the population is surveyed about their employment status and the results are extrapolated to the entire population. The samples are well chosen and the numbers are usually quite accurate. I suspect that a similar method is used in the US, but I am sure someone more knowledgeable about US employment statistics will pop in soon (I have only had reason to look at US numbers a few times myself). Unemployment is usually easier to measure than employment, at least those unemployed who are registered (on the dole). ---
**Author:** ![Mersavets](https://avatars.discourse-cdn.com/v4/letter/m/aca169/32.png) [@Mersavets](https://boards.straightdope.com/u/Mersavets)\ **Post date:** [August 5, 2002, 10:42am UTC](https://boards.straightdope.com/t/where-do-the-governments-job-numbers-come-from/122408/4 "2002-08-05T10:42:10Z")
Unemployment figures are usually manipulated for political reasons. For example, Australian figures are from the Bureau of Statistics (and an extremely professional organisation too), but the ABS is restricted to using survey criteria laid down by the government for determining unemployment levels. These criteria are slanted to indicate lower unemployment eg. those who have undertaken any more than a couple of hours of volunteer work are not classed as unemployed for that period. Obviously, the number receiving benefits can’t be used as that would show a higher (and less flattering) unemployment rate. I would imagine there is some standardisation amongst countries in their statitistical methodology to permit meaningful comparison. ---
**Author:** ![pldennison](https://avatars.discourse-cdn.com/v4/letter/p/9fc29f/32.png) [@pldennison](https://boards.straightdope.com/u/pldennison)\ **Post date:** [August 5, 2002, 12:45pm UTC](https://boards.straightdope.com/t/where-do-the-governments-job-numbers-come-from/122408/5 "2002-08-05T12:45:44Z")
According to the [U.S. Bureau of Labor Statistics](http://www.bls.gov/news.release/empsit.tn.htm), in their periodic press release on unemployment: > [@](#): > > This news release presents statistics from two major surveys, the Current Population Survey (household survey) and the Current Employment Statistics survey (establishment survey). The household survey provides the information on the labor force, employment, and unemployment that appears > in the A tables, marked HOUSEHOLD DATA. It is a sample survey of about 60,000 households conducted by the U.S. Census Bureau for the Bureau of Labor Statistics (BLS). > > The establishment survey provides the information on the employment, hours, and earnings of workers on nonfarm payrolls that appears in the B tables, marked ESTABLISHMENT DATA. This information is collected from payroll records by BLS in cooperation with State agencies. In June 2002, the sample included over 300,000 establishments employing about 37 million people. > > For both surveys, the data for a given month relate to a particular week or pay period. In the household survey, the reference week is generally the calendar week that contains the 12th day of the month. In the establishment survey, the reference period is the pay period including the 12th, which may or may not correspond directly to the calendar week. The notes go on to explain further details about each survey, how the data is collected, and about the reliability of estimates and conclusions. Worth a read for anyone interested in the topic. ---
**Author:** ![hawthorne](https://avatars.discourse-cdn.com/v4/letter/h/c89c15/32.png) [@hawthorne](https://boards.straightdope.com/u/hawthorne)\ **Post date:** [August 5, 2002, 2:51pm UTC](https://boards.straightdope.com/t/where-do-the-governments-job-numbers-come-from/122408/6 "2002-08-05T14:51:15Z")
**Mersavets** ’ view is common but mistaken. There are good reasons for defining unemployment the way the ABS (and other reputable organisations do. You can read the short working paper [here](http://www.abs.gov.au/ausstats/abs%40.nsf/66f306f503e529a5ca25697e0017661f/5b9125c8e59590c1ca256a4d001fac88!OpenDocument). Combined with the participation rate the unemployment number gives a picture of the overall health of the economy, but not of the “real” employment situation. You need to delve deeper into the labour force and underemployment stats to get that. But the definition of unemployment is there to satisfy a standard, to be robust and to satisfy adding-up contraints, not fiddle the figures by order of the government of the day. ---
**Author:** ![photopat](https://avatars.discourse-cdn.com/v4/letter/p/e95f7d/32.png) [@photopat](https://boards.straightdope.com/u/photopat)\ **Post date:** [August 5, 2002, 3:02pm UTC](https://boards.straightdope.com/t/where-do-the-governments-job-numbers-come-from/122408/7 "2002-08-05T15:02:12Z")
Nightshadea-"Well I dont know but one assumes all they had to do is check for new w2 form filings for new jobs " That would work, but do those figures also include the gray area of undocumented workers? ---
**Author:** ![Stoid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/stoid/32/272_2.png) [@Stoid](https://boards.straightdope.com/u/Stoid)\ **Post date:** [August 5, 2002, 7:01pm UTC](https://boards.straightdope.com/t/where-do-the-governments-job-numbers-come-from/122408/8 "2002-08-05T19:01:44Z")
And the W-2 Filings include old jobs filled by new people. What I want to know is > [@](#): > > \*Originally posted by Coil \* > \*\* The samples are well chosen and the numbers are usually quite accurate . > \*\* How do they know? ---
**Author:** ![Coil](https://avatars.discourse-cdn.com/v4/letter/c/bb73d2/32.png) [@Coil](https://boards.straightdope.com/u/Coil)\ **Post date:** [August 6, 2002, 6:43am UTC](https://boards.straightdope.com/t/where-do-the-governments-job-numbers-come-from/122408/9 "2002-08-06T06:43:35Z")
> [@](#): > > \*Originally posted by Stoid \* > \*\* > What I want to know is > How do they know? \*\* Once the slower (but accurate) registerbased data comes in the numbers from the survey can be compared to them. Historically there is a pretty good fit between the estimated employment based on the survey data and the actual numbers. At least this is the case for Sweden and while I cannot from my own experience say how it is done in the US I do not doubt that you have competent statisticians as well.