8.6 % (±15.2 %) - wtf???

[w¡n9Nµ7 §£@¥€r 123]

From the US Census:

Sales of new one-family houses in December 2009 were at a seasonally adjusted annual rate of 342,000, according to estimates released jointly today by the U.S. Census Bureau and the Department of Housing and Urban Development. This is ... 8.6 percent (±15.2%)* below the December 2008 estimate of 374,000.

From the explanatory notes contained within the same document:

These statistics are estimated from sample surveys. They are subject to sampling variability as well as nonsampling error ... Whenever a statement such as "2.5 percent (±3.2%) above" appears in the text, this indicates the range (-0.7 to +5.7 percent) in which the actual percent change is likely to have occurred.

Sooooooooooooo.....

This means the actual December 2009 new home sales may have been as bad as 23.8 % below December 2008 values or as good as 6.6 % above December 2008 values.

More importantly....

####### is the point of reporting on a metric with such a large confidence interval???????? If this is the best they can do, I'd rather see them simply say they don't have the funding to do a better measure and simply not publish one.

Edited January 29, 2010 by w¡n9Nµ7 §£@¥€r

[scandal 951]

In Australia they don't need confidence intervals. They can measure statistics precisely with no error tolerances required.

Clever people, those Australians.

[w¡n9Nµ7 §£@¥€r 123]

In Australia they don't need confidence intervals. They can measure statistics precisely with no error tolerances required.

Clever people, those Australians.

Indeed.

[I AM NOT THAT GUY 5,715]

From the US Census:

From the explanatory notes contained within the same document:

Sooooooooooooo.....

This means the actual December 2009 new home sales may have been as bad as 23.8 % December 2008 values or as good as 6.6 % above December 2008 values.

More importantly....

####### is the point of reporting on a metric with such a large confidence interval???????? If this is the best they can do, I'd rather see them simply say they don't have the funding to do a better measure and simply not publish one.

It takes 6 months to have any confidence in government statistics. This administration is looking for any good numbers, and wants to get them out quickly, since day one. However, the numbers frequently get adjusted away from the early numbers the sycophants keep hanging their hopes on.

[Ban Hammer 13,345]

our tax dollars at work with great math like that.

[w¡n9Nµ7 §£@¥€r 123]

our tax dollars at work with great math like that.

all they have to do is expand the survey and collect more data points.

more data, smaller ci.

[I AM NOT THAT GUY 5,715]

our tax dollars at work with great math like that.

The math is fine. It is more a matter of premature ejaculation.

[What Happened? 4]

It's "fuzzy math".......

[Ali G. 37]

In Australia they don't need confidence intervals. They can measure statistics precisely with no error tolerances required.

Clever people, those Australians.

No need to rub salt on the wound but property prices in Melbourne and Sydney rose yet again this month.

How's it going in Chicago dh?

Edited January 29, 2010 by Booyah

[w¡n9Nµ7 §£@¥€r 123]

dh nice..kathryn will never know what that means

[웃 910]

US Census...

####### is the point of reporting on a metric with such a large confidence interval???????? If this is the best they can do, I'd rather see them simply say they don't have the funding to do a better measure and simply not publish one.

http://www.npr.org/blogs/newsandviews/2008...republican.html

[w¡n9Nµ7 §£@¥€r 123]

http://www.npr.org/blogs/newsandviews/2008...republican.html

kma.

[I AM NOT THAT GUY 5,715]

http://www.npr.org/blogs/newsandviews/2008...republican.html

You might be surprised to find that you could share some of the same views as card-carrying members of the GOP. To find out if you're a closet Republican (or at least, not totally Democrat), take this Black Voices quiz.

ETA: "Sorry! This quiz is temporarily unavailable."

Edited January 29, 2010 by Lone Ranger