More from the bad statistics department

[cute fuzzy bunny]

Was just perusing some information regarding several of the more closely watched 'confidence indexes'.

Fairly disgusting stuff.

In many, they survey under a thousand people. Not even an order of magnitude away from making statistically insignificant.

A lot consist of loaded questions like "Are you happy with the current administrations policies that are improving our economic situation?", or they include complex questions the average joe simply wouldnt understand, but will respond to because they dont want to appear uninformed.

Then several apply 'weighting' and 'slimming' (their words, not mine) where they eliminate 'excessive negative or positive answers' and add a positive bias to sets of answers they 'believe may be unfairly biased to a negative'. One formula added nearly a 50% positive bias to questions the 'researchers' felt led to unfairly negative answers.

Why dont they just make up the numbers they want to publish, wordsmith a statement to that effect that most people wouldnt understand and bury it 3 layers deep in a footnote of an explanation linked to a white paper thats mentioned in the chart but only available by sending a self-addressed stamped envelope to some address in Pueblo Colorado?

Sure would save a lot of time.

What grinds my coffee is that a fair number of these 'statistics' are closely watched by 'market movers', who swoop in to buy and sell based on the absolutely, completely bogus results.

 

[soupcxan]

YouRe: More from the bad statistics department

You can come up with a statistic to prove anything. 47% of people know that.

 

[cute fuzzy bunny]

Actually its 48%...

 

[Mr._johngalt]

It's been proven that 99% of all statistics are wrong
anyway

The old line "figures don't lie but liers figure" comes to mind here.

JG

 

[BUM]

Its getting down to making your own choices in life. In healthcare I'm used to the phrase "informed consent". You know one of the forms you sign before any procedure is done. Yes, you are informed, period. The quality of the info and applicability to you specifically is questionable.

In the northeast too many news items start with "A Marist Public Opinion Institute poll reveals..." Well I live near Marist College and the questions they ask couldnt be less revealing.

Here is an example of one of their routine questions:

In general, thinking about the way things are going in the country, do you feel things are going in the right direction or that things are going in the wrong direction?

Now a bunch of college kids collect a few thousand answers to the above question. What could you do with this info...anything you want.

 

[Sheryl]

I wanted to be sure I was correctly stating one of my favorite quotes: "There are three kinds of lies: lies, damn lies, and statistics."

In the process of searching for it, I found this, which is more interesting anyway:

http://www.zealllc.com/commentary/damnlies.htm

 

[nuisance]

That is an interesting article, but I don't quite buy its doom-and-gloom scenario. He makes a pretty strong point that the total money supply has risen a lot, and more than CPI. Then he says that this will lead to a lot of inflation because when more money chases fewer goods, there is inflation. At no point does he say anything about the amount of available goods, though. I'm very willing to believe that there is a significant, real growth in the amount of goods there are to go around. These are generated eg. by more efficient farming techniques, imports from China, etc.

Tim