DIY Chatroom Home Improvement Forum (http://www.diychatroom.com/)
-   Off Topic (http://www.diychatroom.com/f39/)
-   -   Statistics (http://www.diychatroom.com/f39/statistics-184590/)

 ddawg16 08-01-2013 01:37 AM

Statistics

More than once, some of you have heard me make a statement something to the effect "Did you pull that statistic out of your ...?"

Well....no longer do you have to pull it out of your posterior ....

You can pull it from here....

http://www.statisticbrain.com/

It's not all inclusive.....but....it sure helps to put some things in perspective.

 PoleCat 08-04-2013 08:56 AM

78.6% of all statistics are made up.

 bbo 08-04-2013 09:53 AM

 bbo 08-04-2013 09:53 AM

Quote:
 Originally Posted by PoleCat (Post 1224924) 78.6% of all statistics are made up.
and 5/4 people can't do fractions.

 ddawg16 08-04-2013 10:06 AM

I posted the same above link on another forum I'm on......

I had to ask the mods to close it.....67.3% of the posters were trolling....

Especially one....disagreed with my statements...but would not post ANY #'s to back up his position....

To give an example....we got on the topic of what is safer...planes or cars....he said compare which is safer....travel by car or plane when you look at trips....funny thing is, he was sort of right....but it was so far out of context that it really was meaningless....especially when you consider the # he was alluding to was world wide...but in the US we had 3 years without a single commercial fight fatality.

What he was really trying to do was show that speeding was not responsible for traffic deaths....so when they say speed was a 'factor' in over 10,000 fatalities, it's not the cause. Actually, the cause was the solid object they impacted because they were going too fast.

 MTN REMODEL LLC 08-04-2013 10:11 AM

Give me statistics, and I can proove that the average person has one breast and one testical.:yes:

 ddawg16 08-04-2013 10:32 AM

For those of you bored.....here is the same subject on a different forum....you can see how it went from fun to sh!t in 30 seconds....and without the help of some of our own DIY members.

<Links to other forums are not permitted>

 caveeagle 08-04-2013 11:14 AM

Great topic. I am a Lean/Six Sigma (BB) project manager, and use statistical analysis in many stages of my work. Just getting accurate DATA is only the beginning. Learning how to interpret the DATA is just as important.

Our media constantly bombards us with useless stats and illogical, erroneous conclusions drawing viewers to a cunclusion that they support.

Without getting too political, here is an easy example:

"People who drink one glass of wine/day have a 90% (made up stat) reduced risk of heart attack & stroke.

Its easy to arrive at the conclusion that the 1 glass of wine had a causal implecation with the result, and that we should all make it a point to drink that glass of wine. Right?

Not so fast Maynard!

When you stratify their research, you find that its a pretty narrow cross section of folks that chose wine as a daily routine & limit it to 1 glass. It just so happens that these are also the folks (generally speaking) that tend to take care of themselves, demonstrate self control, live within a well regulated routine, etc.

So their study really only proved correlation, but NOT causation.

So the (if) wine producing industry telling us that a glass/day is "good for you" is really a mis-interpretation of the DATA influenced by their bias.

Tons of examples like this on the news EVERY day.

My other pet peeve is the media reporting on public opinion surveys as if that was news. Most people beleive that 80% (made up) of Americans are morons!! Why the heck should I care what percent of people approve of the 1st Lady's latest trip to Spain?! This tactic is just manufactured news, designed to manipulate public opinion. Its super easy to see this. When they site a given survey, Go find the ENTIRE survey on the web. Then pick out the metrics they chose to report on. .....And the ones they chose NOT to report on.

Sorry for the rant!!
:)

 rossfingal 08-04-2013 11:49 AM

Statistically speaking - it's been shown that: the vast majority of people know enough to be "dangerous".

Not me, though!
It's been shown, "statistically" -
I know enough to be REALLY, dangerous!!! :)

"dawg" - who owns/controls, the vast majority of the media?

 Daniel Holzman 08-04-2013 01:59 PM

I used to work with a businessman who was terrified of flying. He knew there was a one chance in a million there would be a bomb on board, and he just couldn't take that kind of risk.

He was a bit of a statistician however, so he figured out that there was only one chance in a trillion there would be two bombs on board, and he was willing to take that risk. So now whenever he flies, he carries his own bomb.

 TheEplumber 08-04-2013 03:53 PM

“Facts are stubborn things, but statistics are pliable.”
― Mark Twain

 MTN REMODEL LLC 08-04-2013 10:08 PM

Quote:
 Originally Posted by caveeagle (Post 1224967) Great topic. I am a Lean/Six Sigma (BB) project manager, and use statistical analysis in many stages of my work. Just getting accurate DATA is only the beginning. Learning how to interpret the DATA is just as important. Our media constantly bombards us with useless stats and illogical, erroneous conclusions drawing viewers to a cunclusion that they support. Without getting too political, here is an easy example: "People who drink one glass of wine/day have a 90% (made up stat) reduced risk of heart attack & stroke. Its easy to arrive at the conclusion that the 1 glass of wine had a causal implecation with the result, and that we should all make it a point to drink that glass of wine. Right? Not so fast Maynard! When you stratify their research, you find that its a pretty narrow cross section of folks that chose wine as a daily routine & limit it to 1 glass. It just so happens that these are also the folks (generally speaking) that tend to take care of themselves, demonstrate self control, live within a well regulated routine, etc. So their study really only proved correlation, but NOT causation. So the (if) wine producing industry telling us that a glass/day is "good for you" is really a mis-interpretation of the DATA influenced by their bias. Tons of examples like this on the news EVERY day. My other pet peeve is the media reporting on public opinion surveys as if that was news. Most people beleive that 80% (made up) of Americans are morons!! Why the heck should I care what percent of people approve of the 1st Lady's latest trip to Spain?! This tactic is just manufactured news, designed to manipulate public opinion. Its super easy to see this. When they site a given survey, Go find the ENTIRE survey on the web. Then pick out the metrics they chose to report on. .....And the ones they chose NOT to report on. Sorry for the rant!! :)
No rant in my opinion....good points:thumbsup:

 Duckweather 08-05-2013 05:52 PM

They say alcohol is the cause in 10%, (may not be exact), of all accidents. So that would mean that 90% are caused by being sober? With statistics like that, there would be some that would say, maybe police cars should be equipped with a bar, for those who need one for the road.

 rossfingal 08-05-2013 06:07 PM

It's been said -
100% of "accidents" are:
"accidental"!
(Geesh!!! I might be the by-product of an accident)!! :)

Most accidents are preventable!!
Too late!!
:)

 ddawg16 08-05-2013 06:07 PM

Quote:
 Originally Posted by Duckweather (Post 1225474) They say alcohol is the cause in 10%, (may not be exact), of all accidents. So that would mean that 90% are caused by being sober? With statistics like that, there would be some that would say, maybe police cars should be equipped with a bar, for those who need one for the road.
On a serious note....that is a hot topic of discussion in some places.

The actual wording is 1/3 of the fatal accidents involve alcohol. When you start looking at a break down of actual causal factors...it's a lot more complicated...but more obvious. For example, of the fatalities, almost half of the accidents (maybe we should just say crash) are solo vehicle accidents.

So....what do you think the odds are that of those solo vehicle crashes that both alcohol and speed are involved?

At the end of the year, over 33K people are dead....

But what is more important...driving fatalities is one of the most preventable deaths out there....in most cases, it's not an accident....one or more drivers did something they should not have been doing.

All times are GMT -5. The time now is 01:55 AM.