Annoying NPR Article - Mothering Forums
Forum Jump: 
 
Thread Tools
#1 of 6 Old 05-03-2012, 09:21 AM - Thread Starter
 
TiredX2's Avatar
 
Join Date: Jan 2002
Location: it appears to be a handbasket
Posts: 20,500
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)

Put Away The Bell Curve: Most Of Us Aren't 'Average'

 

http://www.npr.org/2012/05/03/151860154/put-away-the-bell-curve-most-of-us-arent-average

 

Please, please, someone tell me I'm missing something!  They seem to be removing a very, very small part of a population and then expressing confusion/surprise that that population isn't also a normal (statistically speaking) population.  Just like looking at NBA players and saying, "Wow, we have a lot of outliers on height here.  I guess height isn't a bell curve distribution!"


 

 

TiredX2 is offline  
#2 of 6 Old 05-03-2012, 09:46 AM
 
JollyGG's Avatar
 
Join Date: Oct 2008
Posts: 1,647
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)

I was as baffled as you. I'm not sure if the writers of the study or the writers of the article failed to understand statistics or study design.


Mom to DS 4/24/03 and DD 4/17/06
JollyGG is offline  
#3 of 6 Old 05-04-2012, 08:59 AM
 
MrsGregory's Avatar
 
Join Date: Dec 2011
Location: The 'burbs of Central Texas.
Posts: 1,030
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)

Ok - I'll risk making myself look like a complete and utter moron here, as I listed to the news piece on the radio (yes, I listen to NPR.  And thrash metal.  That's my car.  2whistle.gif) and I understood it (the parts I heard over the train) this way:

 

You have 100 people.  Say some of them are NBA players.  Some of them aren't.  You measure their heights, and you come up with an average height of, we'll just say 6' even.  But, most people aren't actually near 6', you just wound up with 6' as an average because you had a few super-tall folks in there, screwing things up for you.  Most people are shorter, but some people are much, much taller.

 

In addition, I understood this from the segment:  The bell curve was developed with regards to factory workers.  So, in the factory setting you can make anywhere from 0-100 widgets an hour, but you cannot make more than 100 widgets.  The factory just won't allow it.  So some very slow people make 2 widgets, most make, say 50, and some make 100.  However what wasn't accounted for was the fact that people making 100 widgets might actually be able to make more, but the factory was just too slow for them.  So the 50 made by the peak of the curve isn't wholly accurate because if the factory didn't constrain the top performers they would be able to make more than 100.

 

So, the statement that most of us aren't average is true if we are averaging ourselves against the so-called "superstars" in any given field.  I see it as most of us are one way, and then a couple show-offs come along and make the whole lot look bad.

 

 

I'm not arguing the merits of the study, just chiming in with my understanding of it.  It was an interesting listen.

I realize my discussion of the subject is painfully uneducated, but that's what you get with me.  orngbiggrin.gif
 


lovestory.gif   And on 09/23/2011, we were three;  husband, daughter, and me!

MrsGregory is offline  
#4 of 6 Old 05-04-2012, 10:34 AM
 
JollyGG's Avatar
 
Join Date: Oct 2008
Posts: 1,647
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)

Here's the problem with the study. They didn't start out with a random population. They started with the tail of the curve and then expected it to act like it's own curve.. When instead it acted like, well, the tail of the curve.

 

So let's take a number people are relatively familiar with - IQ. Most people have an IQ within 1 standard deviation of 100; roughly 75% of the population. Of the remaining 25% 1/2 fall below an IQ of 85 and 1/2 fall above an IQ of 120. At the very ends of the curve is the less than 1% who fall way on the tails of the curve. These people on the tail is who they sampled when they sampled a population of published academics, award winning actors, and pro athletes.

 

You would expect in this small sample size to see wide ranging abilities based on the fact that the group is so very elite. So it's not surprising that there is only one person whose broke a world record, or earned more acting awards, or whatever. There are simply fewer and few people at that skill level the further out of the curve you get. For example once you get out to the less than 1% end of the IQ range. There are going to be relatively few people with an IQ of 140, but perhaps only one or two in the world with an IQ of 180.

 

So, in contrast, to the claimed idea that the study disproves the bell curve it actually acts just like you would expect the very tail end of the curve to act.

 

Furthermore the study assumes that journal publications, awards won, and sports acclaim have nothing to do with anything but raw talent and we all know that that is an inaccurate premise.

 

The bell curve is not a premise designed for or by factory work. It is an idea seen over and over in large random population studies.

 

The study would be a perfect example of inappropriate applications for a bell curve, such as an elite pool, or a grouping with external constraints on performance (such as a widget factory with a max production of 100).


Mom to DS 4/24/03 and DD 4/17/06
JollyGG is offline  
#5 of 6 Old 05-04-2012, 11:32 AM - Thread Starter
 
TiredX2's Avatar
 
Join Date: Jan 2002
Location: it appears to be a handbasket
Posts: 20,500
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Quote:
Originally Posted by JollyGG View Post

Here's the problem with the study. They didn't start out with a random population. They started with the tail of the curve and then expected it to act like it's own curve.. When instead it acted like, well, the tail of the curve.

 

So let's take a number people are relatively familiar with - IQ. Most people have an IQ within 1 standard deviation of 100; roughly 75% of the population. Of the remaining 25% 1/2 fall below an IQ of 85 and 1/2 fall above an IQ of 120. At the very ends of the curve is the less than 1% who fall way on the tails of the curve. These people on the tail is who they sampled when they sampled a population of published academics, award winning actors, and pro athletes.

 

You would expect in this small sample size to see wide ranging abilities based on the fact that the group is so very elite. So it's not surprising that there is only one person whose broke a world record, or earned more acting awards, or whatever. There are simply fewer and few people at that skill level the further out of the curve you get. For example once you get out to the less than 1% end of the IQ range. There are going to be relatively few people with an IQ of 140, but perhaps only one or two in the world with an IQ of 180.

...

 

The study would be a perfect example of inappropriate applications for a bell curve, such as an elite pool, or a grouping with external constraints on performance (such as a widget factory with a max production of 100).

 

 

EXACTLY!!!!!! 


 

 

TiredX2 is offline  
#6 of 6 Old 05-07-2012, 09:27 AM
 
MrsGregory's Avatar
 
Join Date: Dec 2011
Location: The 'burbs of Central Texas.
Posts: 1,030
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)

I'll (happily) take your word on the matter;  even if I'd tuned in on time to hear about the selection process, I won't pretend to fully understand it.

 

How was the bell curve developed? 
 


lovestory.gif   And on 09/23/2011, we were three;  husband, daughter, and me!

MrsGregory is offline  
Reply

Tags
News

Quick Reply
Message:
Drag and Drop File Upload
Drag files here to attach!
Upload Progress: 0
Options

Register Now

In order to be able to post messages on the Mothering Forums forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in

Human Verification

In order to verify that you are a human and not a spam bot, please enter the answer into the following box below based on the instructions contained in the graphic.



User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off