Democratic Underground Latest Greatest Lobby Journals Search Options Help Login
Google

anecdote vs statistic

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
This topic is archived.
Home » Discuss » Topic Forums » Guns Donate to DU
 
iverglas Donating Member (1000+ posts) Send PM | Profile | Ignore Sun Feb-15-09 11:55 PM
Original message
anecdote vs statistic
Edited on Sun Feb-15-09 11:56 PM by iverglas

Just because that conflict arises here so often, I thought this might be interesting.

Dan Gardner is one of my favourite columnists, although I don't always agree with him entirely.

Here, his case in point is this week's plane crash, but his points are more general.

http://www.dangardner.ca/Colfeb1409.html


... Our species loves stories. We love to tell stories. We love to hear stories. It is a universal human trait, a behaviour observed by anthropologists in every culture, in every place, in every time. Evolutionary psychologists believe it is biologically hard-wired.

... These statistics may be essential to understanding safety. They may tell us more about the risks of travel than dozens of stories about crashed airplanes. And yet they are only numbers.

But when a pilot lands a plane safely on a river, it is the "Miracle on the Hudson" and all the world stops to watch. And when the widow of a 9/11 victim dies in a crash on her way to commemorate her lost husband, we all share the pain.

Inevitably, the human preference for stories over numbers skews our perceptions of the world. We see danger where there is very little. We ignore what should worry us.

We simply do not see clearly.


Please do note that he says both "We see danger where there is very little" and "We ignore what should worry us".

And he isn't talking about one kind of anecdote only.

Printer Friendly | Permalink |  | Top
dmallind Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Feb-16-09 12:08 AM
Response to Original message
1. Sure
Edited on Mon Feb-16-09 12:08 AM by dmallind
And it's applicable to guns too - when data shows remarkably little chance of a legally owned firearm being used in a crime, and yet everyone jumps on stories such as Va Tech to say we should not allow legally owned firearms.

That's what you meant, right?


Printer Friendly | Permalink |  | Top
 
57_TomCat Donating Member (527 posts) Send PM | Profile | Ignore Mon Feb-16-09 07:12 AM
Response to Reply #1
5. Sure sounds like it to me.
I imagine as in most things that one tends to quote that which defends their argument.

I still think we need to outlaw swimming pools. For the children you know.
Printer Friendly | Permalink |  | Top
 
TheWraith Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Feb-16-09 12:24 AM
Response to Original message
2. I agree.
I've loudly pointed out that statistics are the only meaningful measurement of the real world. Stories have their place, but defining our perceptions is NOT it.
Printer Friendly | Permalink |  | Top
 
pipoman Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Feb-16-09 02:38 AM
Response to Original message
3. I have considered this often
I believe public policy should be based on statistics, not knee jerk reaction to anecdotes. OTOH I don't base decisions effecting my personal safety, well being, or future strictly on statistics, but consider possible anecdotes which may effect me against the odds.
Printer Friendly | Permalink |  | Top
 
zagging Donating Member (531 posts) Send PM | Profile | Ignore Mon Feb-16-09 06:19 AM
Response to Original message
4. Lives are lived anecdotally
Edited on Mon Feb-16-09 06:30 AM by zagging
Not statistically. While an individual event may have a relationship to a statistic, the statistic can only determine probability. It cannot determine the outcome of a singular event or a stream of singular events, and there is no imperative for an individual outcome to satisfy a statistical probability. We understand this intuitively, and also understand that we are bound by our anecdotal, human condition. Hence, worry...or hope.
Printer Friendly | Permalink |  | Top
 
petronius Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Feb-16-09 03:45 PM
Response to Original message
6. Most people seem to understand this halfway - that is, 'if it supports my opinion,
it's data. If it undermines my opinion, it's an anecdote.' :)

That's a nice column, thanks for posting it. I end up thinking about the anecdote/data problem a lot because my primary professional interest is climatology, and I frequently need to answer questions that arise from an incomplete understanding of statistics, correlation, and anecdotal evidence...
Printer Friendly | Permalink |  | Top
 
spin Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Feb-16-09 04:50 PM
Response to Original message
7. Very true, there are differences between anecdotes and statistics...
I often tell two anecdotes about my mother and my daughter that involve self defense and firearms.

In the 1920 time frame my mother was walking home from work at night when a man rushed her from some bushes. She had a .22 cal S&W Ladysmith revolver in her purse. She drew the weapon and fired two shots over the attackers head. He turned and ran.

In the late 1980's my daughter had returned home from work and was in her bedroom when our burglar alarm sounded. She walked into the living room and reset the alarm. Sometimes on a windy night, the wind would trip the sensor on the sliding glass door in the kitchen.

A few minutes later, the alarm sounded again. My daughter, realizing that it was not a windy night, grabbed a large caliber revolver, a S&W model 25-2 .45 acp, and walked into the kitchen.

She found a man forcing the sliding glass door open. She pointed the revolver at him. He ran. When someone points a large frame revolver the size of a .44 magnum at you, it can be very intimidating.

She called the police and when they arrived at the front door, she told them she had a firearm in her hand and she couldn't release it. The police officers told her to point it at the floor with her finger off the trigger and open the door. One officer had to gently pry her fingers off the weapon.The police never caught the intruder.

Now these two incidents happened but they are just anecdotes. They never became statistics. Had someone been shot or my mother or daughter victimized, these stories would have become statistics. (In the case of my mother's incident, I'm not sure shootings were even tracked in the 1920s.)

But while statistics can be used in a scientific study and anecdotes can't, you can still learn lessons from anecdotes. I suspect many incidents similar to the ones I have described happen and happen more often than the incidents recorded in statistics. Of course, I have no way of proving this.

I'm glad that these anecdotes never were elevated to the status of statistics. No one was shot or killed. That's always a good thing.

Yes indeed, the writer is correct when he says, "We see danger where there is very little" and "We ignore what should worry us". That doesn't change the fact that danger exists. While you can statistically analyze your risk and determine that the chances of actually finding yourself in a dangerous situation are very slim, being prepared is not a bad idea.

I have a fire extinguisher ten feet from my computer and a fire alarm in the hallway. Since I take reasonable precautions to avoid starting a fire, I should have nothing statistically to worry about. I seriously doubt that I'll ever have to use the fire extinguisher, and I'll waste a lot of money on the batteries. But if a fire does break out, I have a better chance of saving my property and perhaps my life.

As for statistics and airplanes:

A famous statistician would never travel by airplane, because he had studied air travel and estimated that the probability of there being a bomb on any given flight was one in a million, and he was not prepared to accept these odds.

One day, a colleague met him at a conference far from home. "How did you get here, by train?"

"No, I flew"

"What about the possibility of a bomb?"

"Well, I began thinking that if the odds of one bomb are 1:million, then the odds of two bombs are (1/1,000,000) x (1/1,000,000). This is a very, very small probability, which I can accept. So now I bring my own bomb along!"

http://www.keypress.com/x2815.xml










Printer Friendly | Permalink |  | Top
 
iverglas Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Feb-16-09 07:54 PM
Response to Reply #7
8. groan

I saw "statistics and airplanes" and I was fixing to tell that one.

My version just went: Now what are the odds of there being *two* bombs on the plane??


Printer Friendly | Permalink |  | Top
 
DU AdBot (1000+ posts) Click to send private message to this author Click to view 
this author's profile Click to add 
this author to your buddy list Click to add 
this author to your Ignore list Thu Apr 18th 2024, 11:30 PM
Response to Original message
Advertisements [?]
 Top

Home » Discuss » Topic Forums » Guns Donate to DU

Powered by DCForum+ Version 1.1 Copyright 1997-2002 DCScripts.com
Software has been extensively modified by the DU administrators


Important Notices: By participating on this discussion board, visitors agree to abide by the rules outlined on our Rules page. Messages posted on the Democratic Underground Discussion Forums are the opinions of the individuals who post them, and do not necessarily represent the opinions of Democratic Underground, LLC.

Home  |  Discussion Forums  |  Journals |  Store  |  Donate

About DU  |  Contact Us  |  Privacy Policy

Got a message for Democratic Underground? Click here to send us a message.

© 2001 - 2011 Democratic Underground, LLC