[FoRK] New hypotheses

Stephen D. Williams <sdw at lig.net> on Mon Oct 22 11:25:14 PDT 2007

Correlation does not prove causation, but it might indicate it.
However, both counter examples are bad because they are true of most of 
the population, in addition to the target segment.  That's no 
correlation at all.

Narrow traits shared by narrow target segment and not the rest of the 
population are more of a correlation than not.  Correlation just gives 
you a starting point for exploration of a scientific study.  Dismissing 
a correlation as invalid is premature, unless you have counter evidence.

So, all you have to do, Jeff, is:
If the hypothesis is that owning an F-150 causes jackass driving:
Find random people with no history of jackass driving and give them a 
random assortment of F-150s and non-F-150s, and then measure how many 
become jackass drivers.

Or, oppositely, if the hypothesis is that jackass drivers are drawn to 
F-150s:
You need to find all jackass drivers before they buy their first car and 
find out how many of them choose to buy F-150s.

Or come up with a different hypothesis:
On a recent trip, a rental car type that I had never driven before 
greatly amplified my jackass-drivered-ness.  Either that or I had sleep 
problems or something, because I had a whole string of near accidents in 
the span of a few hours that I couldn't explain and were totally out of 
character.  I felt fine, but didn't perform that way.  I'm usually a 
fairly precision driver and spot things well before there's an issue.  I 
have a pilot's license for Pete's sake.  But in that car, on that day, I 
couldn't pull out of a parking spot without an issue.

Perhaps the F-150 has such poor ergonomics that it effectively lowers 
driver skill.

sdw

Jeff Bone wrote:
>
> 100% of all serial killers drank milk as children...
>
> ;-)
>
> jb
>
> PS - I'm sticking to my story.  There's something about assholes and 
> Ford F-150s...
>
>
> On Oct 22, 2007, at 11:44 AM, Reese wrote:
>
>> Jeff Bone wrote:
>>> Ford F-150 pickup trucks represent an estimated 1-4% of the cars on 
>>> the road in Austin, TX.
>>> 100% of 5 incidents of egregious jackass driving I observed this 
>>> weekend involved Ford F-150s.
>>> .:  A disproportionate number of Ford F-150 buyers are asshole drivers.
>>
>>
>> Error. Allow me to illustrate:
>>
>> Pants-wearers represent an estimated +95% of the people in Austin TX,
>> TX, USA, whatever.
>>
>> 100% of incidents involving rude smokers involve perpetrators who've
>> worn pants at some point in their lives.
>>
>> .: A disproportionate number of pants-wearers are smokers who like to
>> blow smoke in your face.
>>
>

-- 
swilliams at hpti.com http://www.hpti.com Per: sdw at lig.net http://sdw.st
Stephen D. Williams 703-371-9362C 703-995-0407Fax 20147 AIM: sdw


More information about the FoRK mailing list