Ridecamp: [RC] [Guest] Pull Codes/Data Collection

K S SWIGART katswig@xxxxxxxxxxxxx
Roy Drinnan said

>> My profession and expertise is that of a 25 year computer systems
consultant and software
  developer.  I can assure you that you can never go wrong by collecting
data.  The worst you have
  done is waste a little paper.  <<

And he is mistaken.  Bad data IS worse than no data.  If you have
no data, then you know you have no information and know that you are
making an uninformed choice and that your decision is a shot
in the dark.

If you have bad data but don't know it is bad data, you will think
that you are making an informed decision when you are not...or...
even worse, having the wrong informaion will lead you to the
WRONG choice.

Personally, I am glad that they started publishing pull codes,
because they have been collecting this information all along,
but it wasn't until they started publishing it that it became
apparant that they were collecting bad data.

In this particular instance, it didn't matter that the AERC was
collecting bad data, because nobody tried to do anything with
the data anyway.  The first time they did do something with it (i.e.
publish the raw data) it didn't take long for it to become
apparent how inaccurate it was.  However, had somebody tried
to draw specific conclusions from a statistical analysis of the
data before they started publishing it and its inadequacy brought
to light, these specific conclusions may very well have been
wrong conclusions.

And the fact that the pull codes (if used accurately) made no
distinction between
the horse that is consistently ridden with a mild lameness
that for five out of ten starts gets pulled because it is
so obviously lame that the vets cannot miss it and disqualify
the rider despite his/her willingness to go on, and the horse
with a rider who was so cautious that they pulled their own
horse half the time because they were worried that it might
go lame is extremely relevant, and anybody who can't see
that needs their head examined.

The only even marginally valid inferences that might be made from
the data that the AERC currently has is:  "It appears as though
most horses that get pulled from endurance rides get pulled
because they are showing some degree of lameness to somebody."

And since most people who spend any time at all at endurance
rides could make that inference without looking at any of the
AERC's data at all, the AERC's current pull code data comes
pretty close to being under the heading of totally useless.

Now that they have been publishing it, this has become apparent
and hopefully they will make sufficient changes so that data
collected in the future won't be.

Unfortunately, in the mean time, there appear to be lots of
people (even, apparently, people who collect data and perform statistical
analyses professionally) who don't seem to know that bad data
is worse than no data and are actually attempting to draw
conclusions from the AERC's current data (or are worried that
other people will try to draw conclusions from the AERC's data).

Although, I won't say that the AERC's current pull code data is
TOTALLY useless, if for no other reason than that it demonstrates
that collecting good data is not as easy as it might seem. Sort
of like the old saw that says "Nobody's life is totally useless.
All people have some worth even if only to serve as a bad
example."

kat
Orange County, Calif.

p.s.  It is a lot easier to collect useful data if you at least
have some idea of what it is you intend to use it for before
you start collecting it. And, in retrospect, maybe the AERC's
data is good for what they intended to use it for...the ability
to point at it in the event that somebody asked and say, "See
we do collect and keep data.

"ummmmmmm...just don't look at it too closely or actually try
to analyze it." :)

But if that is the case, the publishing it was a mistake, because
if you publish it, it would be foolish not to expect people to
try to use it....for something (even if they are trying to use
it for something that it was never intended to be used for).



=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
 Ridecamp is a service of Endurance Net, http://www.endurance.net.
 Information, Policy, Disclaimer: http://www.endurance.net/Ridecamp
 Subscribe/Unsubscribe http://www.endurance.net/ridecamp/logon.asp

 If you are an AERC member - PLEASE VOTE in the Director at Large 
 and By Laws Elections.

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=