[Date Prev][Date Next][Thread Prev][Thread Next]
[Date Index] [Thread Index] [Author Index]
Return to main CEDA-L Archive Page

straw vote

Actually, if there were 64 respondents to the straw vote out of 160-170,
ASSUMING THE SAMPLE IS RANDOM, then the straw poll would be a relatively
good predictor--and since most NDT schools have joined CEDA and will hence
vote on the topics, it might be more random that we think.

Many social scientific studies are published daily and in many disciplines
with samples of far less that 35 per cent or so of the population--in fact,
presidential election pollsters achieve pretty good accuracy these days
with exit polls of maybe 2,000 voters out of around 80,000,000 or so.

Of course I realize that this comes from years and years of researching how
to come up with a random sample, based on years and years of election
trends and sensitivity to many intervening variables.

The possible intervening variables in Stannard's study:  1)  it is limited
to those who are on ceda-l and ndt-l (and where-ever else he posted it),
and some not on the l are subscribers; 2) one program may have as many as
15-20 different subscribers to the ceda-l--and as far as I know there
wasn't a limit from one school in this straw polls--in other words some who
can't vote in the CEDA poll may have voted in this one; 3) voters were
encouraged to rank all 5 in the straw poll, and the CEDA poll explains how
to rank fewer than five if we so desire.

I don't see sample size per se as a problem--in fact the sample in relation
to the population is pretty big.

I still think this poll was a good experiment and can inform those who were
reserving judgment to see what a listserve that includes students might
want.

An interesting future straw poll would be one in which students alone were
asked to vote individually, and coaches  encouraged all of their students
to write in and cast votes who were subscribers.  Only votes on the
students own id numbers would count.

A lot of programs poll their students and vote as a school anyway--that
isn't always fair as a 6-5 majority gets translated into 11-0 with the
single vote.

What do you all think of the idea of students voting directly for topics?
And how would this work, one per school, all team members?  Or should it be
limited to an "expert" sample (such as those attending nationals)?

Thanks for conducting the poll--I'm sure it will spur discussion on these
and other issues.

Tom
Archive created by Jonathan Stanton (jonathan@cs.jhu.edu)
Return to main CEDA-L Archive Page