Ross Perot and Statistics

Thread View: sci.math.stat

3 messages

[T] Back to Threads [V] Tree View [G] Group Articles

3 total messages Started by howard Fri, 02 Oct 1992 16:18

#624

Author: howard
Date: Fri, 02 Oct 1992 16:18

31 lines
1676 bytes

Was it just me, or did anyone else cringe a bit at Ross Perot's
methods of "asking the volunteers" whether he should run for 
President?

Some problems I saw with his "technique" were:

1) defining the population. Is a "Perot Volunteer" someone who signed
the petition to put him on the ballot? Are people who are paid allowed
to vote?

2) Deciding on the sampling method. Apparently, some states just
got together 30-50 people they knew supported Perot and decided without
asking other Perot supporters in the state. Other states made more of
an effort to contact thousands of supporters. And apparently 
part of the decision was made on the number of calls to the 800 number. 
Trouble was, they did not tell people before they called that they could
not vote no, only yes. 

3) Conducting a sample representative of the population. It seems to
me that the most fair way to determine support would have been to 
conduct a random sample of those who signed petitions, and then 
weight those results by the size of the state. 

I think what surprised me most was the failure of the news media to 
question his sampling technique. Should journalism students be required
to take more statistics?


Anyway, I'm voting for Bill Clinton. 

-Mary Howard

Re: Ross Perot and Statistics

#619

Author: anthony.r.rizzo
Date: Fri, 02 Oct 1992 17:18

14 lines
1063 bytes

In article <1992Oct2.111839.7278@iscsvax.uni.edu> howard@iscsvax.uni.edu writes:
>Was it just me, or did anyone else cringe a bit at Ross Perot's
>methods of "asking the volunteers" whether he should run for 
>President?
>
>Some problems I saw with his "technique" were:

................... much stuff deleted ...............

This, clearly, is one case where sample statistics are
irrelevant.  We'll test the ENTIRE voting population in November.
Lets hope that the correct techniques will be used THEN.

Tony Rizzo

Re: Ross Perot and Statistics

#627

Author: r hardin
Date: Fri, 02 Oct 1992 23:47

13 lines
989 bytes


>This, clearly, is one case where sample statistics are
>irrelevant.  We'll test the ENTIRE voting population in November.
>Lets hope that the correct techniques will be used THEN.

Actually, public opinion doesn't mean anything except in view of
a judgment that decides, eg. an election, which finally produces
a representation.

Otherwise it's a complete surface effect, produced by surveys and newspapers.

Just a point about the grammar. (ref. Derrida, The Other Heading, 1992)

Thread Navigation

This is a paginated view of messages in the thread with full content displayed inline.

Messages are displayed in chronological order, with the original post highlighted in green.

Use pagination controls to navigate through all messages in large threads.

Back to All Threads