Democratic Underground Latest Greatest Lobby Journals Search Options Help Login
Google

Once again my theory of you cant trust polls done by colleges proven right

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 » Archives » General Discussion (Through 2005) Donate to DU
 
59millionmorons Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Apr-28-04 07:52 AM
Original message
Once again my theory of you cant trust polls done by colleges proven right
Edited on Wed Apr-28-04 08:01 AM by demdem
Once again my theory of you cannot trust polls done by colleges is proven correct. Quinipiac poll(college) had Specter winning 48-42. Survey USA(professional) had the race tied. I have told many people here who have been overacting to state and national wide polls released recently, not to trust them if they are done by colleges, they are amateurs and are not ready for the big time. The PA senate primary is another example. I remember a few weeks ago Quinipiac put out a PA poll that had Bush up 6 and I told everyone don't trust it. Last month we had a poll done by a college in NJ that had Bush leading by 10, just last week there was a non college poll that had Kerry up by 9. This also shows how silly it is for people to overreact to Gallup polls and others, when you have a Zogby poll that has Kerry leading 47-44. I guarantee you if the election was held today that the results would be closer to Zogby's numbers than Gallups and others.
Printer Friendly | Permalink |  | Top
rock Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Apr-28-04 07:54 AM
Response to Original message
1. Polls can be accurate if done scientifically
Scientifically of course implies a scientist do it.
Printer Friendly | Permalink |  | Top
 
displacedtexan Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Apr-28-04 08:10 AM
Response to Reply #1
4. Very few polling groups hire doctoral-level demographers.
Political Science profs supervising undergrads, some of whom have no problem dummying data to get projects done on time: What do you expect?

Sadly, even the most respected polling orgs don't hire demographers, either... probably because demographers aren't interested in producing a certain result. They are only interested in the truth, as expressed through the raw data collected.
Printer Friendly | Permalink |  | Top
 
cavebat2000 Donating Member (347 posts) Send PM | Profile | Ignore Wed Apr-28-04 08:11 AM
Response to Reply #1
5. well kinda
It means it will be done according to a scientifically proven method. :)
Printer Friendly | Permalink |  | Top
 
buddy22600 Donating Member (426 posts) Send PM | Profile | Ignore Wed Apr-28-04 07:58 AM
Response to Original message
2. polls this early mean very little
Printer Friendly | Permalink |  | Top
 
RoyGBiv Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Apr-28-04 08:08 AM
Response to Original message
3. Polls in general...
I agree with you on the college polls, generally speaking. Some universities have entire schools dedicated to this sort of thing, Michigan for instance, and their polls tend to be better. I was on the analysis end of polling a couple times in college, and it's really amazing how sloppy it got.

Anyway, I was reading some stuff last night, and I found this. I think we all might want to take note of it, especially now.

Reporting Sampling Error

How the Media Confuse and
Mislead Readers and Viewers

by Humphrey Taylor

On almost every occasion when we release a new survey, someone in the media will ask, "What is the margin of error for this survey?" There is only one honest and accurate answer to this question -- which I sometimes use to the great confusion of my audience -- and that is, "The possible margin of error is infinite."

http://www.pollingreport.com/sampling.htm
Printer Friendly | Permalink |  | Top
 
59millionmorons Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Apr-28-04 08:13 AM
Response to Reply #3
6. "The possible margin of error is infinite."
In other words, they would not bet the farm on there results.
Printer Friendly | Permalink |  | Top
 
displacedtexan Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Apr-28-04 08:16 AM
Response to Reply #3
7. This job link from pollingreport should explain the problem
What kind of qualified scientist will this job listing attract?

Posted 2/9/04

CALVIN COLLEGE, CENTER FOR SOCIAL RESEARCH

ASSISTANT DIRECTOR

Calvin College seeks an Assistant Director for the Center for Social Research (CSR). This position reports to the Director of the Center for Social Research. It is an Administrative Staff III, Grade 6, Full-time, Exempt. Job Code: 5-22178-20-AD3-06. This position is accountable for directing the operations of the CSR, especially its contract-based research. The Assistant Director is under the supervision of the Director and works in collaboration with the Director and other staff members of the CSR. This position requires a Bachelor's Degree from a four-year college or university; three to five years of experience in planning, conducting and analyzing research; or equivalent combination of education and experience. Strong statistical analysis ability, written and verbal communication skills and interpersonal skills are required. Knowledge of computers and related software such as web-based surveys and qualitative research packages is also required. Resumes including Job Code should be submitted by March 31, 2004

All submissions should be sent to D. Honner Calvin College Human Resources 3201 Burton Street, S.E. Grand Rapids, MI 49546 Fax 616-526-6832 E-mail: jobs@calvin.edu

Calvin College is a Christian college in the Reformed tradition. The candidate must embrace the Reformed tradition of the Christian faith and energetically support the mission of Calvin College. Calvin College does not discriminate in the employment of individuals on the basis of race, color, national or ethnic origin, disability, sex, or age. However, Calvin College is a Christian educational institution operated by the Christian Reformed Church and, in compliance with Title VII of the Civil Rights Act of 1964, reserves the right to give preference in employment based upon religion.


DUH!
Printer Friendly | Permalink |  | Top
 
Gman Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Apr-28-04 08:33 AM
Response to Original message
8. Polls are supposed to be a picture or snapshot in time
For a picture to come out correctly it must be focused and proper settings must be used. Polls are the same in that you have to have a good focus or sample and the questions must be appropriately worded. If any of these and a few other things are not done correctly the poll is blurred and out of focus. Polling is something a lot of organizations like to do but not very many are good at it.

Because he was the only pollster to correctly call the 2000 prez election, I trust Zogby the most.
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 Fri May 03rd 2024, 07:11 AM
Response to Original message
Advertisements [?]
 Top

Home » Discuss » Archives » General Discussion (Through 2005) 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