You are viewing an obsolete version of the DU website which is no longer supported by the Administrators. Visit The New DU.
Democratic Underground Latest Greatest Lobby Journals Search Options Help Login

Reply #12: There are many ways to skin the cat... [View All]

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
Home » Discuss » Topic Forums » Election Reform Donate to DU
Sancho Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Mar-31-09 08:00 AM
Response to Reply #10
12. There are many ways to skin the cat...
if there were person/item level poll data, some investigations using SEM would likely be more sensitive to patterns of interest than we've seen from the WPE/parametric analyses. I can imagine many ways that things could or should be done. In my 30 years experience with data analysis including surveys, if you have data that is "too sparse" in 2000, then you fix the problem in 2004 and 2008. We do have computers and things now. I believe that American Idol records 60,000,000 votes in two hours! I would love to see what would have happened if there had been a parallel exit election (voluntary) in Sarasota in the last decade. Everyone is so frustrated that I'd be a very large group of honest voters would have participated and identified the issues in a real and unquestionable way. It would have cost much less than all the court cases! It would also have called into question (possibly) several Presidential elections IF the results had demonstrated manipulation or significant error. The GOP election supervisor, secretary of state in FL, and Governor really, really don't want such a check on elections to occur here. That would require no more than descriptive and obvious analysis.

Sampling is something that is pretty well understood. The problem is "nonignorable nonresponse" (see Wainer).

Exit pollsters don't do what is necessary because their employers (as currently contracted) don't have that interest. WPE is less important with methods that are sample independent (like IRT). Pollsters and "vote counters" have not been using those methods as far as I can determine (I've attended and presented at ASA meeting, etc.). I think that traditionally, it's not in their tool box since those methods are more often used in medical field assessment, test item analysis, and professional license testing.

Likely, an exit poll process needs to be completely public and with the intent of checking on selected new voting processes (like DRE's).
Printer Friendly | Permalink | Reply | Top

Home » Discuss » Topic Forums » Election Reform Donate to DU

Powered by DCForum+ Version 1.1 Copyright 1997-2002
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