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
Google

Reply #94: The "heroic methods" are obvious and I've stated them before...short list... [View All]

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
This topic is locked.
Home » Discuss » Topic Forums » Election Reform Donate to DU
Sancho Donating Member (1000+ posts) Send PM | Profile | Ignore Sun Jun-17-07 11:33 AM
Response to Reply #93
94. The "heroic methods" are obvious and I've stated them before...short list...
all of these are well-documented and are standard operating procedure in the measurement literature and likely could be done with minimal expense given the entire budget:

1.) there have to be targeted samples that are reactive as data come in..as soon as a sampling need is discovered by annedoctal response or early returns...a traveling team or local group need to go to some precincts and hit all the polling stations in that precinct or some such idea (this is old survey methodology)...could even be done with volunteers, local college students, etc...
2.) multivariate (equation modeling) variables have to be correlated with demographics and very careful subsamples have to be identified to test questions about city/rural, reluctant responders, previous voting records, etc...this happens between elections with pilots, etc. and guides the survery creation and leads to number 3 (next)
3.) much better recording of demographics of 100% going in each of targeted polling stations have to be recorded (even by observation and registration records) if responders are "reluctant"; hell, in Florida you can write down auto plates and look them up!
4.) pre and post surveys of actual voters that, again, target specific precincts and unusual voter turnout and absentee participation and voting method have to be contrasted with actual voter results; takes a larger sample in those targeted precincts
5.) websites that voluntarily allow ALL voters to log in and answer a poll, record their precinct, poll statiion, etc. have to be 100% available and advertised; ever heard of American Idol?
6.) all polls have to quickly and automatically compare local and national elections, voting method, unusual pre-election polls results, early unusual turnout, etc. for "before noon" awareness of outlier patterns needing attention
7.) all surveys have to report more sophisticated analysis (IRT) for rater effects, differential person functioning, etc.
8.) precinct level data must be collected and reported in some cases (yes, the idea of identity is a solvable problem...don't even ask it's so easy and it is done every day from kindergarten to medical procedures)
9.) yes, I've had plenty of near 100% participation in surveys from kindergarten parents to patients in medical procedures to participants in a conference evaluating a speaker...it's a matter of the samples motivation/value in filling out the poll vs. time/fear at filling it out. IF an advertisement was broadcast that some precincts would be targeted for polling for quality control, and some people were pre-contacted, and the more frustrated voters were involved....some key polling stations and precincts would have a very large exit poll participation...buy-in takes work and skill to prepare the sample to participate; our current pollsters don't develop a confident set of participants with "secret data" and bad predictions!
10.) much better audio and video recording of voting (within legal limits) and recording of voting experiences (interviews) have to accompany voters as they enter / exit (and yes, if you don't know how to do this annonymously, then take a class); must target any voter problems or those turned away or those who can't wait in long lines, or those who fill out alternative ballots if they aren't on the registration, etc.

This could be a long post, and I'm sure that OTOH will debate all the procedural issues (well documented in the literature) with all the methods (and how to do them correctly) in this quick and easy list, and I'm sure that Freeman and others are very aware of these...our pollsters either don't do these, don't do them well, or don't report them...and there are some local attempts to implement these by watch dog groups as you know...No more excuses!
Printer Friendly | Permalink | Reply | Top
 

Home » Discuss » Topic Forums » Election Reform 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