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 #96: Here's a toy model for TIA to play with [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
Febble Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Feb-12-07 06:27 PM
Response to Reply #86
96. Here's a toy model for TIA to play with

It's not as nicely laid out as TIA's Interactive Model, I'm not such a Excel wizard as TIA. But in this Excel spreadsheet I've pasted in the crosstabulations from the National Exit Poll (from the downloaded spreadsheet, so you have the actual numbers) and I've done them first with the weights off, then with the weights on. Your 12.22 numbers won't be identical, of course, because they would have had some weights, but I don't have the interim weights, so it's all or nothing. Nonetheless, the unweighted numbers are pretty close to your early numbers, and the weighted numbers match your final weights.

I've also set it up so that in the yellow cells you can enter any numbers you want, to represent respondents who might not have recalled their vote correctly. As you will see, it doesn't matter what numbers you enter into the yellow cells, the Kerry-Bush proportions will remain unchanged (see the turquoise cells at the bottom). This is because the "model" assumes that whatever people remember or fail to remember about their vote four years earlier, they correctly reported who they voted for 5 minutes earlier. What you can experiment with is adjusting what they might have misrecalled from four years ago (including whether or not they even voted). The goal, obviously, is to tweak the recall figures to make the "actual" Gore-Bush turnout figures plausible (whatever you consider plausible). And then you can see what implications this has for defection rates.

It wasn't designed to make any point other than to convince you that whatever you think of the likelihood of misreported vote, it DOES make a different to turnout and defection rates but DOESN'T make a difference to the proportions of votes estimated for Kerry and Bush. In fact, given that the Ns in each cell are fairly small, it shows you just how exquisitely sensitive the numbers are to fairly slight mis-report rates (single figure percentages).

And if nothing else, it enables you to increase the number of decimal places in your data.

Have fun.


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