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 #208: So you sided with me, rather than Hout et al? [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 archived.
Home » Discuss » Topic Forums » Election Reform Donate to DU
Febble Donating Member (1000+ posts) Send PM | Profile | Ignore Sat Aug-20-05 05:51 PM
Response to Reply #204
208. So you sided with me, rather than Hout et al?
I found that OpScans benefited Bush. But it WAS apples and oranges. There were no touch screens in the smallest (pop. wise) rural counties, and no opscans in the largest urban counties. This was what undermines Hout's study, IMO - they found that it was touchscreens that benefited Bush. I tried to control for that by considering mid size counties only, where both machines were equally likely to be chosen. But there were still big demographic variables.

The thing is, TIA, that it is one thing to say that the probability of something happening by chance is gazillion to one (though it wasn't that significant once you leave off the rural counties). It is quite another to say that the factor you are considering is the critical one. It's what collinearity is all about. If machine choice was related to some other factor, you can't tell whether it was the other factor or machine that was the critical one. Sometimes you can control for other factors in some way, but in this case the collinearity was high - presumably because touch screens had been introduced to replace the old levers and punchcards that had disenfranchised so many African American voters in 2000.

So that's really where we tend to differ, you and I - you like to keep things simple - and it's often a virtue. But there's a reason for the development of multivariate statistical methods - most things in real life have multivariate causes. It makes analysis complex. And appealing to complexity can look like nitpicking. But I would argue that it isn't - it's trying to get things right. Mebane and Sekhon did a study of Florida where they modelled many demographic variables, and didn't find a machine effect. Which doesn't mean there wasn't fraud - it just means if there was fraud it wasn't particularly associated with one type of technology rather than the other.
Printer Friendly | Permalink |  | 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