Today, 08:34 AM
It wouldn't be that simple for numerous reasons. The above methodology presumes that there is a 1:1 ratio between eligible voters, and actual people, which isn't gonna be true, and to what extent that's not true, is not gonna be revealed by the above methodology. The above methodology is also flawed because it doesn't account for the inability to contact the people most likely to be voted on behalf of by others - e.g. mentally ill, elderly, and dead people.
Not to mention, you'd need to do some control group analysis, to account for the fact that people are naturally going to lie about whether they voted. And of course how many and which direction they lie, is going to change based on location and demographics.
There would also be a generally positive relationship (I assume) between people who are even willing to answer the phone for you at all, and people who vote.
You would also need to ask each of these people what state they voted in, or otherwise control for people who may have moved and/or voted in multiple states.
You would also need to ask if they voted using a normal or a provisional ballot, and then account for the number of people who don't even know what a provisional ballot even is.
Connect With Us