Election Reform
In reply to the discussion: What if we had an election without electronic voting machines? [View all]Stevepol
(4,234 posts)The polling companies are private businesses and if their polls indicate that the Dem is doing well and then the actual (i.e. the machine counted) results are considerably different and this happens several election cycles, that company is not going to be in business very long.
The polling companies thus came up with ways to "adjust" the polls to take account of the over-counting of Dems. The LVCM is used for this purpose now (the Likely Voter Cut-off Model). This is a percentage that is added to the Repub totals to indicate what the likely vote will "actually" be based on results in the past (i.e., the vote as counted on the machines), and supposedly representing the "fact" that certain classes of Dems polled are unlikely to vote as compared with Repubs.
I trust Simon's statistics a lot more than I do the so-called polls nowadays. "Weighting" BTW is perfectly reasonable, in fact, necessary to take account of the sample used and the larger population that is being predicted or statistically measured. But "adjusting" the polls, both the pre-election polls and the exit polls, after a sensible and reasonable "weighting," is mostly just based on the "red shift" which is present now in almost all elections in the US as a direct result I think of the use of electronic voting machines.