2016 Postmortem
Related: About this forumRand American Life Tracker; O 49.1/R 45.3
https://mmicdata.rand.org/alp/index.php?page=electionEssentially sentiment is firming up on this tracker, intention to vote is going up for both and many more people believe that O will win (regardless of who they are actually voting for).
tarheelsunc
(2,117 posts)ProudToBeBlueInRhody
(16,399 posts)Don't they just poll the same people over and over or something?
abumbyanyothername
(2,711 posts)They ask three questions once a week of the sample:
Who do you intend to vote for?
How likely are you to vote?
Regardless of who you are voting for, who do you think will win?
This poll is a good indicator of sentiment shift and they show a chart each day of respondents who shifted from one candidate to the other (usually around 1 to 1.5%).
JackN415
(924 posts)I'm ignorant of RAND specific polling methodology, but I worked with RAND before and they have a certain philosophy of doing research. They tend to think out of the box even before the expression out-of-the-box became popular couple decades ago.
This polling method is like a vertical study in medical/psychological/sociological research. An example of this type of vertical study is:
http://www.theatlantic.com/magazine/archive/2009/06/what-makes-us-happy/307439/
which follow a group of men:
RAND current polling can tract voter sentiment change over time better than other polls that have a different sample population each time, at the sacrifice of absolute accuracy. In other words, it is more useful to gauge the trend than the actual percentage of how many supporting Romney, or Obama.
So, this poll is telling us that after the initial dazzling by the new Romney, some begin to move back to Obama, like children who ran after P. T Barnum's circus are coming home.
abumbyanyothername
(2,711 posts)but I always thought that Rand was just the CIA.
And I am not exactly comfortable having the CIA polling US elections. Although after seeing "Argo" last night I guess they're not entirely evil.
JackN415
(924 posts)various agencies, notably US AF, Army, and Homeland Security in the last decade. Contract work for CIA and NAS (Nat Sec. Agency) would be heavily classified and we would even know their existence. The content of the work for DoD may be classified, but the contract is public record.
They are an independent think tank org. They receive funding to do the research, but no, RAND researchers don't don a trench coat, wear sun glass, felt hat, and pack an Uzi submachine gun. . In fact, they would trip over their pencil sharpeners, and the only products they output are research papers, that sometimes can be use as a stack spacer to lift something up.
underpants
(182,877 posts)The link is great.
flamingdem
(39,321 posts)There're so straight they lack the bells and whistles!
Rand is okay, at least vs. some right wing run poll
abumbyanyothername
(2,711 posts)It is generally updated by the time I get up (out here in CA).
https://mmicdata.rand.org/alp/index.php?page=election#election-forecast
fujiyama
(15,185 posts)It was a very interesting study. A bit limited in its scope in that it only studied white males, but it was a product of its time. Nevertheless, it's amazing that they kept that study going for over 70 years.
Oh, and JFK was one the people in the study...
Jennicut
(25,415 posts)I may not buy that the President is ahead by 49 to 45 but I do think that he has increased his enthusiasm again and that Romney's "bounce" has settled. Reuters/Ipsos and IBC/Tipp show me that Romney is fading a bit. Interesting to see what Gallup results next week show us. Rasmussen will be counted on to give Repub support because of the way they poll.