OK, I'd taken the suggestion for a cooling off as a good one, but I see that our Statist justifiers are unwilling to join me. My perspective has always been, fire if fired upon. If not sooner.
. Therefore:
NY Times poll was discussed and discounted yesterday since at the time, it had no methodology or sampling. The cite at the link has been much improved, and now says this:
The Election Day poll was based on questionnaires completed by 2,376 voters as they left 40 randomly chosen precincts across the state on Tuesday. The polls were conducted by Edison Research of Somerville, N.J., for the National Election Pool, a consortium of ABC News, The Associated Press, CBS News, CNN, Fox News and NBC News. In theory, in 19 cases out of 20, the results from such polls should differ by no more than plus or minus 3 percentage points from what would have been obtained by seeking to interview all voters who cast ballots in Virginia. Results based on smaller subgroups, like demographic groupings, and shifts in results between polls have a larger potential of sampling errors. In addition to sampling error, the practical difficulties of conducting any survey of voter opinion on Election Day, like the reluctance of some voters to take time to fill out the questionnaire, may introduce other sources of error into the poll.
This is an improvement since now they've actually provided enough data to impeach the results. This is a hoot: "the practical difficulties of conducting any survey of voter opinion on Election Day, like the reluctance of some voters to take time to fill out the questionnaire, may introduce other sources of error into the poll". With this phrase, they note that the poll cannot be taken as reflective of the real world. Without data on true randomness of selection, and response rate data to quantify factors like self-selection, we can't assess the validity (to be more accurate, the generalizability) of the findings. Also, WHERE is the questionnaire? I think we are all familiar with push polls constructed to get the right answers (e.g., the ones the statist media will prefer).
Also, I must ask: The New York Times???!!! Does any reader here today belief their swill? IMO, their content today is best characterized as "All the News that Fits our Statism".
(The Liberty Crier cite of) WP Poll attempts more than the NY Times did yesterday: says methodology is presented at bottom of page. What I find is this "Preliminary exit poll results from 2,376 interviews of randomly selected voters as they exited voting places across the commonwealth of Virginia on Tuesday, Nov. 5". This is not a methodology but a throw away line (feel free to review social science, analytical, and sampling techniques to correct my view). We need to know distribution of interviews, questionnaire wording, which "random" responders were chosen and which rejected responses. Since the WP does not present this information, the result is not actual data but anecdotes lumped together to give a patina of 'scientism' to the propaganda.
Sorry, the last link did not come up for me. A little poking around found this, which is close enough I'll figure it is what you were looking at:
http://i2.cdn.turner.com/cnn/2013/images/11/05/va.gov.exit.polls.1120p.110513.v2.final[1].copy.pdf
If this is the one you were referring to, sorry, I don't see (scientific) data here -- again no methodology, sampling criteria (individual or locale), questionnaire wording, stats for those who may have self-selected or refused to participate. etc. Now CNN may
have such data though it's not in this pdf. If you find it, I am truly interested. It would be nice to discover that ANYONE in the media actually cares about the validity of what they publish. So please do add if you find it. If they have
not provided such, this is, in the social sciences analytical world and using the technical term, "wholly bogus".
Finally, and this is most amusing to me, it
appears (could be wrong) that all three of these 'different sources' used the exact same poll by Edison. Without more detail from the latter two, there can remain some uncertainty about this, however the numbers of sampled persons are identical in the first two, and my cross check of a few 'findings' with the CNN poll suggest the same basis for that too. So it appears to me that all three are actually the same report, and so subject to the problems generalizing the poll that the NY Times casually admits.
However, on the plus side, all three cited items (NY Slimes, Wash Compost, and Communist News Network) have nicely re-identified themselves as leading Statist purveyors of deception. And so, though we can't accept these cites as valid data, we can consider them good examples of propaganda. And entertainment.