Wikipedia:Reference desk/Archives/Miscellaneous/2016 October 7

= October 7 =

Psychic surgery in the fashion and modelling industry
Dear sir or madam, is psychic surgery being used in the fashion and modelling industry? Psychic surgery is a type of surgery in which the host person channels or allows the spirit of a deceased surgeon to take over the host's whole self in order to perform surgery on a patient. 213.205.252.81 (talk) 16:55, 7 October 2016 (UTC)


 * Psychic surgery has some info. It is considered medical fraud. SemanticMantis (talk) 17:14, 7 October 2016 (UTC)


 * Yes, and note that the description in our article is not the same thing listed by the OP. StuRat (talk) 19:32, 7 October 2016 (UTC)


 * Indeed. I don't think I've ever heard of what they're describing. Some guy channels Hawkeye Pierce and immediately becomes a great surgeon? Is anybody anywhere doing that? That's, like, wronger than wrong. Maybe even wronger than that. Matt Deres (talk) 19:58, 7 October 2016 (UTC)


 * And since Hawkeye was a fictional doctor, that even adds another layer of wrongness. StuRat (talk) 20:08, 7 October 2016 (UTC)


 * It could easily be dead wrong. Clarityfiend (talk) 21:17, 7 October 2016 (UTC)
 * There are various practitioners of Faith healing and Reiki and similar activities who call themselves as "Psychic Surgeons" and describe their spirit guides in the same terms as the OP, but some quick research seems to indicate that the "surgery" only takes place on the astral plane - "The spirit surgeon detects those parts of the spiritual body in which disease is developing, increases the vibration of the area affected using the physical healers [sic] hands and performs spiritual surgery in a different dimension removing the energetic imbalance." (I won't link to that particular Psychic Surgeon's website, but there are plenty of them out there).  Fortunately, I can't find anyone who actually wields a scalpel while in a trance state. Tevildo (talk) 21:25, 7 October 2016 (UTC)
 * I don't know whether John of God does but some of the testimonials and videos are quite scary! --TammyMoet (talk) 12:40, 8 October 2016 (UTC)
 * João de Deus (medium), rather than the sixteenth-century saint, I think. The former's chosen stage name may very well not be coincidental. Tevildo (talk) 14:58, 8 October 2016 (UTC)

US election - popular polls vs. the Electoral College
Without getting into party politics, I was just wondering what the use of popular polls are in the US elections. Because the president is elected by the Electoral College and not by the popular vote, how are popular vote results representative of anything meaningful when the margin between candidates is anything within a 60-40 split, and perhaps even greater?  DRosenbach  ( Talk 19:23, 7 October 2016 (UTC)


 * That's a great question, but you're probably not going to find the type of discussion you're hoping for here. Reference desk questions are generally closed-ended questions that have factual answers.  If you haven't already, have you read Electoral College (United States)?  RedLinkJ (talk) 19:27, 7 October 2016 (UTC)


 * As we get closer to the elections they do tend to do polling by state and show the math of which combinations of states would lead to victory by each candidate. However, it's expensive to do polls in every state, with enough people to be statistically valid, and not bias the Q in some way (like doing cold phone calls and only getting answers from people with enough free time to take polls).  While I agree that popular polls aren't as accurate of a predictor of who the winner will be, they aren't totally useless.  If all polls show one candidate is down 10% on the day of the election, they aren't likely to win, although perhaps if Republicans are down by 5% they might still win, due to a massive Republican Gerrymandering push after the last census (but note that Gerrymandering tends to be less effective over time, as voters move and/or voting behavior changes, hence the need to repeat the Gerrymandering effort every 10 years).  StuRat (talk) 19:37, 7 October 2016 (UTC)


 * But StuRat, gerrymandering isn't particularly relevant here; the reference to the Electoral College demonstrates that the question is focusing on the presidential election, and only in Maine and Nebraska (both of them small in votes and comparatively homogenous politically) are the votes calculated by any method other than winner-takes-all for the entire state. Nyttend (talk) 20:51, 7 October 2016 (UTC)


 * True, but I expanded from the Q to the more general issue of popular overall vote, versus polling by voting subdivision, whether that subdivision is a voting district or a state. However, your observation is relevant in explaining why states that vote Republican for other offices (where Gerrymandering is a major factor) often vote Democratic (where it's only a minor factor), for President.  StuRat (talk) 22:05, 7 October 2016 (UTC)


 * (ec) The popular vote does not determine the outcome, but it is strongly correlated with it. Only two out of the last 31 elections (if I did that right) had a winner with fewer popular votes than an opponent, and in both cases the difference was less than 1% (less than the margin of error of most polls).  (If you go back three more elections, to 1876, you get one where the difference was 3%.)
 * Unbiased polling is expensive. To get a confidence interval of 3% for the results, if I recall correctly, they typically have to interview about 1200 people (note that this number depends hardly at all on the size of the population; it's just a fixed 1200), and care has to be taken to avoid accidentally correlating your sample with factors that bias the outcome.
 * If you had to do that in fifty states, you would need 60,000 people (because the number does not depend on the population size). So you can see why this is not done so much for routine tracking polls.
 * Just the same, there are organizations that attempt to break down the probable vote state-by-state, using various models and methodologies that I don't know much about but would be interesting to explore. See FiveThirtyEight. --Trovatore (talk) 19:44, 7 October 2016 (UTC)


 * A similar issue arises in Westminster system elections using first past the post voting. The voters in each district elect a member of parliament (MP) and the MPs elect a prime minister. The simplest way to use an opinion poll is as a swingometer:
 * IF
 * opinion poll shows party Y up 2% and party X down 4% since the last election
 * AND 15 MPs from party X finished less than 6% (= 2+4) ahead of the candidate from party Y at the last election
 * AND assuming a uniform swing in every district
 * THEN party Y will capture those 15 districts from party X.
 * Obviously #3 is a simplistic assumption. jnestorius(talk) 19:50, 8 October 2016 (UTC)