Exit Poll Comentary

BeastBeast ArmonkyiMembers, Constellation Join Date: 2003-04-21 Member: 15731Posts: 2,201
edited November 2004 in Discussions
By yours truely.
Well I was bored. So I looked at the CNN exit polls. Some things REALLY caught my attention. So I wrote a short summary. Here follows a report/rant written by a not-so-average Englishman/wolf thing (hah XD), written at 12am on a sugar fueled brain, so please forgive typos, and any direct opinions (shock horror :o) that may exist here. And the rant thingy at the end. And my odd writing style and formatting >.>. Infact, just read the thing and think for a while before posting and make sure I haven't eaten your brain in the process or something.

(original source: http://www.cnn.com/ELECTION/2004/pages/res...0/epolls.0.html )

My report:
Interesting Notes From CNN Exit poll statistics:

Vote by race (generalised)
White: (77%) : majority voted for bush (59%)
Non-white:(22%): Majority voted for kerry (71%)
*comentary: big surprise after the last election, huh? (sarcasm)

Income Related:
Sallary less than 100k/year(82%): slight majority for kerry (50%)
Sallary more than 100k/year(18%): majority for bush (58%)
*comentary: looks like bush's rich cronies simply love him, and how his tax cuts gave them the easy life.

Religion:
Christian: (81%): Majority for bush (55%)
Nonchristian (20%): Majority for kerry (72%)
*oh yeah, big surprise here ¬_¬.

Military:
Served (18%): majority for bush (57%)
Not served (82%): slight majority for kerry (50%)
*Interesting statistics here. I don't like it to be honest.

sex pref:
straight (96%): Majority for bush (53%)
bi/****/lesbian (4%): majority for kerry (77%)
*big duh here, even though kerry doesnt really support **** rights either from what ive seen.

Most important issue:
taxes (5%): majority for bush (57%)
education (4%): majority for kerry (77%)
iraq (15%): majority for kerry (73%)
Terrorism (19%): Majority for bush (86%)
Economy (20%): majority for kerry (80%)
Moral values (22%): majority for bush (80%)
healthcare (8%): majority for kerry (77%)
*looks like only 42% cared about the important things (health, education, economy).

opinion of how good bush is doing:
like (53%): majority for bush (89%)
dislike(46%): majority for kerry (89%)
-love (33%)
-middle (32%)
-hate (34%)
*the majority (compared to "dont care" and "love") hate bush... somewhat. but an almost equal ammount love him. This is somewhat disturbing.

(HA!)most important quality:
Cares about people (9%): Majority for kerry (75%)
religious faith (8%): majority for bush (91%)
honesty (11%): majority for bush (70%)
strong leader (17%): majority for bush (87%)
inteligent(7%): majority for kerry (91%)
will bring change (24%): majority for kerry (95%)
clear stand on issue (17%): majority for bush (79%)
*what people need is a leader who cares about the people, is inteligent, and will bring change. (40%).. and those voted kerry. It seems those who voted bush dont care about that, and would rather have an strong, dumb, christian leader, who is honest and clear about how much he is evil...bah.

terrorism opinion on us:
Safer (54%): majority for bush (79%)
less safe (41%): majority for kerry (85%)
*Looks like those who think america is less safe voted for kerry.. poor bush... what a fool.

decision to goto iraq:
Approve (51%): Majority for bush (85%)
disapprove (45%): majority for kerry (87%)
how iraq is going:
good (44%): majority for bush (90%)
bad (52%): majoirty for kerry (82%)
* Interseting but predictable results. Those who think invading iraq was the best thing ever and it remains so voted bush. those who think not voted kerry.

how "average family's finances" are:
Better (32%): majority for bush (80%)
worse (28%): majority for kerry (79%)
same (39%): slight majority for kerry (50%)
*? Again very interesting statistics. most family's finances were the same or worse, and voted for kerry.

opinion of bush:
Hate (49%): majority for kerry (89%)
Love (48%): majority for bush (94%)
*so many hate bush, yet.. whats this? they didnt vote for him as much as those who loved bush voted bush... WTH? if you really hate him that much WHY THE HELL VOTE FOR HIM?

opinion of national economy:
good (47%) : majority for bush (87%)
bad (52%): majority for kerry (79%)
*the people have spoken! bush has mucked up the economy! why did more people vote for him then?

trust kerry with economy?:
yes (45%): Majority for kerry (88%)
no (53%): majority for bush (86%)
trust bush with economy?
yes (40%): majority for kerry (93%)
no (58%): majority for bush (84%)
*if you can't trust kerry or bush with the economy, why fecking vote for them? there are more candidates than bush and kerry.. shame they get totaly eclipsed by the two party system.

America going in right direction?
yes (49%) : majority for bush (89%)
no (46%): majority for kerry (86%)
*more interesting statistics. Have a LOOONG look at america's economy and tell me its going in a good direction. tell me why one British pound gets me almost $2 sometimes since bush was president?

job situation in av voters area:
better (23%) : majority for bush (90%)
worse (43%): majority for kerry (81%)
about the same (34%): majority for bush 70%)
*more somewhat interesting results. seems like the majority think their job sitation in their area got worse, and surprise surprise, they voted for kerry!

bush tax cuts for the economy:
good (41%): majority for bush (92%)
bad! (32%): majority for kerry (92%)
no dif (25%): majority for kerry (54%)
*Have a look again at american economy. yes.

iraq war made us more secure?:
yes (46%): majority for bush (90%)
no! (52%): majority for kerry (80%)
*d'oh! bush got found out again?

anyone in household lost a job?
yes (33%!): majority for kerry (64%)
no (67%): majority for bush (61%)
*33% of households have someone who lost a job. this is shocking. they also voted for kerry, not as shocking.

have average voter lost job?
yes (17%!): majority for kerry (63%)
no (83%): majority for bush (56%)
*yet again. 17% lost a job with bush in office. They voted for kerry as expected.

Summary:
It seems people really hate both candidates. Why did people vote for one of them? If people hate bush so much, why did some of those who hate him VOTE FOR HIM? it's quite frankly is insane. The next four years are going to be VERY difficult for america. For anyone who supported kerry/other parties, now is the time to make sure bush doesnt mess up america any further. For those who support bush, keep your voted leader in check. if you do nothing bush will be able to do anything. Now is the time to defend your country and it's constitution. <RANT!> DOING NOTHING ACCOMPLISHES NOTHING AND WILL FURTHER SERVE TO RUIN AMERICA. :E</RANT!>

-end of report-

Sooo... opinions? Agreements? disagreements? confusion? (note: keep this related to the exit polls and thusly)
IPB Image
«1

Comments

  • SkulkBaitSkulkBait Members Join Date: 2003-02-11 Member: 13423Posts: 2,543
    QUOTE
    It seems people really hate both candidates. Why did people vote for one of them?


    Some of us didn't. Anyway, I think the statistics are interesting but your comments on them are quite inlfamatory.
    1.04ever
  • ForlornForlorn Banned Join Date: 2002-11-01 Member: 2634Posts: 6,495
    edited November 2004

    - Take a stats class before you comment on them. Just a high school course will do you a lot of good.

    - Exit Polls use conviencing sampling. This is bad. This is not going to give you crap for accurate results.




    You wasted your time with your little analysis there buddy, since you based everything off the rediculous exit polls which don't mean jack.
    Post edited by Unknown User on
    user posted imageuser posted image#sci-ns Want decent mini-maps?

    Bad NS clan scene?Pm me on IRC!
  • KeyserKeyser Members Join Date: 2003-02-15 Member: 13591Posts: 684
    QUOTE (Forlorn @ Nov 6 2004, 10:28 PM)
    - Exit Polls use conviencing sampling. This is bad. This is not going to give you crap for accurate results.

    Usually when arguing a point you have to back up statements like this with fact. Since you decided to ignore my other post...

    What are the grievous sources of error for exit polls?

    Please answer this first so you don't look like too silly making wild assertions.
    Keys [NAPT]
    #pandas
  • The_FinchThe_Finch Members Join Date: 2002-11-13 Member: 8498Posts: 661
    Exit polls usually aren't randomly selected, which accounts for a large standard error. However, the numbers you put up aren't surprising.

    QUOTE
    Vote by race (generalised)
    White: (77%) : majority voted for bush (59%)
    Non-white:(22%): Majority voted for kerry (71%)
    *comentary: big surprise after the last election, huh? (sarcasm)


    What's surprising is that Kerry only pulled in 70% of the minority vote. Minorities are traditionally a huge voting block for Democrats and it would seem that Republicans are starting to successfully reach out to minority communities. You know, being a Republican doesn't automatically mean you're a member of the KKK.

    QUOTE
    Income Related:
    Sallary less than 100k/year(82%): slight majority for kerry (50%)
    Sallary more than 100k/year(18%): majority for bush (58%)
    *comentary: looks like bush's rich cronies simply love him, and how his tax cuts gave them the easy life.


    Kerry still drew 42% of the "rich" vote and the 100k or less was split evenly. The wealthy are probably more likely to vote Republican because they a.) don't benefit from a lot of social programs and b.) benefit from the pro-business stance of the Republican party. It's not cronyism, it's self-interest. People look out for #1.

    QUOTE
    Religion:
    Christian: (81%): Majority for bush (55%)
    Nonchristian (20%): Majority for kerry (72%)
    *oh yeah, big surprise here ¬_¬.


    101% of the population? Anyway, it's not surprising. Bush got 55% of the vote, which means that Kerry got 45%. Both wear their faith on their sleeve, but Bush does it better. Given the standard error for exit polls, I could easily see this as being split evenly. However, you won't see any complaints about "fundies" voting for Kerry.

    QUOTE
    Military:
    Served (18%): majority for bush (57%)
    Not served (82%): slight majority for kerry (50%)
    *Interesting statistics here. I don't like it to be honest.


    Not surprising at all. In the case of active duty, the percentage is actually much higher in favor of Bush. Military members tend towards traditionalism, which the military has in spades. They also tend to be pro-gun.

    QUOTE
    sex pref:
    straight (96%): Majority for bush (53%)
    bi/****/lesbian (4%): majority for kerry (77%)
    *big duh here, even though kerry doesnt really support **** rights either from what ive seen.


    Bush has a slight lead among straight people? At any rate, it's within the margin of error. Homosexuals probably aren't going to vote for somebody who wants to pass a constitutional amendment stating marriage is between a man and a woman.

    QUOTE
    Most important issue:
    taxes (5%): majority for bush (57%)
    education (4%): majority for kerry (77%)
    iraq (15%): majority for kerry (73%)
    Terrorism (19%): Majority for bush (86%)
    Economy (20%): majority for kerry (80%)
    Moral values (22%): majority for bush (80%)
    healthcare (8%): majority for kerry (77%)
    *looks like only 42% cared about the important things (health, education, economy).


    That's pretty subjective. And misleading. Just because somebody rated terrorism as the most important, doesn't mean they think that education is unimportant.

    QUOTE
    opinion of how good bush is doing:
    like (53%): majority for bush (89%)
    dislike(46%): majority for kerry (89%)
    -love (33%)
    -middle (32%)
    -hate (34%)
    *the majority (compared to "dont care" and "love") hate bush... somewhat. but an almost equal ammount love him. This is somewhat disturbing.


    You love Bush or hate him. Kerry doesn't even seem to be a factor in this question, which is probably one of the factors in his loss.

    QUOTE
    (HA!)most important quality:
    Cares about people (9%): Majority for kerry (75%)
    religious faith (8%): majority for bush (91%)
    honesty (11%): majority for bush (70%)
    strong leader (17%): majority for bush (87%)
    inteligent(7%): majority for kerry (91%)
    will bring change (24%): majority for kerry (95%)
    clear stand on issue (17%): majority for bush (79%)
    *what people need is a leader who cares about the people, is inteligent, and will bring change. (40%).. and those voted kerry. It seems those who voted bush dont care about that, and would rather have an strong, dumb, christian leader, who is honest and clear about how much he is evil...bah.


    Again with the deceptive language. Most important doesn't mean the others are unimportant. I think that honesty is important in politicians, but I didn't vote for Bush. You need to get these stereotypes out of your head. Not all Bush supporters are inbred yokels who want to beat bibles and homosexuals.

    The rest is pretty much repeats of the above questions, but you get the point. Just because somebody disagrees with you, it doesn't make them stupid. Keep in mind that America's economy is getting better. In October over 300,000 jobs were created, which was almost double expectations. The markets are relatively steady and things are going well enough for the Fed to continue their plan to hedge inflation. I don't think that Bush has much to do with it, but the economy is recovering and doing so quite nicely.

    Ranting about how "Bush is teh debil!" doesn't mean that Kerry was a good choice. I voted Badnarik because he best represents my views.
    QUOTE (X Stickman)
    America's Army taught me that I'm more likely to be shot in the back by my own teammates, then have my sexuality insulted as well as accusations made towards my mother's sex life. If it's a recruitment tool, it's a damn poor one.
  • ForlornForlorn Banned Join Date: 2002-11-01 Member: 2634Posts: 6,495
    QUOTE (Keyser59 @ Nov 6 2004, 11:38 PM)
    QUOTE (Forlorn @ Nov 6 2004, 10:28 PM)
    -  Exit Polls use conviencing sampling.  This is bad.  This is not going to give you crap for accurate results.

    Usually when arguing a point you have to back up statements like this with fact. Since you decided to ignore my other post...

    What are the grievous sources of error for exit polls?

    Please answer this first so you don't look like too silly making wild assertions.

    Not only is The Finch correct when he says:
    QUOTE
    Exit polls usually aren't randomly selected, which accounts for a large standard error. However, the numbers you put up aren't surprising.


    But also any poll which allows a person to choose to participate on, such as a poll where you call into for TV is bad.

    The best types of polls are the ones where you call up random houses and poll them.

    Exit polls, just look at them, they look into one particular area and interview them. Did these exit polls, evidently they were all held in democratic areas where you get huge Kerry results. They probably went to the urban centers of every state, which typically are democratic, so of course they are going to say Kerry will win!

    They were done very badly.
    user posted imageuser posted image#sci-ns Want decent mini-maps?

    Bad NS clan scene?Pm me on IRC!
  • KeyserKeyser Members Join Date: 2003-02-15 Member: 13591Posts: 684
    How are polls that call up houses good when they are
    1. allowed to choose to participate or not? (as you say)
    2. home-owning families/persons do not nearly fully represent the whole spectre of voters?

    (By the way, Harris also held a phone-survey, which Kerry won on Nov 2nd)

    Do you have any proof that these exit polls were only held in democratic areas? This is surprising when there are done by totally independent corporations like Harris who show extreme professionalism. Of course, taking a course in statistics, I know this. I also know that they scale the results of each county to give an accurate representation of the population. I think someone needs to possibly re-take a course in statistics.

    Please explain how exit polls aren't randomly selected, and also add some suggestions on how exit polls could be conducted better in the future.
    Keys [NAPT]
    #pandas
  • ForlornForlorn Banned Join Date: 2002-11-01 Member: 2634Posts: 6,495
    QUOTE (Keyser59 @ Nov 7 2004, 10:45 AM)
    How are polls that call up houses good when they are
    1. allowed to choose to participate or not? (as you say)
    2. home-owning families/persons do not nearly fully represent the whole spectre of voters?

    (By the way, Harris also held a phone-survey, which Kerry won on Nov 2nd)

    Do you have any proof that these exit polls were only held in democratic areas? This is surprising when there are done by totally independent corporations like Harris who show extreme professionalism. Of course, taking a course in statistics, I know this. I also know that they scale the results of each county to give an accurate representation of the population. I think someone needs to possibly re-take a course in statistics.

    Please explain how exit polls aren't randomly selected, and also add some suggestions on how exit polls could be conducted better in the future.

    1. Easy, because phone polls are done like this:

    - "Excuse me Hi I'm doing this poll could I please speak to the person with the most recent birthday?"

    That's how phone polls are done today. So it randomizes it like that. Furthermore they do not choose if they participate; they get lucky enough to be called.

    Unless there's some kind of fact where democrats are more likely to participate in a poll over a republican then results should be very accurate.

    2. No actually they pretty much do



    And I'm just assuming that's how the exit polls were held, after all they were incredibly off, all saying Kerry was gonna win but just look at it. Obviously these polls weren't done well AT ALL.


    Anyhow here is a site that lists lots of phone surveys (And it lists Bush as the winner too... funny that)

    http://www.realclearpolitics.com/polls.html
    user posted imageuser posted image#sci-ns Want decent mini-maps?

    Bad NS clan scene?Pm me on IRC!
  • CForresterCForrester P0rk(h0p Members, Constellation Join Date: 2002-10-05 Member: 1439Posts: 3,637 Advanced user
    QUOTE (Forlorn @ Nov 7 2004, 12:14 PM)
    Furthermore they do not choose if they participate; they get lucky enough to be called.

    Maybe you should take a good, long look at your phone and tell me if you have a button that you can press to end the call.
  • KeyserKeyser Members Join Date: 2003-02-15 Member: 13591Posts: 684
    QUOTE (Forlorn @ Nov 7 2004, 11:14 AM)
    1. Easy, because phone polls are done like this:

    - "Excuse me Hi I'm doing this poll could I please speak to the person with the most recent birthday?"

    That's how phone polls are done today. So it randomizes it like that. Furthermore they do not choose if they participate; they get lucky enough to be called.

    2. No actually they pretty much do

    Yes, I know how phone polls are done. They aren't forced to participate by any means, and you still haven't explained how home-owning families represent all voters. You essentially just ignored everything that I said. confused-fix.gif

    And please, again, when arguing your point, you have to give reasons on a thesis, example being your point that "Home-owners represent the entire population." A good amount of minorities are left out of these polls. Basic statistics.

    Exit polls are conducted by interviewers approaching a random selection of people as they leave the ballots, which I assume you didn't know.

    Seeing as doing the exit polls have an equal chance of getting all voters, not just home-owners, they should be more accurate, right? Basic statistics.

    QUOTE
    And I'm just assuming that's how the exit polls were held, after all they were incredibly off, all saying Kerry was gonna win but just look at it. Obviously these polls weren't done well AT ALL.


    I'm pointing out the discrepancies between the exit polls of the 2000 elections and this year's elections. Exit polls have proven to be pretty accurate in the past, and even now in states where paper ballots are still being used. However, with the introduction of electronic voting, something is clearly different.



    Keys [NAPT]
    #pandas
  • [WHO]Them[WHO]Them You can call me Dave Members, Constellation Join Date: 2002-12-11 Member: 10593Posts: 2,813 Fully active user
    QUOTE (Keyser59 @ Nov 7 2004, 10:15 AM)
    ...and you still haven't explained how home-owning families represent all voters. You essentially just ignored everything that I said. confused-fix.gif

    I'm sure that when he said that they call up random "houses", he meant to say "homes". Last time I checked, most apartments do have a phone doohickey.
    image
    Zoc.php FTW
    "Build a tent and say the world is dry." -Tay Zonday
  • DubbilexDubbilex Chump Members Join Date: 2002-11-24 Member: 9799Posts: 2,016 Advanced user
    edited November 2004
    This is no Bush-bashing, nor a media-bashing thread. The discussion centers around polling methods. Skeletons in various president's closests don't factor into this. Also, your question has already been posed.
    what
  • Nemesis_ZeroNemesis_Zero Old European Members, Retired Developer, NS1 Playtester, Constellation Join Date: 2002-01-25 Member: 75Posts: 12,841
    By the way, last time I checked, Bush's advantage over Kerry was at less than three percent. Seeing the size of the American population and 'unfactorable' issues such as the relative percentage of non-voters in the case of pre-election polling, I'd argue that this year's election's outcome is no indicator for the validity of a polling method in either direction.
    QUOTE
    And if I haven't seen further, it's because those bloody giants have blocked my sight.
  • TheWizardTheWizard Members, Constellation Join Date: 2002-12-11 Member: 10553Posts: 1,646 Advanced user
    QUOTE
    *looks like only 42% cared about the important things (health, education, economy).



    The problem with this assumption is that most US citizens do not consider these topics to be the 'important' things.

    Anyone in the US can get emergency healthcare and frankly, I do not feel that the federal government has a responsibility to furnish Americans with healthcare.


    Life liberty and the pursuit of happiness. I missed the part where it said government sponsored medicine.
    Sig de-borkified. -- Nem

    I had a sig? --TheWizard
  • ekentekent Members Join Date: 2002-11-08 Member: 7801Posts: 781
    Before this descends into the depths of partisan bickering that it can't escape, I hope I can provide some clarity about telephone polling. I conducted telephone surveys for a commercial company for a little over a year when I was in high school. Most of our surveys were political.

    All of the surveys used random sampling from a database of phone numbers. Where the database comes from depends on the poll. For instance, teacher's unions conduct a lot of polls. Their database comes from their member list. All political surveys I conducted had a database pulled from the public voter registration records.

    A large amount of the numbers in the databases were junk - disconnected phones, businesses, or different resident. When you get a person on the other line you were usually lucky to get better than one in ten people to accept the survey. I will leave the speculation as to whether or not a particular political persuasion is more likely to answer the surveys up to you.

    The political surveys were frequently long, 10-20 minutes (teacher's union surveys often lasted 40-60 minutes, but teachers talk like hell tounge.gif), and often I would have to really whine and beg to get people to finish the surveys. Even when they were finished, there is always a demographics section (optional by law) at the end. Frequently people were uncomfortable disclosing their religious affiliation and income over the phone. Infrequently, people were unwilling to tell you their ethnicity. I don't need to point out that surveys with an incomplete demographics section were much less useful.

    Anyway this should call to light that telephone interviews/surveys are not that much superior to specifically exit polls and non-specifically many other types of random-sample polling. The largest factor is the validity of the statistical model that is used to interpret the data.
    The care of human life and happiness, and not their destruction, is the first and only object of good government.
    - Thomas Jefferson
  • KeyserKeyser Members Join Date: 2003-02-15 Member: 13591Posts: 684
    Thanks for the insight kent. It is interesting to get a more-indepth description of how telephone polls are conducted.

    However, I am still still convinced that exit polls are a more accurate poll. You are much less likely to reach a person living alone working 8 hours a day than a stay-at home mother. Also many people go under the unlisted section for one reason or another, and while I'm not sure how much of an impact this will have, it would have to skew the polling one way or another. To add, I don't know how you contact college students, but it seems like that would be slightly harder.

    If you could clear up any of these questions I would be quite obliged. smile-fix.gif

    With exit polls, you can grab any of those voters with an equal opportunity, at least how I see it.
    Keys [NAPT]
    #pandas
  • ForlornForlorn Banned Join Date: 2002-11-01 Member: 2634Posts: 6,495
    QUOTE (Keyser59 @ Nov 7 2004, 10:18 PM)
    Thanks for the insight kent. It is interesting to get a more-indepth description of how telephone polls are conducted.

    However, I am still still convinced that exit polls are a more accurate poll. You are much less likely to reach a person living alone working 8 hours a day than a stay-at home mother. Also many people go under the unlisted section for one reason or another, and while I'm not sure how much of an impact this will have, it would have to skew the polling one way or another. To add, I don't know how you contact college students, but it seems like that would be slightly harder.

    If you could clear up any of these questions I would be quite obliged. smile-fix.gif

    With exit polls, you can grab any of those voters with an equal opportunity, at least how I see it.

    Look when you do an exit poll you are doing a specific demographic.

    Exit polls therefore are flawed, they are only in certain parts of the state. Therefore they have the chance to be dead wrong.


    Meanwhile, homeowners (as opposed to the homeless) are chosen at random, they either accept or not, and that's it! Their location varies enough to give very very accurate results.


    http://www.realclearpolitics.com/polls.html



    Kent, you fail to realize that not all survey's are held like that. I currently work for a polling center, and they hold polls that vary in length and types of questions. Sometimes they ask if for specific bits of info but in case of these political polls; they were short 3 minute surveys.
    user posted imageuser posted image#sci-ns Want decent mini-maps?

    Bad NS clan scene?Pm me on IRC!
  • KeyserKeyser Members Join Date: 2003-02-15 Member: 13591Posts: 684
    edited November 2004
    QUOTE (Forlorn @ Nov 7 2004, 11:56 PM)
    Look when you do an exit poll you are doing a specific demographic.

    QUOTE (Keyser)
    Exit polls are conducted by interviewers approaching a random selection of people as they leave the ballots, which I assume you didn't know.


    QUOTE (Forlorn)
    Exit polls therefore are flawed, they are only in certain parts of the state. Therefore they have the chance to be dead wrong.
    QUOTE

    On Election Day, temporary workers hired by VNS conducted exit polls in about 1,400 precincts across all states. When actual votes began coming in, VNS checked results in more than 3,500 precincts scientifically selected to predict final results. The actual vote was compared to the poll results to confirm the polls' accuracy.



    QUOTE (Forlorn)
    Meanwhile, homeowners (as opposed to the homeless) are chosen at random, they either accept or not, and that's it! Their location varies enough to give very very accurate results.


    QUOTE (Keyser)
    However, I am still still convinced that exit polls are a more accurate poll. You are much less likely to reach a person living alone working 8 hours a day than a stay-at home mother. Also many people go under the unlisted section for one reason or another, and while I'm not sure how much of an impact this will have, it would have to skew the polling one way or another. To add, I don't know how you contact college students, but it seems like that would be slightly harder.


    Heh, look at that, I didn't even have to type any new material. Ignoring arguments is not a good way to debate, Forlorn.

    EDIT: Found another interesting factoid on that site.

    QUOTE

    How accurate were the exit polls around the country for this presidential election?
    The exit polls predicted the percentage of votes in each state that each presidential candidate would receive. The differences between the exit poll percentages and the actual percentages the candidates received ranged from 0% to 2.6%. In only nine states was the difference greater than 1%, and in only one state (Alaska) was the difference greater than 2%.


    This is in regards to the 2000 election with paper ballots.
    Keys [NAPT]
    #pandas
  • ThE_HeRoThE_HeRo Members Join Date: 2003-01-25 Member: 12723Posts: 1,599
    On the "internets", I could find articles arguing that oranges are purple.
    -tex
  • SkulkBaitSkulkBait Members Join Date: 2003-02-11 Member: 13423Posts: 2,543
    This is interesting:

    user posted image
    1.04ever
  • KeyserKeyser Members Join Date: 2003-02-15 Member: 13591Posts: 684
    Make sure to try to view it objectively though.

    You'll notice that they scale it usually on an axis of 40-60%, making the graph look a bit more significant than it really is.
    Keys [NAPT]
    #pandas
  • SkulkBaitSkulkBait Members Join Date: 2003-02-11 Member: 13423Posts: 2,543
    edited November 2004
    There are many problems with that graph (for instance, it never shows Bush loosing votes, I can only assume that certain states were chosen specifically for this reason), but I still found the results interesting.
    1.04ever
  • moultanomoultano Creator of ns_shiva. Members, NS1 Playtester, Contributor, Constellation, NS2 Playtester, Squad Five Blue, Reinforced - Shadow, WC 2013 - Gold, NS2 Community Developer, Pistachionauts Join Date: 2002-12-14 Member: 10806Posts: 4,219 Advanced user
    edited November 2004
    QUOTE (SkulkBait @ Nov 8 2004, 05:32 PM)
    There are many problems with that graph (for instance, it never shows Bush loosing votes, I can only assume that certain states were chosen specifically for this reason), but I still found the results interesting.

    I'm not sure that there were states in which Bush had more votes in the exit poll than he did in the final tally. Everything I've read said that Kerry was up in the exit polls by 3-4 points in the swing states.

    However, as compelling as this looks, there is a very good chance that it is just the result of response bias. There are a lot of places in the country in which voting for Bush isn't something people would want to disclose because it isn't popular. Kerry supporters tend to be more vocal and defiant. Also, a lot of the "get the government out of my life" people probably voted for Bush in higher numbers than they voted for Kerry, and a significant percentage of them, I suspect, wouldn't respond to an exit poll.

    That said, we can't be sure unless we can somehow verify the results independently. If this is important to you, put your money where your mouth is, and donate to http://www.blackboxvoting.org/
    Post edited by Unknown User on
  • LegatLegat Members Join Date: 2003-07-02 Member: 17868Posts: 817
    edited November 2004
    QUOTE
    honesty (11%): majority for bush (70%)


    Does anybody else see the irony in this one?
  • KradKrad Members Join Date: 2004-08-26 Member: 30914Posts: 273
    All of those comments you put down after the "statistics" were completely unneccesary. Put the facts down and let everyone else make the correlation.

    As for most of your statistics... Yeah, for one, 50% is not a majority. Not even a slight majority. It's dead even. 100-50=50. And yes, exit polls are screwed up majorly, so most of those polls are probably dead wrong (For example, I, for one, didn't pick up any of the political poller's phone calls during the election, so our house remains untallied).

    More to come later when I'm done with school. (RAWR!!!!!1!1111oneone)
  • BathroomMonkeyBathroomMonkey Feces-hurling Monkey Boy Members, Retired Developer, NS1 Playtester, Contributor Join Date: 2002-01-25 Member: 78Posts: 1,345
    A defense of exit polls, by Republican **** Morris:

    QUOTE
    Exit polls are almost never wrong. They eliminate the two major potential fallacies in survey research by correctly separating actual voters from those who pretend they will cast ballots but never do and by substituting actual observation for guesswork in judging the relative turnout of different parts of the state.

    So reliable are the surveys that actually tap voters as they leave the polling places that they are used as guides to the relative honesty of elections in Third World countries. When I worked on Vicente Fox’s campaign in Mexico, for example, I was so fearful that the governing PRI would steal the election that I had the campaign commission two U.S. firms to conduct exit polls to be released immediately after the polls closed to foreclose the possibility of finagling with the returns. When the polls announced a seven-point Fox victory, mobs thronged the streets in a joyous celebration within minutes that made fraud in the actual counting impossible.
    Are you sure you read it in a book? Are you sure it wasn’t . . . nothing?
  • DiscoZombieDiscoZombie Members Join Date: 2003-08-05 Member: 18951Posts: 4,521 Advanced user
    QUOTE (Beast @ Nov 6 2004, 07:31 PM)
    Most important issue:
    taxes (5%): majority for bush (57%)
    education (4%): majority for kerry (77%)
    iraq (15%): majority for kerry (73%)
    Terrorism (19%): Majority for bush (86%)
    Economy (20%): majority for kerry (80%)
    Moral values (22%): majority for bush (80%)
    healthcare (8%): majority for kerry (77%)
    *looks like only 42% cared about the important things (health, education, economy).

    this is the scariest thing IMO. the largest % consider 'moral values' the most important issue in their president...

    yes. whether homosexuals can marry and fetuses can be aborted is more important than education, healthcare, and taxes combined.

    I assume that's what people mean when they check 'moral values'. 450% more people think these things are more important than educating our children. Scary indeed.

    I bet a similar poll even in an islamic extremist country would yield more people caring about education than here in the US...
    image
  • ForlornForlorn Banned Join Date: 2002-11-01 Member: 2634Posts: 6,495
    QUOTE (Keyser59 @ Nov 8 2004, 02:17 AM)
    QUOTE (Forlorn @ Nov 7 2004, 11:56 PM)
    Look when you do an exit poll you are doing a specific demographic.

    QUOTE (Keyser)
    Exit polls are conducted by interviewers approaching a random selection of people as they leave the ballots, which I assume you didn't know.


    QUOTE (Forlorn)
    Exit polls therefore are flawed, they are only in certain parts of the state. Therefore they have the chance to be dead wrong.
    QUOTE

    On Election Day, temporary workers hired by VNS conducted exit polls in about 1,400 precincts across all states. When actual votes began coming in, VNS checked results in more than 3,500 precincts scientifically selected to predict final results. The actual vote was compared to the poll results to confirm the polls' accuracy.



    QUOTE (Forlorn)
    Meanwhile, homeowners (as opposed to the homeless) are chosen at random, they either accept or not, and that's it! Their location varies enough to give very very accurate results.


    QUOTE (Keyser)
    However, I am still still convinced that exit polls are a more accurate poll. You are much less likely to reach a person living alone working 8 hours a day than a stay-at home mother. Also many people go under the unlisted section for one reason or another, and while I'm not sure how much of an impact this will have, it would have to skew the polling one way or another. To add, I don't know how you contact college students, but it seems like that would be slightly harder.


    Heh, look at that, I didn't even have to type any new material. Ignoring arguments is not a good way to debate, Forlorn.

    EDIT: Found another interesting factoid on that site.

    QUOTE

    How accurate were the exit polls around the country for this presidential election?
    The exit polls predicted the percentage of votes in each state that each presidential candidate would receive. The differences between the exit poll percentages and the actual percentages the candidates received ranged from 0% to 2.6%. In only nine states was the difference greater than 1%, and in only one state (Alaska) was the difference greater than 2%.


    This is in regards to the 2000 election with paper ballots.

    Excuse me but do not understand by demographic, I mean demographic as in where you take the votes?

    You get the majority of your exit polls out of a city in state, and of couse it's going to be mostly for Kerry as those area's are democratic. But they ignore the many many other little counties around it who get little say.

    Obviously by looking at the actual poll numbers and the exit polls we see they did a sorry job of where they conducted the polls.


    What I find absolutely hilarious about this whole exit poll issue; instead of looking into possible flaws with the exit polls, the democrats look into the actual voting process. You are viewing facts through your own biased filter and frankly it's pretty disturbing.
    user posted imageuser posted image#sci-ns Want decent mini-maps?

    Bad NS clan scene?Pm me on IRC!
  • KeyserKeyser Members Join Date: 2003-02-15 Member: 13591Posts: 684
    edited November 2004
    QUOTE (Forlorn)
    You get the majority of your exit polls out of a city in state, and of couse it's going to be mostly for Kerry as those area's are democratic. But they ignore the many many other little counties around it who get little say.


    QUOTE
    On Election Day, temporary workers hired by VNS conducted exit polls in about 1,400 precincts across all states. When actual votes began coming in, VNS checked results in more than 3,500 precincts scientifically selected to predict final results. The actual vote was compared to the poll results to confirm the polls' accuracy.


    I'm not sure how many more times I'm going to say this, but since you don't seem to understand...

    That quote means that they don't just head to all the cities and tally them up (as you seem to think what all statistics incorporate for some absurd reason). They do exit polls and scale them on the population of the precinct, and then compare them with the results. Last year results were extremely close to the exit polls, 80% of the states error fell under 1% error. Only Alaska was above 2%. From here you cannot say that they only take polls in urban areas, because that simply doesn't make sense with these statistics. Please stop repeating yourself.

    Now that we have gotten that out of the way...

    QUOTE
    What I find absolutely hilarious about this whole exit poll issue; instead of looking into possible flaws with the exit polls, the democrats look into the actual voting process. You are viewing facts through your own biased filter and frankly it's pretty disturbing.




    I am not suggesting conspiracy so there is no need to throw this issue under partisanship. The fact stands that all the (nine) swing states fell under Bush with an exit poll discrepancy greater than 2-3% as opposed to the exit polls in 2000 when they were far under 2%. That alone is not enough to warrant an investigation, but all these swing states have incorporated electronic voting, while in the paper ballot states the exit polls follow the 2000 trend.

    So I would appreciate you not accusing me of partisanship when I have offered proof and you continue making unsupported statements.

    EDIT: I also support blackboxvoting.org in their efforts.
    Keys [NAPT]
    #pandas
  • ekentekent Members Join Date: 2002-11-08 Member: 7801Posts: 781
    QUOTE (Forlorn @ Nov 7 2004, 08:56 PM)
    Kent, you fail to realize that not all survey's are held like that. I currently work for a polling center, and they hold polls that vary in length and types of questions. Sometimes they ask if for specific bits of info but in case of these political polls; they were short 3 minute surveys.

    I don't fail to realize anything. Your incendiary language aside, I gave polls that varied in length from 2 to 60 minutes. The consumer surveys tended to be the shortest, the political surveys tended to be longer, and the union polls were always the longest. I was trying to explain the limitations of telephone polling and I may have over-sold my case a bit, if you choose to read it that way. From how I understand polling, and let me make this clearer for you -

    telephone polling has limitations not limited to what types of people answer the phone and the small sample sizes; exit polling has limitations too. The strength of a poll lies in how successful its statistical model is. Exit polling choses specific precincts based on their ability to predict larger trends. This is determined through statistical models. Telephone polling uses statistical models to predict percentages based on what they poll. There's no real difference. Statistics are used to predict trends based on limited data, which each of them have. I would love for a real expert to weigh in on this, but unless that happens, don't expect more uneducated blathering to change many minds.
    The care of human life and happiness, and not their destruction, is the first and only object of good government.
    - Thomas Jefferson
  • ForlornForlorn Banned Join Date: 2002-11-01 Member: 2634Posts: 6,495
    Well Keyser here is what I think:

    Exit Polling needs a change, and they are the ones who screwed up this election. How do we know people don't lie? How do we know they are succesfully spacing apart the demographics?

    And why do other polls, such telephone polling pick Bush as the winner? This is clearly an error with the exit polls, and not the election process.

    Sure election errors probably happened but I'm sure they always have.


    QUOTE
    telephone polling has limitations not limited to what types of people answer the phone and the small sample sizes; exit polling has limitations too. The strength of a poll lies in how successful its statistical model is.



    Conviencing polling has severe limitations. Convience polling is great for small communities and quick results; for a nationwide effort there is bound to be problems.


    Can anyone find exit polls done over time, and see how accurate they are?
    user posted imageuser posted image#sci-ns Want decent mini-maps?

    Bad NS clan scene?Pm me on IRC!
Sign In or Register to comment.