One poll shows gay marriage in Maine passing by 53 to 42,
http://www.wmtw.com/politics/21433944/de tail.html

Another poll shows gay marriage in Maine passing 48 to 47.
http://www.dailykos.com/statepoll/2009/1 0/28/ME/412

---

It really depends on who is doing the poll.
I have known personally poll takers just filling in forms on their own with their perspectives.
But all in all I think its gonna pass in Maine this time around.
If not they will put it to vote again.
Im tired of this coming back over and over.

browns_86 | Oct 30, 2009

---

All polls depend on who is being polled and who or what is taking the poll.

Polls are not reliable sources.
Innocent | Oct 30, 2009

---

COuld be the way questions are asked, but they are still close but
maybe out the the so-called margin of error
So. Hamton Institute of Tech LLC | Oct 30, 2009

---

Could very easily be a different sample size producing different results.
ty | Oct 30, 2009

---

You’d have look at their margin of error. For example, DailyKos’ poll have a 4% margin, meaning that with 95% certainty they can predict that 44 to 52 percent of the population suports gay marriage. The polls generally report average numbers, but there’s always room for fluctuation.
Aleksey | Oct 30, 2009

---

The ballot is also misleading when asking a question with a yes or no answer placing yes as the negative to the question MISLEADS.

But I guess that is the point——–F’ ing tricksters political entities disgust me. Left or Right no one seems to bring it straight.
jusme | Oct 30, 2009

---

depends on how the question was phrased and how the responses are weighted when averaged out and applied to the general population.

Sampling size and population is important but the real trick is making sure there aren’t any preconceived biases that affect the outcome of the poll. There are usually some accompanying demographic questions or political questions that the pollster has determined are more important than others so the answerer gets lumped in with that group, thus boosting the average. For example, a 30y/o female republican who is pro-choice and makes $35k/year may say she is against gay marriage so the poll may give more or less weight to her age, sex, income and/or political persuasion than is true in actuality and conclude that most 30 y/o’s making $35k feel that way. Or they may conclude that pro-choice republicans feel that way, etc.
CONvict Shrubya | Oct 30, 2009

---

All I care is that it passes. But we won’t really know until next week. I tend to dismiss polls….
delina_m | Oct 30, 2009

---

A number or reasons, sampling method, some polls, split undecided equally or in proportion to the total votes. How a question is formulated can make a big difference/ Also methodology, telephone, vs home visits. Also ethnicity of poll taker and respondent.It’s not that easy nor that definitive. Lastly the number of refusals.sample size and the closeness to the election when the poll is taken. Many undecided switch to one side and some other committed voters will switch to the opposite side. The Literary Review poll is a classical example of Dewey ahead in the polls and Truman wins Predicted siders may switch to the opposite side or to undecided Another factor is the voter turn out and the ability of each side to bring their voters to the polls
Rosen | Oct 31, 2009