I know it seems like the polls are all over the place swinging wildly this way and that (which they are). This has some people thinking everything from conspiracy theories to outdated models.
What I'm about to tell you will not necessarily clear things up altogether, but I think it could offer a logical reason why the polls are all over the map and further reason why I think the RAND poll offers more poll stability and seems to track what everyone feels more closely since it tracks the same people over time.
There's also a good lesson in here on why you should knock on every door on your list even if the yard looks like a Romney sign store.
Follow me over the fold/orange thingy/insert-your-own-word-here for the story.
So I went canvassing with some folks in OH (Cincinnati) for Obama yesterday. In addition to the satisfaction of supporting the candidate I think is best, I also was genuinely curious to put my finger on the pulse of people in such a large swing state.
First a few things about where we were canvassing. This was in a very suburban area just inside the southeast corner of the I-275 corridor. The neighborhood was what I would clearly consider upper-middle class. One set of doors we knocked on was almost all glass and clearly a Frank Lloyd Wright inspiration. Another had a very nice composite wood front deck that was not inexpensive to have built.
Every single house we knocked on was occupied by caucasians. We only saw one person in the neighborhood the entire time that didn't fit that mold -- a seemingly mixed race young man on an afternoon jog. I don't know if his house was one we knocked on.
Lastly -- there were Romney signs everywhere. I would rate the ratio of Romney to Obama signs at easily 10 to 1 (see note at bottom). Also important of note was that there were many signs that were anti-Obama with slogans such as "Fire Obama" and "It's Not Working" with the "O" in "Not" looking like the Obama campaign "O".
The people were all at least curteous -- even Romney supporters. Most were very nice. The age range was all over the map. Some folks had moved, many were not home. Some were home, but didn't want to come to the door. You know this when you walk back by some of the homes that didn't answer the door and the material you securely stuck in their front door is gone.
Of the homes that DID answer the door, I would say most of the people on our list were Obama supporters or able to be persuaded to be which probably explains why they were on our list.
But there's one thing about the folks we were reaching out to that I think can help explain the polls. I'm not sure how the list we were working from was constructed. At least one woman insisted she was a Republican and always has been. Not every person in each house was on our list.
There are two homes in particular that stand out as a testament to this. One home had no signs in front. The names on the list were clearly two siblings in the 18 to 22 range. We had already been to several homes where the young adults in this age range were away at college. In this instance, both siblings were home for the weekend and the brother answered the door.
The young man said he and his sister were firmly for Obama. Their parents weren't. He even said his mother had ordered "2016" and wanted him to watch it in an effort to sway his vote. I informed him how terribly inaccurate and misleading the movie was. He said he already knew and not to worry about either him or his sister.
A second house had no less than five Romney/anti-Obama signs in the yard. The man on our list was in his early 30's so it would have been easy to presume this home was his and he was not open to discussion. We even had a brief discussion on whether we should move on to give more time to others that were clearly undecided. Thankfully, we decided to knock on the door.
The man on our list was living with his parents. The signs were theirs and he was clearly undecided. He listened, was glad to take some literature and seemed appreciative to have someone talk to him about the issues. The conversation was enough to mark him as "Lean Obama" which means that isn't the last time his door will get knocked on in the next two weeks I'm sure.
So here's where this story might help explain the polls. In a house divided, depending on what day of the week it is, what time of day it is and who decides to answer the phone could greatly affect the outcome of a poll.
I'm not a statistician and maybe they've already accounted for this. I'm sure someone that is a poll detail junkie can shed light on this. At a minimum, if you're canvassing, you should take this as a lesson that even if it looks like a house is a lost cause, do not move on. Knock.
*Note about signs: Most homes didn't have signs of any kind. I found out from one person at the field office that lives in the area that there are at least some folks that are Obama supporters that would prefer not to put signs out. She seemed to suggest that the anti-Obama sentiment is sufficiently strong enough in some folks that they'd rather not deal with it.