Do Latest Election Polls Mean Anything?

scott here is the director of survey research at the pew research center we will talk about the post-debate polls will talk about the dates in general scott first thing to talk about is we've seen in the last couple of q research center of presidential polls we saw present all bomb out with a substantial national leader i think they got as high as eight oracle potentially even and you you tell me if i'm wrong to now looking at mit romney plus for is that really believable that we have actually seen a twelve-point sewing between president obama with the president obama leading in that ronnie or is something else at play here it it is quite a swing i don't think and my history of doing of presidential pop also i've ever seen anything quite as sizable is this uh... you know we've we've done our polling using the same method all your along and you know that you get what you get when you do it and then you ask the question what what underlies this i think that in the in the case of the comparison that picture citing p_ mid-september poll that we did it showed uh... obon along with a eight point lead own like we've got butters it we may have caught of honor and be campaign at really the most positive time for him it was in fact shortly before the the um... forty seven percent take came out but it was so that it was reflecting the the strong bump that you've got mad at the convention about that a lot of other all polling organization shown andy it was a particularly good time for him the narrative and in the media was that uh... marami can't do anything right and he said you know he's got it do something to turn his campaign around bike comparison i think that the poll that we conducted over the weekend caught the electorate at the moment about the greatest ballads permit rock bundy uh... since our poll came out on monday showing the romney with a small allege among likely voters we have seen a number of other polls that also show significant movement their those goals may not have shown as much movement from their previous bolted to this one but all the polls are beginning to converge on a it basically broke break even race so we'll talk about that a little bit uh... i just think that some of the emails i got it i would love for you to address these directly are hated having scott kita ron that's great ask empowered they really uh... adjusting for having that plot marami plus for polls so heavily uh... made up of older voters which tend to go more for romney as well as and again correct me if i'm wrong here showing president old bomber as only having fifty percent support from women which seems very very well how do you would just for those things well you know the basic methodology of the poll is that you conduct a survey 'em fung the general public to start with you know we call and we take anyone who's eighteen years of age or older we then wait the davis so that he demographic characteristics of the sample that we get all the general public match the census statistics so we get the appropriate number of older people younger people with men women uh... african-americans latinos and the like and so the sample that we start with is a perfectly good mirror of the population and doesn't look any different from the polling that we get this summer that was so favorable to obama but there is clear uh... things change when you go to registered voters we're seeing is significantly more um... republican friendly sample a few if you will uh... we've seen these kinds of swings in part b_ affiliation we got a lot of uh... phiri hang greek emails and calls in uh... late july when we put out a paltry autobahn up by ten those calls were coming from from people on the right saying that you're cooking the books your pulling extra democrats into your sample you know we're getting we're not getting as much of that from from the left this time but we're we are getting those kinds of skeptical questions but one of the things that stood out to meet in the poll given that we're talking here about likely voters is how much more enthusiastic republicans are in this poll compared with uh... a month ago one o'clock about that i mean when i actually when i ask myself okay worth less than a month from this election ken this many people really not understand that brocco obama and mit romney represent vastly different ideals broadly can be really be just saying you know what i was going to vote for obama but based on a lukewarm debate performance i'm now going to vote for a candidate who is completely different so many ways is it possible what we're seeing is that as people become moral likely to vote as we close in on the election and enthusiasm changes based on the waiting and that the polling agencies do for how likely you are devoted that's really what we're seeing is a shift that i think there's defy definitely think that that is a part of it that if u tell me in guitar uh... quite in our series of questions that you burke uh... getting a lot of thought to the campaign you're following news about the campaign uh... that your absolutely certain that you're gonna vote you give yourself a ten on on a zero to ten scale about turning out the you're more likely to make it into the final sample if democrats were feeling somewhat down about uh... the president's performance by the other the other night weekend they're not as likely to say that they're paying a lot of attention to the campaign right now 'cause it's not all that pleasant for them then fewer of them are going to make it through through the screen and what we see stressed the people is these are factors that can change again between now and election day well let's talk about that we step back from the day today how often does that happen historically how often has it happened that eight different candidates ended up winning the election than the one that was in the average of all polls ahead before the debates started in other words how often do the debates change who ends up winning it's it's very unusual parts very unusual i would say the one example you could point to would be reagan there wasn't as much polling in nineteen eighty s six years now but jimmy carter had generally lead in polls uh... not all of the polls but many of the polls up until the debate which of course was very late leaving very little time for carter to to recover from not really a bad debate performance but in many respects something maybe analogous to what we saw the other night which is that reagan was able to across a list by threshold of acceptability with a lot of people who had very negative or at least gap practical views of him and i think we're on the you know was able to take that we can see that in the internals in our poll we have but significant spike upwards the percentage of people who said that romney relates well to the to ordinary people yeah who've found him to willing to work with people on the other side these are all things that bit he didn't have before the debate that he was able to to pick up as a result of of how he handled himself laughing touch on uh... quickly here what when we when we forget about the national polling for a second look at those yuki swing states that really are going to kind of building each candidate's potential path to victory has anything really changed drastically in terms of of the scenarios that get one candidate to victory or another given and you know came bearing in mind the importance of ohio to republican candidates and all of those things we know now you know i've been following just like you have the this wednesday polling that's come out there was a uh... for quite a prop today in fact uh... but giving us a little bit more insight and i think the bottom line from the others is that they reflect the balances twelve but he is a more modest and i think that's understandable because people in the swing state tonight live in when i lived in virginia at been bombarded uh... by campaigning uh... for months much more so than people in the rest of the country and as a result of that i think uh... attitudes are more firmly set in the swing states tend and they were elsewhere where people may be more casual about the election though and but they're both is probably not gonna make a difference one way or the other interesting alright so we're gonna keep watching those we will see what happens after the vp today and two more presidential debates scott commuters director of survey research at the pew research center thanks as always got thank you david