let's watch. >> okay. here we go. lightning ground. very, very quick, i'm going to read you the names of vary youls people and i want you to give me a word or phrase to describe your feelings about the person. so, if i were to say laura bush, don't say to me, former first lady, i know that. i want to know what your feelings are. it's going to be, what are your feelings towards each of the people. a word of phrase and we're going to go quickly. top of the mind, here we go. we're going to start out with barack obama. laurel? >> weak. >> christine? >> she stole mine. >> okay. >> don't have one. >> tony? barack obama, real quick. word or phrase. come on. this isn't hard. >> somebody i'd hang out with but don't want to be a

president. >> okay, michael? word or phrase? >> athletic. >> athletic. >> ben? >> arrogant. >> debra? >> uncaring. >> uncaring. chuck? >> evil. who else? >> liar. >> what? >> liar, yeah, that's the best one. >> out of touch. >> okay, out of touch. here we go. >> mitt romney. word or phrase. laurel. >> i like him. >> decent character. >> okay? >> undecided. >> okay? >> identifiable. can relate -- can relate to him. >> chuck, word of phrase? >> i like him. i like -- he speaks out. says what he needs to say. >> michael? >> charismatic. >> ben? >> wishy washy. >> debra? >> manufactured. >> manufactured. christine? >> vanilla. >> laurel? >> i have to change it? >> no, who else?

who haven't i asked? >> jonathan? >> innovative businessman. >> chris? >> unimpressed. >> unimpressed. what was your word? >> he's wishy washy. and you were -- >> manufactured. what does manufactured mean? >> he's made. he's a politician, he's taken all of the things that we care about and said, that's what he is. >> and so, frank newport, these foe kucus groups play an import role in this process. >> they do. i've moderated, like peter hart, a lot of focus groups, primarily for business and industry myself, so, i'm well acquainted with them. they have their place, but it is complicated to really discuss them. as you saw there, they are designed to elicit responses in more depth than you would get from just asking survey questions. now, what peopter hart there wa doing top of mind reaction.

that isn't a bad idea. now, notice, that was a republican group. and we all know intuition and quick reactions are important, because a lot of what we do is not guided byrationalbrain. he was saying top of mind to republican voters. even if he had been commissioned by obama, it was good. that at least gave him a starting point, saying, president obama, is the way that some people who are republicans tend to think of you, this is what comes out. now, typically, how that would be used is -- i heard the world "weak" come out. you would take that, to see what degree they played or didn't play. the way we use focus groups and have historically is a starting point. when you don't really understand what a situation is. you get a feel for the kind of ways people look at products or candidates, the dimensions they use. that guides you in further research. now, i have to say that some people are arguing twitter and facebook can do the same thing.

if you really get into it, you can figure out the words people are using, as well. historically the they have problems. that one, big television cameras zooming in everybody's face, you know, so people may have been preening for the cameras and things along these lines. they have their place. they are useful and people do use them. >> i want to come back to one example how a focus group changed the way a campaign was run, back in 1988 with willy horton's ad. let's go to terranova, i'm going to call on you next, a question for frank newport of the gallup organization. terranova? >> hi. my question is, in an era where informal polling is really popular, how is traditional polling going to keep up? today, you can do the same thing on facebook and twitter. what stems are you going to take to keep poming relevant? >> the polls, these polls that you do on any website are not scientific.

>> no, that's the point. people used to say, what about television stations, used to have call in this 900 number and vote yes or not at the end of the news cast, they report what the absoluters are thinking. the facebook polls are the same way. they are not random. they are not representative. so, they really don't give us a hint at all about what we're trying to do, which is represent the broad population. let me just say that again. what we're trying to do in polling is, take a sample and see what everybody would say, if we generalized to it. if you say on facebook or something, vote yes or no, up and down, thumbs up, like, dislike, however you want to do it. that's really not representative of what everybody is going to say out there in the population. so, i this they're not worth a lot to pay enough attention to. but we've had these kind of things over the years, we probably still will. we need to, in the industry, continue to say, that's fine, but what you really need to pay attention to are polls that can be demonstrated to accurately represent what the population would be thinking or feeling if

we were able to interview all of them. >> bob lichter from george mason university, as you look at abc news, nbc news, newspapers, do these polls create the sense of horse race journalism? that's the question, that sometimes skew the coverage toward a race or toward a candidate? >> boy, we've heard that back to the '36 election. lots of people have complained that the focus on who is ahead distorts coverage, keeps coverage of issues from being more dominant. that's a long-standing complaint about horse race polling. some countries actually have banned release of polling in the last week or two before an election and things along those lines because they think it's so injurious to the process. i don't agree. i think people have a natural desire to know who is ahead we should provide them that information. because, if we don't do it, it's

going to be -- assumptions are going to be made. say we ban polling. next fall, we're going to say, who is ahead? and it's better to have scientific information out there. and news organizations cover lots of aspects of the polling. i have a lot of faith in the american public. if they want to ignore a horse race, they should. just go for the parts of the paper or cable news coverage that are on issues. the fact is, humans like to know who is ahead. there's a real desire to understand and know the horse race and therefore, i think it's hard or bad or not advisable to is up pre sue press that, to say, no, you shouldn't know that, we're going to keep you. they want to know who is ahead and we tell them. >> we're going to go to sarah and brad at the washington center. sarah? you are first. >> well, a question based on what you just said, when people are looking at those polls and they are seeing that the candidate that they favor is

either ahead or behind, do you think that is a factor if they choose to vote ? if their candidate is behind, people may choose to go to the polls to try to make them win. >> yeah, i -- those questions are hard to quantify, that's my first answer, you know, questions deserve a hard core imper call answer and there's no evidence to show that. you could argue the other way, people are winning won't go because they say he's going to win anyhow. we don't have an answer to that, but the bottom line is that that information is out there, the candidates have that information and we think, at gas lllup, and others think it's appropriate to provide that information and if they are not going to vote because they know so and so is behind, that's they require. they can make that decision. better to have accurate information to make that decision than inaccurate

assumptions, which would happen otherwise. i'm not going to vote because i heard he has no chance based on inaccurate data. this way, they can make that decision on accurate data. >> let's go to brad next at the washington center. brad? >> oh, hi. i was wondering, i think it's called the bradley effect, where they found in california, in the '80s, i think, african-american candidate polled 5% or something higher than he received the week of the election, said it was because people were ashamed to admit that they wouldn't vote for someone because they were black. do you believe that people will tell a pollster they are voting for someone other than who they are really voting for because it's the socially acceptable thing to do, like someone in new york might think that, might be embarrassed, people think they are a nut job if they vote for santorum, someone who is black may feel like people are going to think they're a sellout if they don't vote for obama.

do you think that people give answers to pollsters that are not true to how they're going to vote once they get behind the curtain and how do you think polling can account for this? >> thank you, brad. >> well, that's a question that's been looked at a lot. one answer to that is, look at the accuracy of the polls. in this election season, the pochling has been quite accurate in primaries. we are now sitting on a situation where romney is showing ahead in wisconsin it would be very surprising if we were back here wednesday morning and santorum pulls an upset. polling has been accurate. polling showed him ahead in louisiana, he won in louisiana. polling has been very good. one answer to that question is, in general, obviously people haven't lying to pollsters because the results tend to be quite accra about what the actual voting is. now, the bradley effect, we also heard that on a female, hillary

clinton effect, that people would say they would vote for a woman. we heard it about santorum, he overperforms, like the gentleman said, maybe in southern states, people are less likely to say they're going to vote for him but actually vote for him. in answer to his question, it's hard to get evidence that supports that across the board. you know, there may be a race or two where we say we saw it, but other races where exactly the opposite happens, that a black candidate overperforms where the polling was. we certainly didn't see that for brk brg barack obama. so, the data are very unconvincing to support this hypothesis that people in general in that instance are more likely to tell a pollster they would vote for a black candidate but when they get in the booth, they don't do it. it is hard demonstrate that at all. >> i was on your website in preparing for this class and in

this era of transparency and access, basically everything you have, how you go about pofl ill it's all available online. what does this mean for the public who wants to know who you are polling and how you go about doing it? >> i think that's great. you are absolutely right. used to be harder, when we didn't have the website to pull the details out. now pollsters should put all the details out about how it was done. the american association for public opinion research has strong standards that say, if you report publicly a poll result, you need to, by our standards and our code, respond with and tell people, all of the details about the methods, exactly how you worded the question, when the interviewing was done, what your sample was done. if they don't do that, report it to us. we've actually censored firms before, officially,essenti cens

firm. we are urging organizations to go even further in making this information available. that puts the burden on you, the user of polls, because you have to go find that information and use it. but at least the information is there, so that the interested person should be able to find out all the details and in particular, the question wording for polls, to make a better assessment of its value. >> and you also blog. >> yes. i talk a lot about that and others at my blog, as well. now days, you need to be transparent. i have a lot of faith in the american public. even if we make a mistake, we goof, we try to put the data out there. in the long run, people figure things out. we are better off saying to the public, here's how we did it. if you want to make a different interpretati interpretation, go ahead. >> christopher conway. go ahead. >> i had a question. because we elect our

presidential candidates by state and not by national popular vote, what is the role and rell van relevance that your poll takes? >> we're asked that a lot. a couple of things is, we poll nationally for a lot of things. we don't want to disentrance chiz people in ut and massachusetts who are definitely going to vote for republican and democratic candidates, it is good for us to see national little where people are standing, even if some of that information is less relevant. second of all, it's an e feshgt way to pick up what's happening even in the swing states. i mentioned earlier, we can't poll, like, we poll every night, 1,000 people nationally. huge commitment on the part of gallup, 350,000 interviews a year plus. we can't do that in all 50 states. it is just unfeasible. we can't track it. so, we think it's efficient to track nationally. those trends we see nationally almost always are translated into the swing states and the

states, as well. so, in the election, where we see movement in the direction of one candidate or the other, nationally, generally speaking, that's absolutely going to happen in the swing state is,s, well. we think the national really gives us some valuable information that's important to have. >> and how do you pay for these polls? they're expensive. how do you make money? >> good question. our primary business is working for business and industry. not doing publicly released polls but we work for business and industry and do client satisfaction, employee engagement. we consult with them on how to increase the engagement and satisfaction of customers and employees. that's what we mainly do and that's where we make our money. what we do in terms of polling, both domestically and internationally, we like to make money out of it and we have an arrangement with "usa today" where they reimburse us for some of the cost of pofling. internationally, we have organizations that will pay us

to have access to the data to be able to analyze it. that's partly the way that we pay for our kind of polling we're talking about here. but the bottom line is, gallup as a whole is an organization that is profitable because we work for business and industry and we help them out. they like what we give them. >> i want to go back to something you said earlier. the 2000 campaign, a black mark for news organizations on election night, jumping the gun, believing the exit polls and then having to retract the story. here is just one example to go back to november of 2000, cbs dan rather. >> florida goes for al gore. now, folks, the equation changes. cbs newest mates when all the votes are in and counted, the sunshine state will have plenty of sunshine for al gore. error is human. that's what's happened with florida. everybody that we know of had put florida into the gore column

some time aim go. now, we're pulling it back into the undecided column because some bald data came from certain precincts in florida. quite honestly, we now don't know what to make of florida. bush wins. florida goes bush. the presidential is bush. that's it. cbs newest mates when all the votes are in and counted, that george bush will have won the presidency of the united states. >> revenge is his and revenge is sweet, as in austin they begin celebrating. one of the closest elections in american history, bush has won it. cbs news has now, for the second time tonight, pulled back florida. we are now putting florida in the undecided category. reason we're doing that is that in actual votes counted, our cbs news -- bush leads by about

2,000 votes. and it being that low, charges of irregularitieirregularities, going to be called for by the push people, apparently, we're pulling back florida. >> brings back memories from november of 2000. a key point to put on the table. these are exit polls. not the kind of pofling that you necessarily do. what's the difference? >> well, exit polling, and let me say a little bit more about what was going on there in a minute. but exit polls are random samples of actual voters as they leave places from the precinct. we do pre-election polling. exit polling, they randomly sect precincts, send people out and every 12th that ximents from a voting place, they are stopped, they say, can you please answer these eight quells, who did you vote for, what are your de demograph demographics? using that, it's all sent back and that gives the people a pretty good feel, generalizing to what would have happen if they interviewed everybody. one of the primary reasons we have exit polls is not so much to do prediction but to give us

a sense of demographics and issue importance. so, you've heard it, this year, for example, high little religious white christians are more likely to vote in primaries for santorum than romney. well, how do we know that? well, patieit's from exit polls. now, what happens on election night is, people start off with, at cbs, nbc, cnn, all the places with the exit polls, they start mixing in the real data. they pay less and less attention to what the exit polls show and more to real data. kind of a mention suixture. in 2000, they didn't wait to get enough data in. they had initial data, called it for bush, gore. and dan rather said there, in the last part of the clip, it's too close. and it wasn't until december that they finally, at least, the supreme court said, all right, enough of this, bush won. so, it was a complex issue. and now, of course, lessons have been learned and you'll watch

polling organizations, media organizations, are much less likely to call things early. >> especially before the polls close, because keep in mind the first call was at 7:50, the7:50. the polls were still open in the panhandle area. >> they were very careful about that and nationally, they won't start calling the election next november until california and hawaii, because carter complained i believe back in 1980 that they were saying reagan won and people were still voting out in california. that 2000 is a fascinating and complex case. basically the networks were too eager to declare a winner for competitive reasons. lot of networks like to be the first and they've learned their lesson. as the exit polls come in, you hear the anchors say too close to call, we're not making the call so there's a lot more waiting to see actual votes come in and start to tally those before they'll make a call than there is to make a call off an exit poll. if an exit poll comes in shows

so and so is ahead by 30 points the analyst will feel confident saying he's going to win the election. if it's close they'll say too close to call. >> isaiah mohammed from dickens, go ahead. >> hi, earlier you had an article that touched on how americans dislike and distrust the government and our health care system. in your career of polling have you seen the inverse of this to be true, and by this i mean do americans fear the intrusion of the private sector for example during the bush years, verizon implicated on spying on american citizens, and that's part one of the question. have you seen that in your career, and secondly, people who feared the government in our health care system, did you ever find out what exactly is it that they fear? i know they talked about death panels and things like that. >> those are excellent questions and there is he aa lot to it, we'd need another four hours to

answer it. we can say in general americans are extremely distrustful of the government. last year we had ten different measures at near record lows for the government. congress's approval now is down in the low teens, at or near the lowest it's ever been, even with obama's approval is at 47%, the congress is 12%. americans are suspicious about big business. americans don't like big anything so when we rate organizations, big business, americans are skeptical about it, not about the specific companies you mention but just generically, americans don't like big business. they love small business and americans are very suspicious of big government as well. so it's both of the above but americans distrust of government is particularly low now, partly fueled by last july and august saying the debt debate when congress was organizing back and forth and couldn't make a decision and we're facing the same thing now. we have these huge cuts in the

budget that will be automatic january 1st of next year, if congress doesn't do something, they're already built into the system and since it's an election year, people doubt congress is going to do anything about it until december so we're going to see the same phenomena where americans say if they're in washington the last minute, clock ticking, trying to make decisions on billions of dollars, that's not something that induces a lot of legitimate support for the public. >> let me pick up on the other point that isaiah was making when it comes to some of the terminology you might be hearing and i'll use his health care example as a case in point. insurance mandate versus being part of the insurance pool, how do you phrase it, will determine what your instant reaction is and how politicians frame the argue mgt. >> that's right. i would say if someone has a reaction to the health care bill or part of it look at how they worded the question. here we have six examples people asked about the individual mandate. does the wording seem to make a difference? by the way it doesn't. americans are more negative

about the individual mandate than any other part of the health care act. some folks try to read a whole paragraph and others more brief americans are suspicious of it and the health care act, either splits even or more negative than positive so americans aren't positive about it. he asked the question, why are americans negative about it and that's a complex question. i think in part it's because it's all bundled together and taps into the suspicion of big government. lot of the provisions of the patient protection and affordable health care act or affordable health care act are approved of by the majority of independents, ride on their parents' insurance longer, making sure those with individual assistance. lot get majority support but i think what the obama administration may not have realized deep distrust of government is there so when you bundle it all together, we'll put the nice provisions together

and make it a big government thing that tapped into the underlying feeling of big government. obama supporters say you have to bundle it. we've heard in the last week or two in the supreme court you can't do the nice things without the individual mandate and so on and so forth so it is a complex issue that i believe the less than positive reaction to the affordable care act is based in part on americans kind of what i would call instinctive negativity about big government and when somebody comes along and says government's going to control all of this and do all these things, americans say wait a minute, i'm not sure i want big government in charge of all this stuff. >> a key component of what this class is all about is learning from history. let's take a step back at one campaign, 1988. i want to share with you and the students just three different points in that election, is you remember in the summer of 1988, michael dukakis was ahead in the polls going into the convention in new orleans, george herbert walker bush as vice president had a double-digit loss in the

polls, 12% to 15%. he gave what many regard as one of the best speeches at a national convention in new orleans. let's watch. >> and there is another tradition, and that's the idea of community. beautiful word with a big meaning. the liberal democrats have an odd view of it. they see community as a limited cluster of interest groups locked in odd conformity. the country waits passive while washington sets the rules, but that's not what community means, not to me. for we're a nation of community, of thousands and tens of thousands of ethnic, religious, social, business, labor union, neighborhood, regional and other organizations, all of them varied, voluntary and unique. this is america, the knights of

columbus, the grange, hadasa, the disabled american veterans, the order of the hepa, the business and professional women of america, the union hall, the bible study group, lulac, holy name, a brilliant diversity spread like stars, like a thousand points of light in a broad and peaceful sky. does government have a place? yes. government is part of the nation of communities, not the whole, just a part, and i don't hate government. government that remembers that the people are its master is a good and needed thing. >> so that speech referred to as the thousand points of light basically framed the debate framed the argument for the bush campaign, keeping in mind it's very difficult for any party to win a third term, we've seen historically the challenges that faced even franklin roosevelt and the problems al gore faced in 2000 trying to get the third term. behind the scenes efforts by the bush campaign to figure out

where they could go after the michael dukakis support and they held a focus group and they came up with the willy horton commercial, an issue we should point out first brought up in the primary by then senator al gore, turned into an issue the bush campaign used in the summer and fall, this one of the most effective ads in presidential politics. >> bush and dukakis on crime. bush supports the death penalty for first-degree murderers. dukakis not only opposes the death penalty, he allowed first-degree murderers to have weekend passes from prison, one was willy horton, murdered a boy in a robbery, stabbing him 19 times. despite a life sentence he received ten weekend passes. weekend prison passes, dukakis on crime. >> so that was september of 1988, a month later, this debate, the question that kicked off the debate with cnn's bernard shaw, let's watch.

>> governor, if kitty dukakis were raped and murdered, would you favor an irrevocable death penalty for the killer? >> no, i don't bernard and i think you know i've opposed the death penalty during all of my life. i don't see any evidence it's a deterrent, and i think there are better and more effective ways to deal with violent crime. we've done so in my own state and one of the reasons why we have had the biggest drop in crime of any industrial state in america, why we have the lowest murder rate of any industrial state in america, but we have work to do in this nation. we have work to do to fight a real war, not a phony war against drugs and that's something i want to lead, something we haven't had over the course of the past many years, even though the vice president has been at least allegedly in charge of that war. we have much to do to step up that war, to double the number of drug enforcement agents, t