carolina and nevada. be sure to join "washington journal" live tuesday morning. join the discussion. ♪ [applause] >> every election cycle, we are reminded how important it is for citizens to be informed. >> c-span is a home for political junkies and a way to track the government as it happens. >> i think it's a great way for us to stay informed. >> there are a lot of c-span fans on the help you like colleagues are going to say i saw you on c-span. >> their summits more that c-span does to make sure that people outside the beltway know what's going on inside it. on " history bookshelf," former gallup senior editor david moore talks about his book, "the opinion makers." he discusses the flaws in

american polls and the way that poll results are reported by the news media. waterss recorded at street bookstore in exeter, new hampshire in 2008 and is about one hour and 10 minutes. dr. moore: i want to turn to essentially the theme of the book. the subtitle says, "an insider reveals the truth about the polls." what is the truth behind the polls? a major objection to the way that polls operate is that they don't tell the truth about the electric and they don't tell the truth about the public. that doesn't mean they don't tell the truth all the time. i think we find a lot of very interesting information from polls. particularly when the polls address what people know about their own experiences and what people genuinely know. for example, we know, based upon early gallup polls that a majority of people in the 1930's

approved of topless bathing suits for men. we know that a majority of women and a majority of men opposed having women go to work in business and industry if their husbands could afford to support them. those kinds of questions, we could not even ask today. whether or not women should be allowed to work? that's if their husbands could afford to support them. it shows you what a tremendous change there has been in american history. polls can provide that kind of information, particularly about the culture. where i object to the polls is when they deliberately mislead us, with respect to elections, reelection polling, and respect to public policy. when it comes to public policy polling, the polls do not tell us the truth.

the main purpose is not to find out what people are thinking, the main purpose of the polls is to provide information for the news media which is plausible and which can be dovetailed in with regular news stories. let me just give you an example. we all know that prior to the war in iraq, the polls were showing about 2-1 support among the public for going to iraq. this was in february and march, leading up to the invasion. the question asked whether or not people would support american troops going to the gulf region in order to remove saddam hussein from power. that phraseology came directly from the administration. "to remove saddam hussein from power." of course, the war was more than that. it was not just to remove him from power because obviously we

did that and we are still there. it was to do a lot of other things, which were never part of the administration tagline or what the media reported or what we asked the public. even then despite what we now , know was biased coverage of all the information about weapons of mass destruction and an imminent threat of saddam hussein, and we know that the news media, particularly the new "the new york times" and "washington post" admitted they went overboard in supporting the administration viewpoint. despite that fact it does not , mean the american public was persuaded they had to go to war. the reason i say that is while i was working at gallup at the time, we decided to run an experiment to get below superficial opinion. we know a lot of times when you ask people about their opinions, they give an opinion because

they are pressured to give it by the interviewers. they may not know what they're talking about. they may not know about the issue, but they are given what is called a "forced choice" question. do you favor or oppose it? it's never do you have an opinion or have you thought about it. by pressuring people to give their opinion, they will come up with an "off the top of their head" response. we wanted to go and find out whether people were really committed to the view that they just expressed. we asked the question, for people who said they supported the war, whether or not they would be upset if we did not go to war. we asked people opposed to the war whether they would be upset if we did go to war. now if you express an opinion and say, "but i don't care if my political leaders do the opposite of what i just said" and you admit that within 30 seconds of having expressed an opinion, chances are you are not

very firmly attached to that idea. that is what we wanted to measure. what was the net effect? the net result shows 29% of the public favored the war and would be upset if we didn't go. 30% opposed it. and would be upset if we didn't go. 41% did not have an opinion or didn't care. when i say didn't care, i don't mean it in a cavalier way. they did not know whether we should go to war or shouldn't go to war. they were truly divided. some did not care, probably, and some just simply did not know. it didn't matter. the main point is that a plurality of americans, going into the iraq war, one of the most heavily covered current events at the time, a plurality of americans were so unengaged that they did not have an opinion as to whether or not we should or shouldn't be going to war.

what was the picture presented by the news media? the picture presented by the news media -- 2-1 support or greater. later on, we find many commentators saying the american as ae are as much to blame government for going into war because the american people were clamoring for war. we hear that kind of justification or that kind of condemnation of the american public. the fact is, had we measured public opinion, without trying to manipulate people into giving a response, but if we had measured it accurately, we would have seen a divided public. would that have made a difference to president bush? probably not. [laughter] dr. moore: would that have made a difference to the democrats who we might have expected to oppose the war? there were 10 democrats in the senate, all named as possible

presidential candidates or eventually did run, or like senator reid who became a majority leader, who supported the war. there is always the suspicion that they supported it for political reasons because they what the publicr wanted. the important point is that this wasn't just one issue. on the issue of opera grade and on the issue of closing guantanamo bay we are soliciting , the public for an opinion off the top of their head so that the news media can say the majority is for this or the majority oppose that. they want what is the net a majority. effect? let me give you an example with respect to guantanamo bay. a lot of people did not even know guantanamo bay existed, much less that we had a prison.

a lot of people did, some did but some did and some didn't. when it gallup asked the question, our first effort with this was to give people information so that we could then question them on their opinion. let me tell you what is wrong with that to start off with. a sample of people that we choose, based upon very precise scientific statistical methods, is supposed to represent the larger population so that we can generalize from what these people say to what these people believe. but if these people are given information that these people do not have the sample no longer , represents the general public, and that is what we do on a routine basis. what you have then is not a real result. you have a hypothetical result. a hypothetical result is only correct if everyone in the general population is informed in exactly the same way.

so let us go back to the specific example. how did we inform this sample of people, who we knew, a large number of them, had no idea what guantanamo bay was, or that the prison was there, or anything of that nature? so we said as you may know, the united states has a presence at in cuba where bay it houses suspected terrorists. do you think the united states should close down guantanamo bay or not? if you have never heard of the guantanamo bay prison, why would you say yes, close it down? it houses terrorists. you don't know anything about tortureroversy over the

the possible violation of geneva , conventions. there are problems with guantanamo bay. you say "yes." guess what? we reported the american public supports the bush administration keeping one time obey open -- guantanamo bay open. i was a part of the poll that came up with those results. we had no intention of being part of what i now call, "the legitimacy spin cycle." the administration comes up with a point of view, and they push that point of view. as you know, the bush administration is very effective with that . they push that point of view, and the news media in the united states tend to be fairly unimaginative when it comes to looking for new sources. they rely disproportionately on what the government tells them. they will go to other government sources or two other people in

power, such as democrats, if the democrats oppose what is happening. they don't do very much research. they don't go to the u.n. or to the red cross or other organizations that might have information about what is happening at guantanamo bay. so, instead, the media tends to reflect, pretty much, the dominant theme of the administration in power. then, the media, who owns their own polls, cbs, abc, washington post, the fox poll, the cnn poll, the usa today poll, they all own these polls. i know from working on it that we have phrased the question pretty much the way that the media have been phrasing the question. the media has been phrasing the much the way

the administration phrases the issue. when we asked the public what they think, it tends to support the administration's point of view. this happened right after abu ghraib. the administration was very concerned about the notion that we were torturing prisoners. "the united states does not torture." that is the theme that went through. there were some polls, several polls that touched on that issue. there was one poll, the abc-washington post poll, that really looked into the issue with greater depth. all of the polls, with the exception of one that asked only one question one time, all of the polls that asked about it always talked about the abuse of prisoners. the talked about torture in the abuse of prisoners. -- they talked about torture and the abuse of prisoners.

abc was even more creative. they talked about the "apparent" abuse of prisoners and asked several questions of the sample about the "apparent" abuse of prisoners. twice, three times, and the fourth time they said "abuse" of prisoners. a couple questions later they , said, "do you think of the treatment of prisoners in abu ghraib was torture or abuse?" what do you expect them to say? they have been told four times that it's abuse or apparent abuse. how many are going to say, oh, that must be torture? the polls announced the good news to the administration -- the public did not consider what was happening at abu ghraib to be torture. they bought onto the notion that it was "abuse" and not torture. not only that they bought onto , the notion that it was an

aberration. that most of it was caused by low-level soldiers rather than any policy within the administration. the way polls work is essentially to complete, in many cases, to complete a spin cycle for the administration. i will also give you an example that happened with the clinton administration and the attack on waco. that happened so many years ago, you probably don't necessarily remember, it was about 15 years ago. i think it was 1993 when the justice department attacked wake of because of the cold that was there. the question was, "did they tooadministration attacked quickly?" of course, the public did not know. they were essentially pressured into giving an answer. everybody said, "they did what

was right, they waited the proper amount of time." sure enough, that kind of reaffirmed the clinton administration had done right. the point i want to make with respect to the way that polls operate is that instead of forcing people to come up with decisions and opinions when they do not have them, the polls ought to at least allow people to admit that they don't have an opinion. "do you think they did this or did that or don't you know?" that is not what the polls are interested in doing. if they reported on a regular basis that somewhere around 60% of the public was unengaged in the issue, and of the rest, 16% favored it, 21% opposed it, there were still even more unsure, and there are these

people who haven't even heard anything about it, that would not be very interesting. that is exactly the kind of result that george gallup produced in 1953, when asking about taft-hartley. the taft-hartley act was an act that had received a great deal of publicity. but, when george gallup preceded the policy question with a question about whether they had even heard of taft-hartley, 60% had no idea what it was about, which by itself, was amazing. given the publicity. that those who had heard about it an additional 7% said they , had no opinion as to whether or not it should be extended or curtailed. two thirds of the public, not knowing anything about it, you will never see a poll result in a newspaper or on tv these days that says 67% of the public

doesn't know. if they did, you would say, why are you polling them? what is the point ?if people are not engaged, why are we looking for opinion? that goes back to what george gallup wanted to do in 1935. he wanted a poll to monitor the pulse of democracy. his polling came after a period, the progressive period. that was from about 1880's-1920 one primaries began to be adopted and women were given the right to vote. it is when they provided for the direct election of senators. there is a real change in the orientation of the constitution, allowing people more input. that is when primary elections were designed, in order to take

away power from party bosses so that we had leaders or candidates were put up by the parties, at least the rank-and-file were able to have some kind of say. from that a lot of people felt , there should be direct democracy. james bryce an intellectual of , the period, said there ought to be some method for monitoring on a week to week basis what people are thinking. wrote 1925man who said that's ridiculous. he said the public was not too interested, too uninformed. "we cannot rely on the public to make policy." instead what we need is a ruling class. that wasn't necessarily a good solution. came the 1930's and can george gallup. polling is an

outgrowth of the idea that if there is not direct democracy, the public ought to be part of the discussion on any given issue. fi.l very idealistic i don't oppose the idea of having public opinion polls reveal how many people are in favor, oppose, but also, how many people are truly unengaged. i think that is important part of this, a realistic description of the electorate. the electorate is neither completely useless, as litman would've argued, nor is it an entity that can guide public policy. it is useful to know how many people are in favor or opposed and how many people are undecided and what are the issues could it could be very educational.

that is not the way the polls operate right now. not only do they not operate that way with respect to public policy, they don't operate that way with respect to elections. one of the biggest embarrassments the news media should have right now is giuliani versus clinton as the two candidates they were continuing to run in hypothetical matchups in 2007. juligiuliani was the dominant front runner of the republican party and hillary clinton, who according to gallup polls, had a solid lock on the democratic vote. there are two reasons, major, fundamental reasons why the polls were so wrong. pollsters will say, well, the voters changed their minds. the answer is no, they had never made up their minds in the first place.

not only that, but you are not even polling the right voters. was talk about polling the right voters. who are we talking about when we say giuliani was the front runner and that hillary clinton had this solid lock? they were talking about republicans nationally. they were not talking about republicans who were in iowa new , hampshire, south carolina. they were talking about polls they did of republicans nationwide. we know there is no such thing as a national primary electorate. it doesn't exist. we don't have a national primary. we have a series of contests. people who vote at the end of it are not paying attention and don't have to. by the time it comes down to march or april, a lot of things are going to have changed. why should they pay attention to

who is there? if you force them to, with a name, they will say, ok, i have heard of giuliani. "if the election were held today, who would you vote for?" that's the way the question goes. is they forced choice hypothetical situation. it is "who would you vote for among the following candidates?" well, republicans nationally are not the right voters. the interesting thing about giuliani is that while he was leading among reptil republicans nationally, he was trailing in iowa, in new hampshire, in south carolina and michigan. he was barely competitive in florida. guess what happened? by the time you get to florida, all of these other contests showing him to be a loser, he ended up with not a single delegate in the republican national convention.

think of all the money and all the effort and time people spend in support of the candidacy that was mythological from the beginning. he never had a lead anywhere in the early contest and was a leader only in this electorate. the same thing was happening with hillary clinton. although hillary clinton was doing well in new hampshire and some of the other contests, she did not have a solid lead, but she was leading by double digits, 20 points among national democrats. democrats would not really paid attention to what was going on. in iowa, the very first contest, and everybody knows iowa can reshape everything.

we certainly know what happened four years ago when howard dean was the overwhelming front-runner among national democrats until he lost in iowa and suddenly john kerry came over. there wasn't any shock at the fact that hillary clinton wasn't leading in iowa. detrimente a serious to her bid for the nomination. yet, everybody, and by everybody i mean all of the major media organizations, retreated to the expected race between hillary clinton and giuliani. that is the first thing. we were not even measuring the right voters. let me just say another thing. many in the national media should really be ashamed. there were several media polls who did do polling in new hampshire and remained until the end. that is, until election. they got burned.

gallup said they were 2% undecided and hillary and barack obama had a nine point lead. how did that work out? after the election, gallup found out, sure enough, there were somewhere around 25% of the voters who made up their minds in just the last couple of days. you see a lot of undecideds could that is what the exit polls showed, too. by the way, the unh survey center asked upfront whether people had made up their minds and reported, accurately, that about 21% of democratic voters, going into the election, were completely undecided. another 27% were leaning. it means that if there were less media eventsute that favored one candidate over the other, which there was, that was the case. in any case, the media got burned. what happened in south carolina,

in the subsequent polls? the major media polls would go into those states a few weeks before the election and made sure they didn't poll before the election so that their results could be checked against the final results and see whether or not they were accurate. so the only people that were polling were not the major media polls in all the subsequent primaries. major media polls had it easy. they were doing national polls. they were talking to national republicans, national democrats, talking about the race. the real fight, state-by-state, was being measured by a lot of other organizations, who is centrally come out of the woodwork during election time to get media attention. it doesn't mean they aren't always very good, but some of them you don't hear about until , election time. that is the first thing. what is the second thing we know? the second thing we know is even

among national democrats, national republicans, they were not even measuring correctly. why do i say that? in october, 2007, gallup abandoned its normal process of asking, "if the election were held today, would you support these candidates?" in a separate poll, they asked upfront, if you made up your mind, who would you support for president. instead, they asked have you made up your mind in the democratic primary or republican primary. they found that 75% of republicans have not made up their mind and 65% among the democrats. of those who did make up their mind, who did you support? no republican candidate got more than 5% of the vote. giuliani got 5%. ron paul got 4%.

i think mitt romney got 5%. no one got more than 5%. what was the true state of the national republicans when they were saying giuliani was the front runner? the true state of opinion that the time was, "i don't know." that was the truth. and yet by having giuliani as the front-runner, it really cut off a lot of funds for the lower tier candidates, cut off a lot of people who would've participated. it certainly did that with the democratic side. on the democratic side, it did show hillary clinton was a hit. she had 17%. barack obama had 9%. but that is not a 22 point lead. not only that, but 69% are still undecided. no one could say she has a lock on the nomination. if the polls had told us the truth even for the national

primary electorate, it would've shown us the race was truly wide open. there's a certain self-fulfilling prophecy as far as money goes, being able to raise money, being able to attract the candidate coul and somewhat can we say about the polls during the election? what we can say is they did a disservice to the democratic process. they did not tell us the truth. they manipulated people to come up with these names so that they can report them as though they meant something. they affected the democratic process and they did it all because they did not want to admit that a large number of people were undecided. if they did that, people would say, "why are you pulling them-- polling them?" that is, by the way when george , gallup first did this in 1935, he knew that when he asked the question, "who will you vote for in the november election 18

months away," if you asked about it that way, so few people would have known that they would not have picked up the question. i just ask them who would you vote for an election if it were held today. what turned out to be a minor way for george gallup to push poland, which at that time was still in its infancy, has now become the dominant paradigm misleading approach for measuring the way electorate thinks about candidates. i think that is really unfortunate. one last thing with respect to the current polls. the current polls suggest that 95%-98% of the voters have made up their mind already about who they will support in november.

of course, we know that's not true. there's a significant number of people who are undecided. bouncecalled convention were really people who are undecided who are being forced , to come up with an opinion. during the democratic convention at the time when barack obama was getting a lot of attention, undecided voters said, "barack obama." during the time republicans got a lot of media attention, the undecided voters say, "john mccain." we have got this apparent crazy electorate, totally decided, but constantly changing its mind? that doesn't make sense. [laughter] dr. moore: what is the truth of it? the truth of the matter is thatd at by a cbs poll showed 35% of voters could be called "swing voters." they did it kind of backwards. they ask people who they will

vote for and then they say "is , that a firm decision or will you change your mind?" they got about 35% in august, who would probably be swing voters. their conclusions were that most of the changes after the , conventions, were undecided voters who began to make up their minds. what really bothered me was that cbs didn't go and say, "how did many people are still undecided and what is the status of the electorate?" the campaign manager for barack obama said that he found the gallup polls were completely worthless, because so many voters were undecided. the truth this is that if you want to understand the electric right now, you have to recognize that a certain number of people are thinking about the campaign, they are going to be influenced by it. if you don't have any sense of how large that this and think it

is always this amount, "wow, the numbers are going back and forth," you don't get a good idea of what voters actually want. the media wants the race to keep going up and down, because it gets commentators something to talk about. i think they still could talk meaningfully about a race if it was accurately measured by the polls, which show a large number of people undecided and that number declining over time, trying to figure out where it is that's persuading them, rather the methoduing with of asking poll questions that doesn't tell the truth about the electorate and is completely

misleading about the state of the election campaign. so that is my complaint with the , news media. in the book, i do suggest and its implicit in the way i said here that there are ways to make the polls useful, for the polls to tell the truth about the public. i still think polls can be very helpful. i'm really mostly concerned about public policy polls and preference polls and pre-election polls. when they talk about personal experiences, they can provide good information. when they refuse to tell the truth about the electorate, or about the size of the unengaged public, i do think they do a disservice to democracy. let me open it up to questions. >> if i follow you correctly, from your first anecdote about the poll where 30% of the voters

originally said they were against going to war and 30% also said they would be upset if we went to war, is that correct, ? that implies that all the people that didn't care ended up saying, "yes?" dr. moore: that's true. most of the people who said they favor the war in the first question, the superficial one, the people that said they were not upset if the opposite happened disproportionally were people who supported the war. there was only a small group of people who opposed it who would not have been upset if we did. >> what can you conclude from that? dr. moore: i think that there are some issues where the opposition is very strong. that when they express opposition, they have to know more sometimes in order to

express opposition could there's is a tendency for people to say "yes" to policy. ,it is called response acquiescence. if you don't know anything about it, rather than being negative and say i oppose something you , tend to go along with it. if you are pressured into a response, you're more likely to italong with it then against . it could be anything. that's just kind of the nature of it. >> did they take that into account when making decisions? dr. moore: our political leaders? i think the interpretation of political leaders of polls is quite varied, as you can imagine, given the great number of politicians. there seems to be a lot of evidence that some politicians use poll results primarily in a manipulative kind way. it is also true that a lot of hi politicians don't know

very much about what the public is thinking and pay attention to media polls and what they say. with respect to the administration's point of view on abu ghraib, on gitmo, on the iraq war, the politicians felt the public was "overwhelmingly in support." they really couldn't do a lot against it cou. it's a combination, but i think the polls create an environment for the politicians, and certainly for the news media. the news media believe their own polls and their own coverage is shaped by what they think the majority is as well. ok. >> how do you get your supposedly representative samples? once you get them, how do you get past caller id? dr. moore: could you ask that question into the microphone? >> how do you find your

supposedly representative samples? dr. moore: how do we find our samples? good question. a lot of people want to know why they have never been called. with respect to telephone polling, we use computer-generated, randomly generated telephone numbers. every residential telephone number, in theory, has an equal probability to be selected. we know that a lot of telephone numbers have already been assigned to businesses, have already been assigned to nonresidential enterprises. there are companies, now, that make it their business to help design samples. you can go to those companies and obtain a representative sample of residents with telephone numbers. i'm talking about land lines right now across the country. it is not that much different when it comes to cell phones. the reason most pollsters don't call cell phones is because it

costs the interviewee as much as it costs the interviewer. with respect to cell phones, there is a growing number of people who only have cell phones and they are systematically excluded from the vast majority of media polling. to come back to your question, i can't go into great depth with it except to say that we use computer-generated random numbers to ensure that there is an equal representation in all regions of the country, that all telephone numbers have an equal chance of being selected. the sample size is typically around 1000, nationwide. sometimes more, sometimes less. >> there is something known as effect.der affecte the first black governor of

virginia who had advised obama about this said that in the end, the polls cannot detect racial attitudes, that they are often hidden and are volatile, and that at the last moment, decisive. this is called the "wilder effect." what is your response to that? polling techniques, determining the racial component in the election, is there a wilder effect? dr. moore: the question has to ,o with the wilder effect referring to governor doug wilder of virginia, when he ran for governor, polls indicated he would win heavy handedly, but he barely won. that suggested that a lot of white voters, particularly democrats, who might have been expected to support the democratic nominee, told the polls he would support him, but

didn't because of his race. that is called the "wilder effect." there was something like that for mayor dinkins who ran in 1989 and 1993 for reelection. i forgot the dates exactly. he barely won even though the polls suggested he would win handily. when he ran for reelection, i was with gallup at the time. -- thatd that he was the percentage vote that he got was close to the polls. the question is whether or not we will find with obama the same kind of thing and that there are white voters who say they will support him, but then actually won't. is very difficult to predict right now what will happen. we know that when he won in the pollsthere was

someone saying are accurate, saying thes someone polls are accurate, suggesting that there was no effect because of his race. what is happening right now are not, it is hard to say. the feeling is that that effect does not really pertain anymore. we are talking about examples from various states. you talk about the nation as a whole, whether there is a racial effect in some states, it is very difficult to say. until afterlly know the election and we do a very close analysis of all the polls. i guess i'm inclined to think there won't be much of an effect. black candidates have become fairly common, by comparison with 15 or 20 years ago. that is just speculation. a lot of other statisticians could very well disagree with me. >> are you suggesting that if

people were given the chance to say i don't know or i am undecided, and this that questions were framed more fairly, more people would choose that? from my understanding of psychological research a lot of , people don't like to admit ignorance and probably would choose, even if not coerced, just for fear of looking dumb? is that a valid idea? dr. moore: that's valid research. it is very difficult to measure an opinion when you talk about socially desirable responses did is it socially desirable to have an opinion when you are in a context where you are being interviewed by an organization and they want you to have an opinion? yes, it is. can we get at the notion that some people may not have an opinion? would they be willing to admit it? suggestsical evidence and a lot of people are willing to admit.

it usually gets a larger "i don't know." so you do get a significant number of. i can't say we can't measure precisely, because people themselves don't always fully know what their opinion is. they are kind of confused about a lot of things, a lot of ideas come to mind. the major point i will make is that we should not systematically try to ignore the undecided. we should allow the possibility for people to admit that they don't know. if we have long polls where there are a bunch of questions, favor, oppose, unsure, then that makes unsure ok.

i think if you have a whole series of questions, even if you say, "unsure is illegitimate option," some people will so resist saying "i don't know" to all of those questions. i do not think we can overcome it completely, but i think we should not systematically oppress the undecided, giving the impression that the electorate is fully informed and fully engaged. >> a follow-up to that. that question. if you say somebody is undecided, it sounds like they are just forming an opinion and don't know. the media wants to put left against right, yes against no. why can't you telik and selling -- can't you tell a compelling story about 35% of the electorate being undecided? if so much polling has been

hijacked by people manipulating it, isn't it your responsibility to come up with a good housekeeping seal of approval? [laughter] >> to get the word out there about which pulls you can stand behind, which ones are doing it correctly. how do you take back your own field and get people to understand which ones they can believe in? to simply say, "you can't believe anything," i think people are tired of it. dr. moore: to the first question, it had to do with whether or not we can make the

news media -- make it a salable to the news media that there's a significant portion of the public that is not involve. -- involved. there is a paradigm shift right now. for so many years, the framework has emphasized that the public is rational, fully informed, fully engaged. that has been persistent since the time of gallup. he is to ignore people that said they didn't know and take them out of the equation, until he was approached by lindsay rogers, who said instead of measuring public will, he is measuring "baby talk." he came up with a new plan. it started off by asking whether people had even heard of the issue, and then would ask them

what they thought about it, and a few other things. the problem is that the plan never went anywhere with the news media. he could never sell it to newspapers. that was his major outlet at the time. they did not want that depth, they want superficial response. i propose a simple follow-up question. "do you favor a person or are you unsure?" and then we asked, "would you be upset if the opposite happened?" it should not be too much for the news media, as far as time and resources go, if they want to get at least a little bit below surface opinion. can we make that salable?

some say my prescription sounds like a pipe dream. i know from my colleagues in political science, not necessarily the polling industry, that they also have a feeling that it could be done, if somehow we could change the mindset, make it so that people recognize that there is a certain public that is unengaged. let's not manipulate them into coming up with responses that suggested that we have overwhelming majorities that favor war or this or that. i don't know whether you will be

able to see whether my argument has any staying power. some want to modify the questions according to what i want to propose, and others will try to do it and come we will see whether or not it works. the second question, whether or not there is a "good housekeeping" seal of approval. so far, i haven't established such a group. [laughter] dr. moore: i doubt there will be one. let me speak to the difficulty of the polling industry itself. it is very concerned about some polls. a lot of people think that internet polls that measure the general public cannot work and are not good because you cannot get a random sample. but you have got harris interactive that claims it can make a living off of it, doing polling.

a lot of people think that is bad polling. but when the national council of public polls, a group of major pollsters, decided to establish a board that would judge some polls, one of the persons on there was from harris interactive. there could never be any combination of internet polling, because it is a part of the industry. this illustrates that there are a lot of standards that are difficult to police. the american association of public opinion research does set standards, good practices, best practices, but on the other hand, it doesn't address the kind of issues i'm concerned

about. i don't think you will see the polling industry police itself the way you have suggested. >> the news would have us believe that sarah palin is the best and since sliced bread. are we being manipulated? dr. moore: i will talk about as far as the polls go. the polls, at this point, when the polls measure public opinion about sarah palin, they do at the same way they measure opinion about other candidates. cbs will say, "do you have a favorable opinion, and unfavorable -- an unfavorable opinion, or don't you have one opinion the one way or the other?" cbs is the only organization that bends over backwards to allow people to say they don't have an opinion. all the others say, "favorable, unfavorable?"

i would say, with respect to measuring favorability towards sarah palin, that the polls are measured pre-much the way you would measure any other candidate. i think there is manipulation everywhere when it comes to election. i don't want to comment specifically on that. >> to get a better idea of how you think polls ought to be conducted, if you were trying to determine, right now, who would carry new hampshire between mccain and obama, how would you

rephrase those questions? dr. moore: if i were phrasing the question, i would mention the names. i would mention that in the presidential election in november, who do you expect to vote for? then i would ask for democrats, barack obama and joe biden, republicans, john mccain, and surveillance -- john mccain and sarah palin? i would list, "some other candidate" for the other independent candidates and have an option for those who have not made up their minds yet. for those who haven't, i would follow-up, by asking them whether they are firm or haven't changed their minds.

would ask those who are undecided if they are leaning. it gives me a white shade of opinion, from firm to not firm. what i would tracker the results of the first question, which would be the measurement of the two major candidates, other or undecided. i don't want to minimize the undecided as the campaign goes along. but, for further analysis, asking follow-up questions, you can get a better insight into what the shade of opinion is. that would be my recommendation. i know some will be following it, and others won't. any others?

>> there is a lot of discussion on your blog about internals. can you talk about that? dr. moore: you're talking about the extent to which a candidates support varies by age, by region of the country, by party identification or income. the race has not changed, but has polarized. the internals of suggests there has been a polarization of the race, while the overall results suggest no different. a lot of times, internals will shift because you have smaller groups. if you see some changes among the four income groups, at best, you have around 250 people. change can be kind of random. a lot of the time, you will find

stories about internals, that are what you might call, "analytical overkill." they're trying to find a real change here, change there. "he is doing better with white blue-collar workers who work at night." verses, you know, that kind of thing. [laughter] dr. moore: it doesn't mean internals aren't important, but given the general sample size of most of the major media polls, too much internal analysis is misleading. it can be helpful if you do it right. yes? >> it must be discouraging to

you and the pollsters to realize that there are people like me and a few of my friends who really don't care about your polls. we don't believe your polls, we don't believe everything in the newspaper, and we don't believe everything on television. we tend to think and apparently. what does that do to you? mr. moore: as far as i know you are the only person. [laughter] we will have to rethink our whole profession. >> i feel manipulated. mr. moore: i do too. all i can say is that there are a lot of people interested in the polls, and i am one of those people.

i always take a look at them. but i get frustrated because i know that they are not done right. but apart from that, i try to read them. we look at internals to see what they might mean. it gets to a point where i am interested in the technique into what people are doing. as far as you concerned, that is a different story and there is a significant portion of the american public that is not interested, that is why we have a decline in response rates. they are so low, out of every five people that are supposed to be included, we are lucky if we can get one. 80% of the people who we want to take part, we cannot get a hold of. some refuse outright, some refuse to respond, and others happen to be not there.

20% overall response rate is fairly common and even lower. so there are a lot of people who do not like polls in general. whether they are disgusted by them or feel manipulated by them, we have asked. in a poll. [laughter] mr. moore: i do not think that we've ever asked, do you still disgusted? or do you not know. [laughter] >> the, you hear people commenting on the polls, plus or minus a percent within this range, does that have statistical validity? how do they come up with that number?

mr. moore: the margin of error tells you how close the sample is, compared to the population at large. if you have 1000 people randomly selected for a cause and there are 2 million people across the country, they would represent the general population within plus or -2%. 95% of the time when you do a sample at large, you say so many percentage whites -- points. that is what statisticians tell us and i believe them. there is evidence to support, it is theoretical but something to be done, but the problem with polls is that they are not perfect, random samples. we have a high refusal rate or

nonresponse rate that we cannot even measure how different the sample is from the larger population as a whole. we cannot measure it. so, the plus or minus 3% is something i we can put out there, but it is a minor consideration when you try to figure out the real meaning of the poll. >> you mentioned that we have a romantic idea of ourselves as an active, well informed, concerned population. and that that ideal is held closely by the press and public institutions in general. in fact, the truth seems to be the opposite of that, that we are uninformed and generally self absorbed, that with most issues we do not even care.

does a poll support one or the other of those views? mr. moore: the current poll results show that everybody is informed. the reason is, if they are not informed enough, we tell them. but i think objective polls, we tried to get the undecided that will show that there is a substantial portion of the public that is not gauged on a given issue. but there is times when it is not. it varies by issue. on issues with foreign policy i'm a you can be sure that you'll have somewhere around 70%-90% of the public and engaged. should we -- unengaged.

should we recognize cuba? you ask, some people say yes, you get 70% may be who have no opinion. so, that is a fairly easy foreign policy to talk about it when you talk about whether or not whether we should support the initiative with china coming into the world trade organization, we found the 92% of the people who didn't. sure enough, we had 92% who expressed an opinion. i think, the word china or the country china, that might have triggered some kind of gut reaction.

but an actual question that dealt with china's trade, how much trade it they do, this is evidence of it, i am sure that you would find 90% of people unengaged. people have lives to lead. that is why we elect representatives. i do not fault the american public to be unengaged on a lot of issues. i do not have the time to keep up with it. and i do know that many times, the gallup, u.s. representatives, we have this very complicated question and we say, how will you answer it? we have to admit, we've no idea ourselves. we were probably as well informed as any of the general public given the fact that we are asking questions about the news and talking to news reporters, yet we cannot answer some of our own questions. occasionally, that would happen.

and so that comes back to the notion, i would never describe the american public as ignorant. i would say, we need to make a distinction for those who are and are not engaged, those who do not care at all, those who have an opinion. it is useful to know the truth about it, rather than to continually pretend as though you have a monolithic public. >> how do you square that with the decline and low percentage of the population that votes in elections, even presidential elections? mr. moore: i cannot remember what the exact figures were, but in the state of new hampshire, it was 75%, 80% of the people

vote. so it varies by state. in the primaries, it is very low. people are not very much engaged in the primary process. but in any case, it is measurable. the major point i would come away with is, it is not wrong to admit that on any issue, and a number of people are not engaged. it could be a greater engaged public. just let us know as well as we can. the truth is, this is about the public. >> do you have a concern about the low polls in news programs, and how much focuses on the horse race instead of the issues themselves?

mr. moore: polls have increased in the news cycle. i was reading -- and there is a chart on how polls were becoming essential part of news stories. it has increased. these polls to provide very important flavor of the political environment. it they are interpreted in a lot of different ways depending on who is talking. i do think that they make it easy for journalists to be able to talk about something. then they do not have to do as much research. but when you have as much cable news 24-seven kind of news, they , need more to talk about. i would not argue that we need more polls, but i would argue

that if they are going to be discussed, at least have the polls tell us the truth about what is going on so that we can be talking about that aspect, rather than about other interpretations. [applause] [captioning performed by the national captioning institute, which is responsible for its caption content and accuracy. visit ncicap.org] [captions copyright national cable satellite corp. 2016] >> you're watching american history tv. 48 hours of programming on american history every weekend on c-span3. , forw us on twitter information on her schedule. and to keep up with the latest history news.

>> the national wildlife refuge system came into being at the dawn of the 20th century. sought to safeguard the wildlife bounty. in that era, original hunters were pursuing big game into oblivion, and taking birds for their meat and feathers. conservationists are to stand at florida's pelican island. 1903, president theodore roosevelt established this tiny 5.5 acre island as america's first national wildlife refuge. and soon, the refuges followed. refuge was mountains created in oklahoma in 1905. to provide a haven for the american bison.

although these great shaggy beast's had once numbered in the millions, by the earliest 20th century, so few bison remained that the government had to ship 50 survivors from the new york zoo to form a starter heard for the new refuge. decreednctuaries were for from one and the open bighorn sheep. and after 10,000 elk starved to death by striking wintering grounds in wyoming, the government established the national elk refuge in 1912 and launched a winter feeding program. in the 1930's, the devastating droughts of the dustbowl years dealt another blow to wildlife, drying out the prairie marsh is that migratory waterfowl needed for nesting, feeding, and resting stops. hunters came to the aid of waterfowl, that proposal to purchase duck stamps, to

generate funds to buy and restore wetlands. after the act passed in 1934, the nations refuge chief, j clark sawyer, drove around the country scouring the landscape for likely duck habitats. by the end of the dustbowl era, he had added more than 100 new refuges to the system, many located along traditional waterfowl migration routes. in 1980, the system grew once again, as the addition of vast acreages in alaska provided protection for nesting grounds for millions of birds. and habitats for everything from giant kodiak bears to pacific salmon. now, more than a century after its founding, the system has expanded to include refuges in every state, as well as five territories.

and the system takes in virtually every type of habitat in the nation. , to the rain forest desert. tundra to the tropics. provide laces to rest and reproduce for literally hundreds of species of birds, mammals, fish, reptiles, and other creatures. are 250 in that tally threatened and endangered species. might not survive but for the presence of these protected lands. to enhance the survival of all the species under their care, refuge employees work behind the ,cenes, improving habitats building wildlife populations, and conducting research to learn

how to better meet wildlife needs. system way, the refuge is helping to ensure that the remain webs of life intact for future generations. >> this year marks the 130th annual meeting of the american historical association. american history tv was in atlanta for their congress -- for their conference. up next, and expiration of the death penalty in america, specifically the >> the georgia