POLITICAL POLLS

October 13, 1998
	With the November elections weeks away, politicians will be watching the polls closely to see what effect the impeachment inquiry decision has had on the electorate. After a look at current polls, Jim Lehrer and guests discuss the history and impact of political polling.

Sept. 23, 1998: Three pollsters examine opinion surveys on President Clinton. Sept. 22, 1998: Members of the House Judiciary Committee discuss Clinton's grand jury testimony. Sept. 15, 1998: Kohut analyzes public opinion of President Clinton. Aug. 19, 1998: Is the public sick of the Lewinsky story? Aug. 18, 1998: Kohut explores America's opinion of its president. Browse the NewsHour's coverage of the Starr Investigation, Coversations On Clinton, and Omaha's Views.   Pew Research Center White House	JIM LEHRER: And now speaking of history, we have some history and context on polling from NewsHour regulars presidential historians Doris Kearns Goodwin and Michael Beschloss and journalist/author Haynes Johnson joined tonight by Ben Wattenberg, senior fellow at the American Enterprise Institute, and host of the PBS program "Think Tank." Michael, when and how did the use of political polls begin?
	In the beginning.
	MICHAEL BESCHLOSS: Well, go back to the Harrisburg, Pennsylvania, in 1824, as well we might, and their editors got the idea of asking their readers to write in and say which presidential candidate you're supporting, and other newspapers got the idea, and gradually they got the idea of actually printing ballots in the newspapers. And not only was this not very accurate because it just rested on whoever happened to send in this printed ballot, but oftentimes these newspapers would use these ballots and these polls to log roll for their own party and their own candidates, because in the late 19th century many newspapers will link to a political party and so for instance the Springfield Republican would pool its readers and 90 percent would say we're voting for the Republican candidate, so on the front page of the paper there would be a big article saying guess what, there's a big bandwagon for the Republican. They felt that this was editorially a great way to help their own candidates. It got better in the 20th century. 1916, the Literary Digest began polling in presidential races. JIM LEHRER: How did they do it? MICHAEL BESCHLOSS: They did it much more scientifically. The idea was to try to get a sample that penetrated a cross section of the American people. They did well for 20 years using techniques that we now would find inaccurate and caused them in 1936 to have an enormous gaffe. That was when they found that Alf Landon would win over Franklin Roosevelt with about 59 percent of the vote. The problem was they were polling people who owned automobiles, people who owned telephones. And they tended to be pretty Republican in 1936 because those people were fairly rich. JIM LEHRER: And Doris, when did it start to get serious and start to get more professional?
	Polls become more scientific.
	DORIS KEARNS GOODWIN: Well, it seems like in the 1960's, television networks and newspapers came together to form much more scientific polling. And I think it's true that it made it more accurate. And you didn't have the kind of huge mistakes that you had with the Literary Digest in 1936. But one of the things we've seen now, as it becomes more scientific, as it becomes so pervasive that the poll, itself, becomes part of the story, when you think about those last weeks before a presidential election, for example, the main topic of conversation in the newspapers often is how is the person doing, the horse race becomes the story. And then you combine that with a Teddy White kind of reporting on presidents, the inside story and the tactics and the strategy, and you have the inside story mattering more than the outside story. And I think that's part of what diminished our politics in these last few years. Not only what we always say – that leaders are too dependent on the polls and the public is too passive -- but that the whole inside reporting has become the story, which means we don't trust words anymore. There's no authenticity if you know that everyone's polled every sentence of every speech they're making. So even as they become scientific and much more accurate, they become much too obtrusive, in my judgment. JIM LEHRER: And, Haynes, when you were a newspaper reporter, you used to at election times, you used to travel the country talking to people. And now a lot of people in journalism would say, well, reporters don't do that anymore; they just get their organization's polls and that's what they know about what the American people think.
	Focus groups.
	HAYNES JOHNSON: What you do is go to a focus group. You sit there behind a – you go somewhere off to a communal place and you have a window. They can't see you, and you look at the people there, and you have a pollster who's asking the questions. And out of that comes the story that tells you how the people are thinking. There may be 20 people in the room. You're not getting your shoes dirty and so forth. You don't go around precinct to precinct. I remember when Ben and I used to interview Ben Wattenberg about his expert polling data and so forth back in the – years ago – but I – the use of the polls – Doris is exactly right – they -- BEN WATTENBERG: With Richard -- HAYNES JOHNSON: With Richard -- the real America. That's right – little plugs here back – but I mean, the change – you really have to go back and see, Jim, the idea of not going out of town. Everything is – almost -- we're polled to death, and everybody feels that this thing has taken it out of control, almost. JIM LEHRER: Now, Ben, when did politicians start using them seriously in a way that affected maybe a position or a speech or a decision? BEN WATTENBERG: I would guess that was probably in the 1970's when the – I mean, in the – at the time of Watergate there were only three national regular public opinion polls – not partisan polls – Gallup, Harris, and I guess CBS or something like that. And now there are just scores have become a mass industry and it's a great tool of democracy. But, you know, anyone who relies on it seriously regularly as a sole lens to see what's going on in the electorate is loopy. It's just not that good a tool. It can be perverted in a dozen different ways.
	Improving polling techniques.
	JIM LEHRER: I want to get to the perversions in a minute. But just – refresh us on the technique. For instance, Literary Digest, Michael, what did they do? They asked their people to write in, or did they – how did Literary Digest gather their information in the -- MICHAEL BESCHLOSS: They sent out postcards. JIM LEHRER: Postcards. MICHAEL BESCHLOSS: They sent them to these car owners and telephone owners, and then they got them back but -- JIM LEHRER: They didn't attempt to weight them by -- MICHAEL BESCHLOSS: No. JIM LEHRER: By income, or by race, or by region, anything like that? MICHAEL BESCHLOSS: Not even by state. And that poll included only 21 states. And so the result was that you had this very bizarre sampling. JIM LEHRER: Now when, Ben, did it start that you could do 1,000 people – Andy was just telling Margaret about his group – just polled 1,000 people. When did it get to be 1,000/1,200 people and then be able to extrapolate that into what the total public opinion in America is? BEN WATTENBERG: I think that would have been with John Kennedy and Lyndon Johnson in the 60's, but reaching its full flower in the 1970's, and I guess by the time Jimmy Carter came to the White House with Pat Cadell, I guess that was the first of the regular presidential pollsters, and Pat used to – we used to call the Watershed of the Week. I mean, every week, he'd say, my God, everything is turned upside down and the whole world has changed, and people don't like their families, and the fact is, the big argument in public opinion polling is always whether – we have a pretty good record going further back in the 60's and 70's -- whether it shows change or continuity in America. And there is a surprising amount of continuity, notwithstanding these vast social and economic – we've been – during depressions, world war – Korea – the 60's – all that stuff -- on a number of issues – patriotism, family, that kind of stuff – it's just very high, very level, very little change, and very important. JIM LEHRER: Speaking of importance, Michael, has there ever been a case before when polls have become as important as they appear to be now? In other words, all the events of the Clinton investigation – the release of the Starr Report – oh, we got to wait for the polls – the playing of the videotape – we wait for the polls -- and they become increasingly important. More important is it's just my imagination, or are they more important --
	A public opinion platform.
	MICHAEL BESCHLOSS: No, I think you're absolutely right, because in the 1990's this has become essentially licensed -- Read Dick Morris's memoirs. He tells about – I've read them so I can give you an executive summery -- and that is in about four sentences. 1995 and 1996 he basically said to Bill Clinton you want to get re-elected, you have to move to the center, and I'm going to use polling to find all sorts of initiatives that will have 60 or 70 or 80 percent of support. The acceptance speech that Bill Clinton gave at the Democratic Convention in Chicago, we were all there, was 90 minutes basically of Dick Morris's poll tested initiatives. The point is that that was pretty much known at the time. In the early 1960's, when John Kennedy had a pollster, Lew Harris discreetly makes soundings on things like civil rights and Kennedy's policy toward Cuba, he would actually send Kennedy memos that in some cases were unsigned in case they got into malicious hands so that it would not become known -- JIM LEHRER: Haynes Johnson and other reporters -- MICHAEL BESCHLOSS: Haynes Johnson might have uncovered this covert relationship between Kennedy and Harris, which Kennedy oftentimes used, especially in thinking about language to use in speeches. JIM LEHRER: Doris, help us now. Let's go back to say Franklin Delano Roosevelt's time. How did FDR decide what public opinion was at any given time on any given issue? DORIS KEARNS GOODWIN: You know, that's part of the artistry of what makes a leader. He would have public opinion polls brought in to him. There were some polls. Anna Rosenberg, his labor adviser, used to bring them in. He would read letters to the editor. He would talk to newspaper reporters. He would talk to people when he went around the country, as much as he could. And somehow he infused all of this into his very being and got a sense -- Isaiah Berlin, the British philosopher, said he had the best understanding of the trends of a country of any leader he's ever known, but the most important thing is then he took that sense of where the country was at and he figured out how do I move the country forward. What happens with these sophisticated polls now, if he had listened to the poll on lend lease in the fall of 1940, the majority of the people said, no way, we don't want to help England. But, instead, he said, okay, they don't want to help him now. Through my fireside chats, through my speeches, through the debate in Congress I will educate and shape and mold them to want to. By the end of that debate the overwhelming majority wanted lend lease. That's what leadership is. And now you get the feeling that they're hamstrung. There's too many Dick Morrises running around, giving polls in the ear of the presidents. JIM LEHRER: Take the polls first and then decide what you think, rather than the other way around. DORIS KEARNS GOODWIN: That's right. JIM LEHRER: Haynes, as somebody who is – as a journalist – not as a professional pollster or any of that sort of thing – what guidance would you give to an average newspaper reader or television viewer as to how – because there are --every day now there are polls -- what attitude should they have when they read them?
	Take with a grain of salt.
	HAYNES JOHNSON: Take a deep breath, step back, look at the long trend. Polls are very useful if you take the long trend on say college students entering – freshmen every year – it tells you about values for 30 years now. It tells you a lot about where the country is going. JIM LEHRER: The continuity point that -- HAYNES JOHNSON: These are very useful. But the overnight polls, the exit polls are just a quick snapshot and it can change. The other problem, which we haven't really kind of – us the political consultants – the Morrises – who use these to form and forge false attitudes about an issue. You can take an issue like Willie Horton. Crime is a problem. Now we'll make that our big campaign thing. BEN WATTENBERG: Crime was a very important issue. HAYNES JOHNSON: Of course. But you can take it and magnify it to a point for political purposes. And now what I think we're seeing on the television out there – that camera – that screen – we're seeing the consultants every day coming on, talking about their view of the truth of the poll, even though they're taking only a snippet of this or that. BEN WATTENBERG: Let me – something -- just the kind of stuff they shouldn't pay attention to – this is a Fox News opinion dynamics poll. Do you think Lewinsky was in love with Clinton or the office of the president? Do you think political power is an aphrodisiac? Do you agree with James Carville that Ross Perot is about a half a quarter low? Do you think sexual relations with Clinton and Lewinsky were normal or kinky? This is what you're getting now. It's a lot of fun to read but – JIM LEHRER: What does history tell us, Michael, finally, about how accurate polls are now? MICHAEL BESCHLOSS: Polls are hugely accurate nowadays as a snapshot, but as leaders use them, they have to be willing to throw them away. If Kennedy for instance in ‘63 had used Lewis Harris's polls on civil rights, he would have said, I shouldn't send a bill to Congress because they're going to drop my public approval ratings. Jimmy Carter, who used Pat Cadell then, is absolutely right, and Haynes elaborately more so than perhaps any president had used a pollster got a poll saying, you support a Panama Canal Treaty, it'll kill you, Carter said, thank you for the advice, but forget it; I'm doing what I want. JIM LEHRER: Okay. And we must leave it there. Doris, gentlemen, thank you all very much.