Witnesses:

Joan Konner, Professor of Journalism, Columbia University; Michael Traugott, Chair, Department of Communication Studies, University of Michigan

Mr. Philip Zelikow: Thank you very much. I think we have heard from some competent and honest election officials, and very beneficially. We�re going to go without a break directly to the next panel. It�s panel three on media projections, if they�ll take their seats. This panel has only two witnesses and will run for a shorter time span than our previous panels. Our first witness is Joan Konner. Joan Konner is a professor of journalism at Columbia University. She is former dean of that school. She came to our notice for those reasons but also because she ran the CNN project on election coverage that followed the 2000 election, essentially the principal force behind the task force which CNN, to its credit, commissioned to examine its own behavior and that of the other news services and the conduct of their affairs and, I think, has some suggestions to make as a result of that work. Dr. Konner.

Dr. Joan Konner: President Ford, distinguished Commissioners, thank you for inviting me to testify at this hearing; it�s a great honor. I�m Joan Konner, professor of journalism at the Columbia Graduate School of Journalism and former dean of the school from 1988 to 1997. Before going to Columbia as dean, I was a television producer for 25 years in commercial and public television. The subject today is media projections, an issue that was a key concern in a report I co-authored with Jim Risser and Ben Wattenberg this year, commissioned by CNN. We were asked to look into what went wrong in CNN�s television coverage of the presidential election and what might be done to prevent such mistakes from happening in the future. A copy of our full report, "Television Performance on Election Night 2000," is available to this Commission as part of the record of these hearings.

The CNN report, and all the other reports that have been issued about Election Night reporting, recognized that something went terribly wrong. CNN executives, correspondents, and producers themselves described Election Night coverage as a debacle, a disaster, and a fiasco. We agreed. The report states, "On Election Day 2000, television news organizations staged a collective drag race on the crowded highway of democracy, recklessly endangering the electoral process, the political life of the country, and their own credibility, all for reasons that may be conceptually flawed and commercially questionable."

I�d like to address my remarks today to two main points. The first is to the context of the report and the second to some of its substance. Our inquiry, judgments, and recommendations were based on the ideals, the principals, and the best practices of journalism. It was not a political statement or a legal opinion. Our panel�s criticism was based on journalistic principles, stated in the report, that the central purpose of a free press and a democratic society is to provide the public with information upon which the people can form intelligent decisions concerning important public matters on which they have the power to act. And that public affairs journalism is the pursuit of truth in the public interest, and its major values are accuracy, fairness, balance, responsibility, accountability, independence, integrity, and timeliness. We concluded that because of several key factors, all the networks failed in their core mission to inform the public accurately about the outcome of the election. Specifically, CNN and the other networks failed in the reporting [of] election results in Florida, which turned out to have the key to the outcome of the election.

About the substance: we found that the faulty journalism resulted from excessive speed and hyper-competition, combined with an overconfidence in experts and reliance on increasingly dubious polls. We reported an impulse to speed over accuracy, and we attributed that impulse to the business imperatives of television news: to win the highest rating, which is not a journalistic standard, but a commercial standard. Ratings-that is, the size of the audience-drive the commercials, and commercials determine the bottom line profits of the corporations that own the news networks.

Our report also found several flaws in the system set up to cover the election. We questioned the overall concept of the Voter News Service, which was the single source of information and data on which all the networks relied. VNS was set up as a partnership among competing news organizations. This unusual collaboration among competitors was conceived principally as a cost-saving measure, although the network maintained pooling resources enabled the networks to greatly expand their polling reach. We believe that relying on a single source of information contradicts well-known, deeply entrenched, best journalistic practices. Relying on a single source eliminates the checks and balances built into a competitive vote-gathering and polling system. It eliminates the possibility of a second source for validating key, and possibly conflicting, information.

We further question the purpose of then introducing the element of competition through independent decision desks at each of the networks, all of whom rely on the same data and information received at exactly the same time. What results is a speed trap in which all the networks are doing their complicated calculations under maximum competitive pressure in minimum time, usually making their so-called competitive projections minutes apart. The compulsion to be first led CNN and others to project results without checking other possible sources of information, and at the time that the Bush call was made, there were in fact two other sources available, the Associated Press, which does its own vote counts, and the official returns of the state.

Our inquiry also indicated serious flaws in the polling methods used by VNS, including exit polling, outdated polling models, and outdated technology. We noted that polls inadequately take into account the growing number of absentee ballots and early mailed ballots or the variations caused by a wide variety of factors, from non-responses to the quality of the questionnaire and the questioner. We note the polls are, in general, statistical calculations, not factual realities, and, as such, they are an imperfect measure of voter intent and actual voting, especially in close elections. Although the overall record of projections based on polls is good; in close elections the margin of error, statistically, and the human and technological margin of error, actually make any projection based on polls more risky than presumed. We concluded that this overconfidence in polls is not deserved.

Our recommendations included the following: that exit polling no longer be used to project or call winners of states, and that exit polling be used for analysis only, that returns from sample and key precincts no longer be used for projecting or calling winners. We believe that model precincts are subject to many errors and can lead to faulty calls. We recommend that all calls be based on actual counted returns and that no calls be made in states where polls are still open, something the networks have already agreed to. We recommend that no call be made until all available sources of information are checked, and we recommended that the Voter News Service be re-examined, repaired or reinvented.

We note that many of these recommendations would probably slow down the process of reporting, and we believe that�s a good outcome. We believe that slowing down would improve network performance and would demonstrate that accuracy was more important than speed. Our report expressed the view that the mistakes in the reporting of the presidential election were damaging to journalism and to the country. The erroneous call for Gore and the later call for Bush, declaring him prematurely the next president based on faulty numbers, undermined the credibility of the news organizations and distorted the real result of the election at that point.

Some have charged the networks with bias in their reporting: that is, deliberately of unwittingly calling or withholding the results of the race to benefit one candidate over another. We found no evidence to support that view. We also found no convincing evidence that calls made before polls are closed, within a state or in another state, have impact on voter turnout. All of CNN�s election coverage was made with the best journalistic intentions, but mistakes were made and they, along with other networks, contributed to the public atmosphere of rancor and distortion during the painful post-election challenge and debate.

Because I have this privilege of testifying before this distinguished Commission, I�d like to take another minute to go beyond the scope of the CNN report and to speak as an individual, as journalist and citizen.

Mr. Zelikow: I do need to warn you that you�ve been out of time for some time, Dr. Konner, so if you could come to a close.

Dr. Konner: I�ll cut it there. I do not believe that what happened in Florida in the last election was just an accident of one unfortunate Election Day, rather that this Election Day was an accident waiting to happen. The events should not be permitted to disappear into history as if they were an [unclear] in a system that fundamentally works. In fact, the system did not work.

Mr. Zelikow: Our next witness is Michael Traugott. Professor Traugott is professor of communication studies and the chair of that department at the University of Michigan. He is, perhaps, the outstanding expert in the country on the issue of media reporting of election results and, I understand, has actually been seen in the precincts of VNS from time to time assisting them with their labors. Dr. Traugott, thank you for joining us today.

Professor Michael Traugott: Thank you very much. I want to first thank President Ford and the Commission for inviting me to testify today. I participated in the past conference at the Ford Library on the presidential nomination process organized by George Grasbark, and I am especially pleased to be back and have the chance to speak to you again. My comments today are actually divided into two parts: the first bearing upon your general effort to gather information about administrative reforms that might facilitate voter participation and the second on the impact of media polls and election projections.

Elections are the most important thing that we do in a democracy. They provide a regular opportunity for citizen participation in government; they�re the essence of our representative process that�s a hallmark of our system, and they legitimate the orderly transfer of political power in a way that citizens in many other political systems envy. Even during the often contentious and ambiguous month following last November�s election, Americans never seriously doubted that our system would produce a winner or that whoever it was would be invested with the full authority of the office of president and the support of the American people.

On the general matter of election reform and its impact on participation rates, I brought along a copy of a recent paper reviewing recent efforts to stimulate political participation and evaluating their success. To summarize those findings very briefly, first of all, most reforms are designed to serve one or both of two purposes: to increase participation and to make the participating electorate more reflective of the population as a whole: in the smaller sense, to make it more democratic. The best empirical evidence from a large number of scholars engaged in a wide variety of studies is that these reforms have had only modest success in either regard. That is to say, the increase in turnout has generally been small, rarely exceeding the high single digits in magnitude, and as a result they haven�t changed the composition of the electorate very much.

Both of these findings are the results of the fact that reforms can work in one of two ways; they can act as a way to mobilize citizens and bring new people into the electorate, and they can make it easier for relatively active citizens to continue to participate. The research shows that most reforms seem to have their greatest impact on retention, compared to mobilization. As a result the electorate shifts in ways that make it look more like past voters than those who have traditionally not been participants. There are five different sets of explanations for electoral participation: legal and administrative restrictions on registration and voting, social/economic characteristics of citizens, and the resources they provide to voters, social/psychological precursors of political involvement, personal efficacy, [and] the impact of economic conditions and political mobilization.

In the short term the government can only deal with the first set of legal and administrative restrictions, but these may not be as important in determining participation as attitudes or personal economic circumstances. The prospects for wholesale changes in participation rates is low, and the character of the electorate is unlikely to change very much as a result. Does that mean that we shouldn�t try new reforms? The answer is clearly �no.� Normatively, we�re obliged to do what we can to spur participation and change the character of the electorate with reasonable cost. But one thing that�s needed now, and that the Commission can help with, is a conversation that distinguishes normative goals from reasonable expectations for change or improvement based on our current state or empirical knowledge. How much of an increase in turnout should we expect from any particular reform? What is enough change? Is there a reasonable amount of change? And does it make any difference in how much change we get when some reforms are combined with others?

On the matter of the impact of polling, generally-and media projections, specifically-we have a very complicated public policy issue of quite a different sort. I believe that there�s an accumulation of incontrovertible evidence that there are effects, usually small, of exposure to such information in the public. This evidence comes primarily from experiments rather than from surveys, and that means that while we�re confident that the effects occur, we cannot estimate their prevalence in the population very well. At the same time we have a clear and well established set of precedents in case law that make it clear under the First Amendment to our Constitution that there is virtually no law that can be passed to limit the rights of news organizations to disseminate such information. This is clearly a feature of the American system of government that distinguishes us from most other democracies in which there is some constraint in the publication of polls late in the campaign or on Election Night projections.

It should also be pointed out that one of the most important functions that�s served by pre-election and exit polls in both political and social terms is their ability to give American voters a voice in the campaign in interpreting and explaining the meaning of an election. The analysis of why voters decided what they did, rather than the preoccupation of the horserace aspects of coverage and the emphasis on who�s going to win, provides an independent check on the stories that spin-masters promote about the meaning of an election to advance one or another political agendas. This is an important function that exit polls can serve based upon a long historical record of accuracy. If you are interested during the question period, I could say a little more about the nature of these effects on the mass public and how they vary at different times in the campaign period, and I can also discuss the fact that these effects are not all negative; some of them are quite positive.

But I will conclude by indicating that nowhere could you find a more powerful potential consequence of exposure to exit polls than when you observe one of the major party candidates about to make a concession speech based upon an exit poll projection. Thank you again for the opportunity to appear here today, and I look forward to discussing these issues with you.

Mr. Zelikow: Thank you very much, Professor. We�ll open the panel for questions; begin with Mr. Seigenthaler, then Ms. McAndrews, then Senator Boschwitz.

Mr. Seigenthaler: Thank you, Mr. Chairman. I would first of all say that in Joan Konner we have one of this nation�s distinguished journalists. She was and is one of this nation�s distinguished journalism educators. I�d like to call on you to use your expertise and, if I can, get you to voice a considerate opinion on just three or four very quick questions. First of all, what�s the journalistic value in a network calling an election for president three, four, five minutes ahead of the other networks?

Dr. Konner: The journalistic value is to receive the praise of the critics who play a role in this and always point out who called the race first. The implicit judgment [is] that they have a keener, smarter, faster team, and time is of the essence; it�s part of our field: deadlines are what we live with and who got to the deadline first, and also to garner the audience, and to get a larger audience.

Mr. Seigenthaler: Well, does it enhance ratings? I mean, if, looking back on this election, the extent that states were called accurate and most accurately-and most were-does it enhance the ratings in one network over another? To be able to say�

Dr. Konner: I think you could actually have a controlled experiment, and PBS, which did not call the race and did not come up with projections and winners-unless they were quoting other networks-and their ratings are lower.

Mr. Seigenthaler: Their ratings are lower. Now let me ask you about your expert opinion with regard to credibility. Just take all of the networks. Take NBC, CBS, ABC, MSNBC, FOX, CNN, Public Broadcasting, and C-SPAN. In your judgment, which of those would you rate with the highest credibility, audience aside? Which one would you rate the highest credibility?

Dr. Konner: Outside of Election Night?

Mr. Seigenthaler: Outside of Election Night.

Dr. Konner: I think that CNN, in being totally dedicated to news, comes out very high on that list, and they further reinforce their credibility by being the only network to take a serious look with outside peer review at what was the performance on Election Night. Other than that, the three networks are neck and neck and are usually only rated in terms of their audience size.

Mr. Seigenthaler: And how would you rate Jim Lehrer in terms of credibility?

Dr. Konner: I would say very credible, but PBS is really not in the running on breaking news.

Mr. Seigenthaler: Well, I understand the audience is not there, but that . . .

Dr. Konner: Well, audience can be rated on quality as well as quantity, and PBS and Jim Lehrer get a very substantial audience in terms of interested, active listeners.

Mr. Seigenthaler: And final question: looking back on what happened, is the risk of being wrong, considering the variables you studied and found, worth the gamble of winning what amounts to a three or four or five-minute horserace?

Dr. Konner: I have to answer that question personally: I think obviously not. I think that credibility is the most important by-product for new organizations, and all of them lost credibility and public trust in this election.

Mr. Zelikow: Ms. McAndrews.

Ms. Colleen McAndrews: I�m from California, and we�re very concerned in the West over the impact on depression of voting based on the early call. As you�ll recall, in 1980, the networks called the election for President Reagan very early, and the Democrats in California thought some of the local legislative races went Republican because Democrats then failed to come out to the polls because it was already known what was going to be the winner. The same thing was accused in the 2000 election when the polls were called for Gore in Florida. I heard anecdotal evidence of a phone bank that shut down in a close Congressional district in California because the Republicans just said, �Well, it�s all over,� and they all went home and that Republican Congressman lost very narrowly.

So there is this fear-it seems intuitive on a human level, even though I know it�s hard to measure-that there would be this impact. People think the election for the top office is decided, so they don�t come out to vote. I heard you very carefully use the word, "no convincing evidence" of any kind of reduction in the vote. I also note that I don�t care that much that the networks embarrassed themselves by wrong calls after the polls are closed-after all, The Chicago Tribune did that in 1948 and every student gets to see that newspaper being held up and their embarrassment-I do care that they call the vote, rightly or wrongly, before the polls are closed. If the networks have agreed to not project until the polls are closed in one state, what do they think is different that allows them to call it nationally, before the polls are closed nationally? So, wouldn�t the values that are used to say, �We won�t call it in one state and have the Florida effect, where the panhandle was still open after they called� apply to limiting themselves until the polls are closed nationally?

Dr. Konner: I don�t think so, actually. I don�t think they�re parallel. I think that they have information about a particular state, and they have a right to broadcast that information. And I don�t think it�s parallel to say one state is equal to, say, 50 states.

Ms. McAndrews: Well, I would disagree. Secondly, I agree the First Amendment is probably not going to let us limit the networks; they do have the right to exercise that. However, I�d like you to comment on a suggestion that�s come before our Commission that there be some kind of embargo on election results by elections officials at the local level or the state level and that by denying the networks this information until a uniform poll closing time, such as 11:00 on the East Coast, 8:00 in California-we would, of course, be discounting Hawaii and Alaska-that at least for 48 states, there would be an embargo on this kind of information. And I would like to know what you think about this and especially if, Professor Traugott, if you think that would inhibit the networks if they were denied the information of official or semi-official poll results to marry up with the exit poll information.

Professor Traugott: Well, you have several distinguished legal scholars on the Commission panel who know a great deal about the First Amendment and the case law in this area. As I expressed in my testimony, my own view is that, given this history of case law, the Commission could better devote its attention to other issues that might affect participation rates than this particular element.

Mr. Zelikow: You may misunderstand the question, Mr. Traugott. It�s not an embargo on reporting by the news media, it�s an embargo on the disclosure of official results by election officials.

Professor Traugott: Yes, I understand, but in terms of the existing models for projecting the outcomes of particular states, there are a variety of components that are involved here. One of which is the rowboat, which comes into play with special significance when the outcome looks to be very close. But if the outcome is not very close, and in most contests in the United States the outcomes are not very close, then the embargo on the states might make it more risky to make projections, and the networks would have to face that consequence. But it wouldn�t make it impossible to make projections just from exit poll interviews. It would be a very complicated kind of decision for them to have to make. But it would not, strictly speaking, prohibit them from doing so.

Dr. Konner: Also, they do their own vote counting. The AP does its own vote counting, so the official state count is only another source. And the AP vote counting plus their own VNS vote counting would give them some kind of result. And there is an inconsistency in what you say; so what about Hawaii and Alaska? You know, they�re states, too, and if you�re going to do it in 48 states, you should do it in 50. And I also say that knowing projections in advance could play either way, and it has not been proved that it plays in favor of the winner. The panhandle has independent voters lined up at the polls; they can vote to turn the state around. Nobody really knew the vote count and that it could have played in favor of Bush and might have played in favor of Bush. And there are other races on that ballot; the presidential race is not the only one. The presidential race is a spur to get more voters out to vote in local races and state races, so why voters turn around, or even if you could prove that they did turn around, is a mystery. And I do believe a voter has some responsibility for staying in that line as long as the polls are open.

Ms. McAndrews: How does the AP do this vote count if they aren�t relying on�

Dr. Konner: People at the polling places.

Ms. McAndrews: Just as the vote is counted, they call it in?

Dr. Konner: As the vote is counted. Yes, yes.

Mr. Zelikow: Excuse me. Are you saying that local election officials in counties allow reporters to sit with them, monitoring the tallies?

Dr. Konner: Yes.

Mr. Zelikow: Then why don�t the reporters report the early vote tallies that are available to those same officials at noon?

Dr. Konner: That�s too early. It�s too early. It�s not too early statistically; they could do it. I guess when the polls close, the county poll close[s], as it did in Florida before the panhandle closed, the AP is there at various precincts getting the official tally.

Mr. Zelikow: Now, our understanding is that reporters get the official tallies from the officials, that reporters aren�t actually participating in the vote-counting processes themselves, that local officials provide that information to the reporters who are present.

Dr. Konner: And if you forbid local officials to pass on any information whatsoever, would it affect the AP count? I suppose it would.

Dr. Robert Pastor: I think they can observe the count in many different counties; in some counties they can�t. An observer/reporter can observe the count.

Mr. Seigenthaler: I know that in many states newspapers poll the precincts, people in larger communities. And every daily newspaper is a member of the Associated Press and feeds its results into the Associated Press. Now, in some rural areas there are not enough personnel to man every precinct in every state, and in those rural counties they do rely, indeed, on the local officials to give them the returns. But the massive numbers of votes that pour into AP come largely from the newspaper members who man the precincts, not the actual county election commission office.

Dr. Konner: The L.A. Times also did its own vote count in its area.

Mr. Zelikow: I have Senator Boschwitz, Senator Gorton, and Ms. Ravitch on my list, and I�m afraid we have to close the list. Senator Boschwitz?

Senator Rudy Boschwitz: You say that Jim Lehrer has high credibility; I�m a persistent watcher of that, and I don�t know what that does to his credibility, in the eyes of some, you know? But I would like to know-I enjoyed your testimony very much, Ms. Joan Konner-what did the networks say to your report? You indicated that CNN apparently was forthcoming, but were the networks defensive? Did they listen? Did you get any indication that they might change their ways?

Dr. Konner: As far as I know, CNN is the only one that published some modification of its reporting procedures. They certainly did not accept every recommendation we made, but they did, in fact, accept some of them. I�m not aware that there was any other official response from the networks. I was present at the congressional hearing when all of the network presidents appeared, and all of them acknowledged what a fiasco election night coverage was. That doesn�t answer your questions. They acknowledged the fiasco. Have they taken any action? I�m not aware that they�ve taken any action. Although I do know that they have, in fact, started a review of VNS, Voter News Service, procedures and started to put in place machinery to help correct some of the failing.

Professor Traugott: Senator, one proposal was to ban VNS, and in the last 72 hours there has been a press release that VNS will continue with the support of the networks but with an infusion of additional financial support to help them to revise their models and their data collection procedures.

Dr. Konner: But that does not address some of the fundamental problems. You can�t really, by polling a statistical analysis, account for the growing number of absentee voters. You cannot compensate for non-responses, both in exit polls and in pre-polling. Polling is becoming more dubious-less dubious than it was in the 1980s-procedure today than it used to be.

Mr. Zelikow: Senator Gorton?

Senator Slade Gorton: In Oregon we have a mail-in system. What are your thoughts on how that affects the existence or potentiality for fraud? And what are your own subjective views on what it does to a communitarian spirit or civic attitude to have no polling place and the impact of a very large percentage of voters voting well before Election Day?

Professor Traugott: I can do that because I have been studying voting-by-mail in Oregon since 1996. Oregon, as you know, because it is a neighboring state to yours, is a very high participation state to begin with. Vote-by-mail does have a positive effect on turnout; our estimate from our research is that it is somewhere between four and six percentage points. We think that this is probably durable although there has been some tick down in participation rates in Oregon since they have gone into holding all of their elections by mail.

With regard to the question of fraud, there is obviously a wide variety of ways in which fraud can be introduced, and also attempts at fraud can be hindered depending on your willingness to pay the cost. Oregon has a system in which there is a 100% signature verification. Every ballot comes back inside a blank envelope, a so-called secrecy envelope, which is placed inside a security envelope that has a signature on it. And they have 100% signature verification. In the largest counties in Oregon, they have very high levels of computerization. In fact, they have files of digitized signatures that they can use to compare to the envelopes. They generally describe the cost of election administration as quite low in Oregon because of voting-by-mail, but that ignores, really, the capitalization of this computerization. For other jurisdictions to take up something like this, they would have to make a pretty substantial investment in computers.

They have a variety of systems to try to detect fraud, including an 800 number that people can call in and register complaints. There has been a lot of survey research in Oregon. There has been no indication of significant fraud: any complaints by citizens or any discovery of fraud by journalists, who play a very important part in this process. I agree, however, with previous witnesses that you may not be able to translate the Oregon experience directly to other jurisdictions.

The question about communitarian values is a very complicated one. One of the things that you lose in a system like voting-by-mail, it is claimed, is standing in line with your neighbors and having a chance to discuss politics. Some people in Oregon, first of all, maintain that receiving ballots at home allows-and in the Oregon system they also get a voter pamphlet-kind of a conversation to take place at the dinner table during this twenty-day process that may be more rationalized and informed. I would say that it is also important to point out that when you see the final data from the year 2000; it is probably going to be the case that more than one-third of Americans voted before Election Day. If you take into account early voting, voting-by-mail, permanent absentee registration, I believe that in Washington one-half of the voters, or 60%, were cast before Election Day.

On the flow of the vote, our study reveals some interesting things. First of all, that there does not seem to be any partisan differentiation in the flow of the vote. Of course, there is this twenty-day period, but the most partisan and most involved voters cast the first votes. The votes that trickle in at the end-in the Oregon system they can also be carried in and dropped off-come from the least involved and least motivated voters. They have not seen this yet in Oregon, and there is a peculiarity in the Oregon system: that they make available, on a daily basis, information about who has voted and who has not voted. But we expect to see, because of their adaptability, that candidates would modify their campaigns eventually and alter their messages so that they are speaking in one way and mobilizing their most committed supporters early in the campaign and then maybe later in the campaign having more issue-oriented messages to get the less interested and motivated voters to take part. So there is a very interesting set of consequences for candidate behavior as well as for voter behavior.

Mr. Zelikow: Ms. Ravitch, and then President Ford will close this panel.

Dr. Diane Ravitch: I will try to make this fast. First of all Dean Konner, that was a really fine report, and I congratulate you on it. I wanted to ask Professor Traugott a question because I am a historian and I tend to look for patterns. And historians are always getting it wrong because they find the wrong patterns, but I noticed in your paper you talked both about participation and then about polling. I began to think about the connection between the two of them: how we have seen participation steadily decline as polling has steadily increased. And, of course, I wonder whether there could be some causal relationship. I have this kind of secret theory that people - when they see their candidate very far behind or very far ahead - don�t think it is important to vote. They think that they can be a spectator. I am just curious to know whether there is any indication one way or the other and whether there might be a relationship there.

Professor Traugott: I don�t believe there is a relationship there, but let me explain to you why that is. Journalists have always been interested in portraying political campaigns and elections with a particular style. Even before the advent of the period of modern polling and current methods, there were various kinds of attempts dating back to the turn of the century to make assessments of where the electorate stood and to report on this. So I think that one thing that modern polling does with improved methods is to provide generally more reliable estimates than might have been done in the past. We think about the contemporary period of turnout decline starting with the election of 1966 when the turnout was at its peak about 65%. But, of course, the modern polling period we think of starting right after the war. There was a big boost in turnout in the post-war period up until 1960.

I would add one other thing. The comment that you make about polls that show one candidate relatively far ahead of another, this experimental evidence that I referred to probably does have a depressing effect on participation, but there are occasions, of course, that polls show that a race is very close. We could think of that as circumstance when it would be just as likely that a reporting of those results would stimulate interest in the campaign and increase levels of participation. This is partly why I made this reference in my testimony to both positive and negative effects.

President Ford: In those states where you have a very high percentage of absentee voters such as in Washington-in California it is forty-some percent-has anyone ever made a study as to how the people who voted by absentee compared to the people who voted on Election Day?

Professor Traugott: There actually have been a fair number of such studies. I would summarize the results of that research in the following way, Mr. President. The Republicans seem to have a kind of organizational advantage usually when these new reforms are adopted. We see sometimes, for example, in the case of the Bradley-Deukmajian race in California where initially there might be a slight imbalance, a partisan imbalance, in the returns on absentee ballots, which are really a function of the mobilization efforts of the party organizations themselves. But the Democrats seem to catch up pretty quickly, and after a period of stability sets in, we don�t see very much by way of partisan differences.

President Ford: So a candidate, by getting more absentee ballots to the poll, doesn�t necessarily get a better advantage over his opponent?

Professor Traugott: If the opponent is just as well organized, or the opponent�s party, on that assumption, yes.

Mr. Zelikow: Thank you very much. We appreciate your testimony and the expertise that you have offered to the Commission. We will now go again without a break, in the long march of the Commission this day, to panel four, which addresses special problems with voter access.