What if you had a choice?
G. R. Boynton
Presented at the 2009 meeting of the APSA

In the 'golden age of television' the chief source of information about presidential candidates was campaign ads the candidates ran on television. We have known this since the 1972 election when Tom Patterson and Robert McClure conducted a panel study investigating the information sources used by citizens and what was learned from each source (Patterson and McClure, 1976). Two quotations from their book make the point.

Stated differently, those individuals who faithfully tuned in network news during the election learned not much more than people who spent that time doing something else. (Patterson and McClure, p. 50.)

During the 1972 presidential election, people who were heavily exposed to political spots became more informed about the candidates' issue positions (Table 2). On every single issue emphasized in presidential commercials, persons with high exposure to television advertising showed a greater increase in knowledge than persons with low exposure. (Patterson and McClure, p.116)

If you ask, even today, an overwhelming percentage of Americans say that television is their most important source of news. (Pew, December 23, 2008) But Patterson and McClure found that people who "faithfully" got their news from network news programs might as well have been doing something else in terms of what they learned about the candidates.

How does the combination of learning little from news broadcasters and learning much more from the television ads of the candidates work? The answer for the news broadcasters is twofold: the horserace and the seven second sound bite. News broadcasters talk at great length about the election campaign, but their chief topic for reporting and interpreting is who is ahead and who is behind and why. What the candidate has in mind for the country only slips in from time to time. When the candidate appears in the report his or her appearance is remarkably brief; it is seven seconds or less. (Brabazon, November 27, 2008) The candidates' ads, however, are very specific about what the candidate has in mind. If you see a campaign ad a dozen times about some subject, whatever that subject may be, you get it. The campaign ad is simple and direct. It is designed exactly to communicate what the candidate has in mind. Of course, you do not learn much about who is ahead from the candidates. For that you have to go to the news broadcasts.

Where does this practice of television journalism leave the candidate? Candidates speak only when spoken to. They have to wait to be asked a question, and if the journalists do not think you are a very important candidate you may wait a long time. And speaking time is now seven seconds which is hardly enough time for the candidate to articulate his or her view about the subject of the question. This is the candidate's fate unless the candidate buys television time for their own 30 sound bites.

Where does the practice leave citizens? Television viewers are left without much choice about what they can learn. The viewer has a choice over which of the programs being broadcast to watch. But the viewer has no control over the ads being aired. They appear in between segments of choice, and the viewer does not know if it is going to be soap or a presidential candidate. Even if it is the candidate the viewer wants to learn about it may well be a subject the viewer does not want to learn about. It is there: take it or leave it.

The golden age of television was golden for network profits. But it left candidates spending much of their time raising hundreds of millions of dollars to buy television time, which boosted network profits very nicely. And it left citizens with very little choice about learning what the candidates had in mind.

YouTube changes both constraints. The candidate may speak about any subject and at whatever length he or she chooses. And citizens have a choice. They can learn what they want to learn about the candidates' views.

YouTube is about choice. How did they exercise their choices in 2008? The McCain campaign produced and posted 120 videos to YouTube. The Obama campaign posted more than 700 videos to YouTube. On average the McCain campaign videos attracted 163,000 views, and that was substantially more than the average of 53,000 views of the Obama campaign videos. Citizens chose to view the campaign videos a total of more than sixty-two million times.

It is possible to push the analysis of choices well beyond these few numbers. The analysis will be constructed as an examination of views of the videos, that is, what viewers were doing when they had a choice. However, in examining choices about viewing the videos the choices of the campaigns will be illuminated simultaneously.

The mean views were 163,000 and 53,000. The figures below provide more information about the distributions.

Views of McCain Campaign Videos	Views of Obama Campaign Videos

There are many fewer videos in the figure for McCain videos and their upper value is 2.2 million whereas the upper value is just under 1.8 million for the Obama campaign videos. In both cases the number of views for most videos is very low relative to the upper values. This is even clearer if you look at the mean values for the 20 least viewed. The mean number of views for the 20 videos receiving the fewest views was 5725 for the McCain campaign and 2142 for the Obama campaign. The numbers rise very gradually and then the increase is very sharp. In both cases most of the videos are well below the mean value, and only a few exceed it. But they exceed it by such a margin that it produces distributions like these.

So, what were the favorites? What did people view most frequently when they had a choice?

The table has too much information to take in at a glance. But there are three characteristics of the table that are particularly important.

One, notwithstanding the very large differences in the mean number of views of the videos of the McCain campaign and the Obama campaign when you look at the views of the top 20 of each campaign the views are essentially the same. The Mean number for the top twenty videos of the McCain campaign was 595,132 and the mean for the top twenty of the Obama campaign was 609,940. The very minor difference is in the direction of the Obama campaign.

Two, three of the top four videos are about the opponent. The two most viewed videos of the McCain campaign were "Celeb" and "The One." Both were famous or infamous. "Celeb" compared the celebrity of Obama to the celebrity status of Britney Spears and Paris Hilton. An empty headed celebrity is not likely to be a good leader was the implicit message of the video. And "The One" is a reprise of Cecil B. deMille's The Ten Commandments with Obama cast in the role of Moses. Obama was too full of himself to be a plausible candidate. "American Stories, American Solutions," the most viewed of the Obama campaign videos, is the only one of the four about the candidate. And it is a twenty-seven minute, made for TV video, that the Obama campaign broadcast on network TV. It must have been a hit on TV because it was certainly a hit on YouTube. The second most viewed Obama campaign video was "Keating Economics." It reviewed McCain's participation in the major scandal of the 1980s when he was censured by the Senate for the type of support he gave to the chief villain of the scandal.

Three, the focus of the top twenty videos of the two campaign was quite different. Focus: a video might give reasons for voting for the candidate or it might focus on reasons to not vote for the opponent. Three of the top four are reasons to not vote for the opponent. For the McCain campaign 13 of the top 20 give reasons to not vote for Obama. Five give reasons for voting for McCain. And two did both. The Obama campaign videos are just the reverse. Thirteen give reasons for voting for Obama and seven give reasons for not voting for McCain.

The point here is what viewers chose. Individual supporters and opponents of the candidates decided to view "Celeb" and it became the most viewed video of the campaign. The viewers of the McCain campaign videos were more interested in viewing videos about Obama than about McCain. The viewers of the Obama campaign videos were also more interested in videos about Obama than about McCain -- though anti-McCain videos were not too far behind. The individuals going to YouTube to learn about the candidates wanted both focii. They wanted to know why to vote for each candidate and why to not vote for each candidate.

Which were the videos that very few wanted to view? What were viewers not into?

The table is again very big. Two things are important. First, the mean number of viewers for the 20 least viewed McCain campaign videos is 5725 and the mean for the Obama campaign videos is 2142. These are videos that not many people saw.

The McCain side of the table is easy; they are policy related videos. The Virtual Town Hall videos came in dead last. That format may have worked well on the campaign trail, but it did not attract many viewers on YouTube. And his invitation to Obama to join him in town halls, "Anywhere, Anytime," was only viewed 5986 times. Clean coal and jobs are most of the rest. The upshot: McCain policy oriented videos did not attract many viewers. They were the bottom of the barrel. Policy was not what they wanted to learn from McCain on YouTube.

It is not as clear looking at the Obama side of the table. Most of these were part of a strategic decision to use videos to reach out to local areas, which is something that cannot be done very effectively through television advertising. In this table you have Virginia, Delaware, New Mexico, Wisconsin, New Hampshire, Minnesota, Michigan, and Nevada. Seventy-one of the 100 least watched videos of the Obama campaign referenced a town, a state or both in the title. How this was used is clear in this table.

The table shows the states and localities mentioned in 362, or half, of the Obama campaign videos.

A large majority mentioned swing states, states in which the outcome had been close in 2008, and most mentioned swing states that had voted Republican. Video is cheap. They used the videos to reach out to localities knowing that their appeal would not be broad. If you lived in Michigan you might want be reminded that you could register by October 6. But most people do not live in Michigan, and are not likely to care to watch a video with that title.

Policy: in the golden age of television you got information about the horse race from the news media and learned what the candidate had in mind from candidate ads. So, when people went to YouTube to learn about McCain and Obama how much did they look for policy?

When viewers go to YouTube searching for videos about all they have to go on is the title of the video. If it is entertainment there will be a small photo that may help, but with politics the thumbnail images all look pretty much alike. There is plenty of policy talk in videos that have titles that do not indicate the subject covered. But a viewer has no way of knowing which do and which do not except via the title. So, the counts are counts that are based on titles.

These are the basic numbers:

The 28 policy oriented videos of the McCain campaign was a greater percentage of their total videos than was the 61 of the Obama campaign. The difference in total views was not as great as the difference in the means per video. There was also a substantial difference in the shape of the distribution. The top 5 McCain videos are much, much greater relative to the rest than are the top 7 Obama videos.

While the subject of all the videos might be examined, the real question is what did viewers want to learn? Looking at the videos that were viewed most frequently is a way to answer that question.

The difference between the two is striking. The top policy oriented videos of the McCain campaign were about Obama. "Education" was a video claiming that Obama had supported sex education for kindergarten students when he was in the state legislature. Sex sells! to the tune of 580 thousand views. "Pump" was a video about Obama's [Democrats'] failure to keep the prices of gasoline down. "Troop funding" was the claim that Obama had not supported the troops in Iraq. Taxman was a claim about Obama raising taxes. And "Barack Obama Forgot Latin America" is odd. But they are all reasons to vote against Obama. Viewers did not seem to want to learn about McCain's vision for the country. His town hall videos that were explicitly policy oriented, for example, bombed by any standard. They were interested in reasons to not vote for Obama.

Obama's most frequently viewed videos look very standard. Two were about taxes, and how Obama will cut them for most people. Three are about the economy. One is about education, and one is about energy. They are his vision for the country, and his vision is what viewers seemed to be seeking.

Policy was not the most frequently viewed subject on YouTube. 3.4 out of 62 million is not a big number. This is, of course, a considerable underestimate of the policy content that was in the candidates' videos. But it suggests very clearly the difference in what viewers wanted to learn from McCain and Obama.

Vice presidents: One of the carnards of American politics is that no one is really interested in the vice presidential candidates. At best they might appeal to a region or a state, and some could not even bring the voters in their home state to favor the ticket. There is not much evidence to back up such a claim, however. The viewing of videos featuring the vice presidential candidates provides a means to estimate the interest in vice presidential candidates relative to the interest in the presidential candidate.

The McCain campaign is not much help here. They were assuming the carnard and Sarah Palin was only featured in three videos and shared a fourth with McCain. "Introducing Gov. Sarah Palin," run before the Republican convention, was viewed 68,078 times. "Sarah Palin Speech Highlights," taken from her speech at the convention, was much more popular; it was viewed 483,862 times. "Alaska Maverick," immediately following the convention was viewed 291,036 times. And "Original Mavericks," also right after the convention, featured both McCain and Palin and was viewed 300,555 times. Those are pretty good numbers, but that is where it stops.

The Obama campaign had many more hours of video available, and they could use Biden much more frequently than the Republicans used Palin. As with the section on policy, the title of the video is used in counting the appearance of Obama and Biden. If neither name was in the title the video was not counted.

One hundred and fifty-eight of the more than 700 Obama campaign videos had Oama's name in the title, and 68 had Biden's name in the title. The difference in videos between the two is reduced if you count only the videos including Obama's name in the title after the Democratic convention when Biden became the vice presidential nominee. The difference from that point is 109 to 68. The mean number of views per video is the best way of estimating the difference in interest in the candidates. After the convention that ratio is three to one: 61,196 compared to 21,249. A second way to look at this relationship is given by looking only at the work-a-day videos. There were big events featuring Obama. But there were even more events that were not big. If you ignore the top 10 percent what does the ratio between presidential and vice presidential candidate become. The ratio is then 2 to 1.

It is not a surprise that more attention is paid to the presidential candidate than to the vice presidential candidate. But people going to YouTube to learn about the campaign and the candidates did not ignore the vice presidential candidates. They wanted to learn what they could about Palin and Biden. The Republicans might have exploited this interest more fully. And a 3 to 1 ratio for the Democrats is a better estimate of interest than "no one really cares."

Celebrities/entertainers: After Obama, the star of the 2008 election campaign was "Yes we can," a video made by Will.i.am in support of Obama. It has been viewed more than 30 million times, which far exceeds anything the campaigns placed on YouTube. YouTube is a venue for entertainment. When politicians beard the entertainment lion even the best of them do not come out ahead.

The Obama campaign wanted the youth vote. What better way to appeal to young voters than through the entertainers they followed so closely? Will.i.am had shown it could be done. So they called on entertainers to make videos in support of the Obama campaign. And they particularly focused on getting out the youth vote.

There were three in August, 6 in September and 40 in October and the first days of November. Forty was their effort to encourage young voters to register and vote.

That was the idea. How well do the results suggest that celebrity can be transferred into political effectiveness. The results suggest not very well. The total views for all 49 videos was 876,224 and the mean number of views was 19,472. Those are small numbers. The videos were used late in the campaign. There were only a few weeks, or even a few days, they were available. That may help explain the numbers.

If you look for the entertainers that were viewed most frequently there is an interesting finding.

Ashton and Demi made a video with a cardboard cutout of Obama that was viewed more frequently than any of the others. Jay-Z was included with two videos. Natalie Portman, James Taylor, Bruce Springsteen, and Dave Matthews are all well known entertainers. Chris Paul is the only entertainer, basketball player, who is not an entertainer in normal parlance, and he was at the bottom. But at the bottom, with most of the 49 having more views than this, is Will.i.am -- from 30 million to eleven thousand.

The other interesting feature of this table is the difference between the views by November 4 and the views as of almost a year later. These videos have staying power. The increase in the views of the two Jay-Z videos are quite remarkable; 80 to 200 thousand and 80 to 571 thousand. Views of the Bruce Springsteen video doubles. And all of the others have gone up in the interim. Celebrity and search on YouTube makes for views that came too late to do what they hoped.

Viewers went to YouTube to learn about the candidates. They do not seem to have been 'persuaded' by the entertainers. The mean number of views of 19,472 is far below the mean for all Obama campaign videos. Even in this venue viewers choose what they want to see, and in this case they wanted to see politics instead of entertainment.

What is the upshot? Multimedia communication has changed decisively. The socio-economic colussus that dominated communication for 50 years is being transformed. The networks will surely continue to be major producers of entertainment material. But the transmission of that video will no longer be under their control. They are currently scrambling to figure out how they can continue to operate in the new communication environment that is the web.

The 2008 campaign is the leading edge of that change in political campaigning. The McCain campaign campaigned as though nothing had changed. The Obama campaign took advantage of the new freedom to produce materials for every audience. And viewers were able to make choices; they could learn what they wanted to learn. Neither, the freedom of the candidates to speak nor the freedom of citizens to learn what they wanted to learn, had been possible for 50 years. We are entering a new era of political communication.

References

Tara Brabazon (November 27, 2008) Mantras For Modern Times, Times Higher Education.

Thomas E. Patterson, Robert D. McClure (1976) The Unseeing Eye: The Myth of Television Power in National Politics, G. P. Putnam's Sons, New York.

Pew Research Center for the People and the Press (December 23, 2008) Internet Overtakes Newspapers as New Outlet.

(c) G. R. Boynton, 2009.