By Michael Getler
February 8, 2008
On a night like Super Tuesday, even one mistake projecting a winner in any of the 22 Democratic primary races and 21 Republican contests is too much for some television viewers. And you can't blame them. This is serious business. Memories of the historically-botched Florida returns in the Bush vs. Gore presidential election of 2000 are still fresh. Even fresher are the forecasts of a big primary victory for Sen. Barack Obama (D-Ill.) in New Hampshire last month based on numerous pre-election polls, almost all of which turned out to be wrong.
So, when, at 11:18 EST Tuesday night, PBS's special election coverage provided by The NewsHour with Jim Lehrer projected that Sen. Hillary Clinton (D-N.Y.) had narrowly defeated Obama in the closely watched and hard-fought Democratic primary in Missouri, a fair number of viewers who had been paying close attention were probably surprised because the race had been extremely tight all night. A little later, some of these viewers got mad at PBS, not at the candidates, when it turned out that Obama had actually won a narrow victory.
Those who called or e-mailed me the next day invoked the memories of Florida, especially, and asked the traditional election night question of why not simply wait till all the votes are counted. Others said they understood such things happen but said they expect more of PBS. Actually, PBS was not alone but the episode is worth explaining because it sheds light on how election news unfolds on PBS, and how its operations differ from other news outlets.
Before I get to that, I want to make a momentary detour to say that this column also deals, farther down, with another thing about PBS — the rebroadcast of documentaries, in this case involving some earlier Frontline programs — that can also occasionally cause confusion for viewers and that is also worthy of some explanation.
An Unlikely Culprit
The main reason PBS got it wrong was that The Associated Press, which provides about as straightforward and reliable a news service as you can get, got this one wrong, and the NewsHour, it turns out, relies exclusively on the AP for such things.
Here is how NewsHour Executive Producer Linda Winslow puts it: "For the record, our policy is to only call a race when the Associated Press does, since they are consistently the most conservative. AP is also the major source of exit poll information and projections for all news outlets. Many broadcasters supplement AP's efforts with key precinct calculations of their own. We strictly rely on AP. Alternatively, we will report a race that has been called by two or more major networks using the phrase 'major news organizations have projected . . . ' and no projections are made until all the polls have closed in a given state. Missouri," she continued, "was the only state 'called' incorrectly last night. Obviously, we wish we hadn't been among those using the incorrect call, but we have no independent resources for checking the numbers."
And here is how this unfolded. At 11:03 p.m. EST, the AP declared Sen. Clinton the winner of the Missouri Democratic primary. At 11:18 p.m., Lehrer told PBS viewers, "We've just been told that Missouri has been called for Clinton." He didn't say who told them. A little more than an hour later, at 12:32 a.m., the AP backed away from that call. But the NewsHour special went off the air at 12:30 a.m., so viewers who only watched PBS went to bed with the wrong information. Then, at 2:48 a.m., AP declared Obama as the winner.
If you happened to switch to pbs.org for coverage online in those early morning hours after the PBS television special went off the air, you could have read an entry on the "Reporters' Blog" written by NewsHour staffer Alexis Matsui and posted online at 1:01 a.m. that was headlined, "Missouri too close to call; Networks Present Mixed Projections." It reported that "some major news networks" were now projecting an Obama victory and that these moves came "after many networks, including NPR (National Public Radio) and the NewsHour" had projected a Clinton victory. Matsui also reported that by the time of this posting "the Newshour had rescinded its call."
Then, at 2:34 a.m., Matsui posted again, reporting that "following mixed projections from major networks," Sen. Obama narrowly won Missouri. It said again that "other networks, including the Associated Press," had projected earlier in the evening that Sen. Clinton had won, but made no mention of the NewsHour. To which a reader responded: "Hey Alexis, aren't you going to 'mention' that you got it wrong? PBS signed off last night giving Missouri to Hillary."
Actually, I don't usually think of the AP and NPR as "networks," and using that term could be viewed as misleading because it implies that other television networks had also called this race incorrectly. It may be that I've missed something, but from all the reports I've read thus far, I can't find any indication that any of the other major television news networks — broadcast or cable — actually called this one wrong.
PBS and NPR did get this wrong, and the AP, which quickly wrote a good story attempting to explain what went wrong, said it found at least 18 newspapers that went to bed earlier Tuesday night using an AP story or graphic saying Clinton had won Missouri. But CBS, according to a couple of published reports, saw the original AP call, questioned it and waited, making their own call for Obama at 12:34 a.m. ABC, NBC, Fox, CNN and MSNBC also waited and got it right.
It's Not What You're Thinking, the AP States
The AP explanatory story said the initial bad call wasn't caused by "competitive haste" within the news organization. Rather, AP said, it was because at the time it was felt that while Obama was running well ahead of Clinton in St. Louis County, it would not be enough to erase Clinton's state-wide lead. The news agency said it was reviewing its procedures, that it takes "the credibility of our calls very seriously" and "we need to understand what happened here."
That also seems like good advice for the NewsHour. Normally, the AP is as solid a source as you can get. But relying exclusively on them obviously has been shown to contain at least some risk. This was only one primary in one state for one party, and PBS otherwise did a solid job, and is respected precisely because it is usually so careful. But should this happen again, or be repeated on election day in November, it will be a much bigger deal, which is why episodes such as these always should get candid, no-punches-pulled scrutiny internally that challenges the easy answers.
Other television networks, as I mentioned, questioned the call and waited. On the other hand, PBS is not like the other networks — indeed, it is a service rather than a network — and does not have its own reporting infrastructure. PBS doesn't produce content. Individual affiliated stations and production companies do that. So it is the work of MacNeil/Lehrer Productions and WETA in Arlington, Va., the flagship PBS station serving the Washington area and co-producer of the NewsHour, that PBS relies on for the nightly national news. They don't have anywhere near the number of reporters or producers that CBS or other commercial networks do.
NPR, on the other hand, does produce a great deal of its own content and has a large reporting, editing and production corps of its own. My fellow ombudsperson at NPR, Alicia Shepard, has produced an illuminating report on how the Missouri episode unfolded there.
The NewsHour did do an extensive follow-up segment on Missouri Wednesday night but mentioned only very briefly that the race was "so close that many news organizations, including this one, prematurely called the state for Hillary Clinton." I would have said "some" news organizations, since PBS's TV competitors, as far as I can tell, did not make a call for Clinton, and a few added sentences of explanation about how this happened, including mention of the AP, might also have been appreciated by viewers.
Watching Repeats of 'Frontline'
Frontline is PBS's top-notch public affairs documentary series produced by WGBH in Boston. When a program airs for the first time, it usually gets a lot of viewers and media attention. Because of their quality and subject matter, many of these programs are frequently re-broadcast by many PBS-affiliated stations around the country.
But last week, a couple of viewers wrote to call attention to confusion that can accompany these re-broadcasts. This gets complicated, but it is an interesting issue because it involves programs that may or may not be updated to reflect developments that have taken place between the original airing of the program and the time when one or more stations chose to rebroadcast it. It also involves other programs that Frontline has updated but in ways in which some viewers, one of whom wrote to me, feel that the updates are superficial and misleading in order to make an older program feel new. And, it also is another reminder that all of these stations are independent and can air what they want when they want.
Here's how this unfolded in letters from viewers and responses from Frontline:
Last night I watched Frontline's opening segment on KCET about the new journalism, a rerun as I had seen it before. It's a fine program. However, I was stunned by the fact that, although the program gave a complete history of the L.A. Times takeover by the Tribune Co., and the following turmoil, there was no update or comment whatever about the fact that the Times has since been resold to Sam Zell. So anyone not in the know would be completely in the dark concerning recent events that have taken place since the program originally aired. There should have been an update.
Ron Bottorff, Newbury Park, CA
Here's the explanation from Frontline's story editor, Catherine Wright. It first explains that most PBS stations that night carried a different Frontline program, one that had been updated:
"Last night, February 5, most PBS stations broadcast Frontline's updated report 'A Dangerous Business Revisited,' which was a significant update of our 2003 investigation into the cast iron pipe foundry business. Since our original report, quite a few things had happened, and so we replayed most of the original broadcast with correspondent Lowell Bergman, who then reported on what happened after our program aired, what has changed at McWane, the central focus of the report, and whether the company has become a less dangerous business. While we don't always undertake to update older programs as significantly, we do update our programs for national repeat broadcasts to bring up to date the important facts, events, and cases in the stories.
"However, individual PBS stations are also free to rebroadcast other Frontline programs in their schedules, and unless they alert us that they are doing so, we don't generally know that they do. If a station contacts us, we are happy to work to provide them with updated material to include in the rebroadcast, but many stations don't have the resources to do so on their own. Should we decide to rebroadcast 'News War' we would include an update on the changes at the Los Angeles Times and the sale to Sam Zell — you're right that it's a key event with respect to the LA Times story, and the larger landscape of media ownership."
And there was this from a viewer in Athens, Ohio:
"Last Tuesday (Jan. 29), you aired 'Return of the Taliban' with a new introduction, and a new copyright date of 2008 at the end of the show. This was completely dishonest, as it was an already-aired show, except for the new introduction that made it sound like it was new. Throughout the show, the narrator and those interviewed put a time-stamp on what they are saying, using words like, recently, a few months ago, today, a year ago, etc . . . that makes it seems like it was at the end of 2007, etc . . . BUT IT WASN'T, so, it alters the perception of what has gone on with the Taliban by not dating it correctly. What a sneaky thing to do. Please clarify to your viewers that this was an old show. I think you owe your viewers a correction."
In this case, Wright responded to the viewer's criticism and what she described as "the impression that it left that it was being broadcast as a new program." Wright continued: "I wanted to write to thank you for making us aware of the potential confusion. It has not been our intention to mislead viewers about rebroadcasts or pretend older programs are new, but rather to reframe past programs in light of more recent events. Given the confusion, we are looking at ways that we can make this more clear going forward. Thanks again for calling this to our attention."
That sounds reasonable to me.