Why the polls missed in New Hampshire

So far the only answer the polling gurus can offer is both obvious and trivial: there were a large number of late deciders and they broke for Hillary.

However, an article in Salon.com by Rebecca Traister suggests it was motivated in no small part by outrage on the part of women about the media treatment of Hillary Clinton. This explanation actually makes a great deal of sense because it would account for why it was late-breaking, and in some cases involved women voting for Hillary Clinton who actually preferred Barack Obama, and a powerful unidirectional trend. If there is a strong surge of support in one direction TC thinks there must be an explanation. So far this is the one that makes the most sense.

I can't believe I'm saying this, but had I been a New Hampshire voter on Tuesday, I would have pulled a lever for the former first lady with a song in my heart and a bird flipped at MSNBC's Chris Matthews, a man whose interest in bringing Clinton down hovers on the pathological, and whose drooling excitement at the prospect of her humiliation began to pulse from the television last week before most Iowa precincts had even begun to report results.

Before any tallies were in, Matthews was observing, based on early projections, that if Clinton received the expected 30 percent, it would mean that seven of 10 Iowa voters did not like her, a mean little metric that he did not apply to the other candidates. "It's hard to call yourself the people's choice if two-thirds of the Democrats are voting against you!" he burbled.

He was not alone in his glee. There was the Washington Post's Eugene Robinson, announcing before the caucuses had concluded that if Clinton lost Iowa, she would likely lose New Hampshire too! And South Carolina! She'd be lucky to scrape by with small states like Nevada, Matthews crowed. Newsweek's Howard Fineman was also excited. "If [Obama] wins this thing, even by one vote in Iowa, then that five-point lead of Hillary's [in New Hampshire] is going to disappear in a second," he said. Pat Buchanan recommended that in her still purely imagined concession speech, Clinton "be very, very gracious." It wasn't just the guys. Andrea Mitchell might as well have had canary feathers hanging from her mouth as she reported from Clinton's Iowa campaign headquarters on the "manufactured" crowd gathered for Clinton's concession speech.

Ding-dong, the witch is dead! Which old witch? The Clinton witch!

Read the whole thing here. Also look at the Gloria Steinem article cited in the Salon piece.

Kathleen Hall Jamieson on Obama, Huckabee and the Press

I have long been an admirer of the analysis of Kathleen Hall Jamieson - of polls, politics and political communications.

She was interviewed by Bill Moyers this week about Iowa. Here are two excerpts:

BILL MOYERS: What did you hear with Obama and with Huckabee? With Obama?

KATHLEEN HALL JAMIESON: Obama delivered a speech in which he cast himself in the role of the President of the United States, as opposed to a candidate seeking that office alone. The speech is an attempt to try on the presidency and see that it fits. Obama's a very strong stump orator. And one of the things that we realize when we see the extended speech of Obama is that he is a much weaker debater. He's much weaker when he's speaking one on one to reporters. He's much weaker when he's speaking to camera. And he's good in all those formats than he is as a stump speaker. As a stump speaker, he is a master.

BILL MOYERS: And Huckabee?

KATHLEEN HALL JAMIESON: Huckabee demonstrated that he is very good at speaking intimately. Less well crafted speech — he wasted a lot of time at the beginning of the speech. But where Obama's a natural stump orator, Huckabee's much more effective at intimate use of a stump platform. Ronald Reagan could do both. He was a great stump orator, and he was great at intimate communication. Huckabee is very good at kind of low key, intimate conversational engagement. Obama, much better at rallying the masses.

But here's why both of those speeches were important. They were good speeches. They talked to the nation in the role of a candidate who is speaking as a president to a people. Giving people a chance to say how would you fit in that role. And we forget sometimes that speech making is a very important role in the presidency. There are times in the nation in which the president is the only one who can speak to us and for us. And whether it's the president we wanted elected or not, that person has to be able to play that role for all of us. Obama has that capacity, and I believe Huckabee does as well.

And an interesting point on media distortion of Hillary:

BILL MOYERS: Let's turn to the press. You and I both know that every primary creates a new reality, just as every experience creates a new reality, so that the press today has a new narrative. What's the narrative you're reading now about the primary process?

KATHLEEN HALL JAMIESON: Something pernicious happened last night in press commentary. The commentators on each of the networks that were covering live — so the major cable networks — managed to say at, at least one point, that two-thirds of the Democratic voters had rejected Hillary Clinton. And then they provided explanations for why they had rejected Hillary Clinton. Nothing in the polling data tells you that anyone rejected Hillary Clinton. But the press frame is an either-or frame, a zero sum frame game. And as a result, it doesn't open the possibility for its viewers that people could look at the Democratic field and the Republican field and say, "Those are fine candidates. Any of those would be a good president. I would support any of those, but I prefer this candidate."

BILL MOYERS: It's a statement of preference, right?

KATHLEEN HALL JAMIESON: It was a statement of preference.

BILL MOYERS: Not opposition.

KATHLEEN HALL JAMIESON: And then, when after having set up this rejection of Hillary Clinton by two-thirds of the Democrats, then they provide the rationale for what the rejection means. Well, it's because she's too polarizing, she's too divisive. They also don't know that from the available evidence.

Full transcript is here.

My favourite book of hers is Eloquence in an Electronic Age.

New Hampshire and the Nomination

One of the traditional voices in American politics whose judgement I respect is Republican consultant Mike Murphy (who managed John McCain's presidential campaign in 2000). He was on Meet the Press today. Here is an excerpt:

MR. RUSSERT: Give me the nominees of the party.

MR. MURPHY: Oh, man. Obama.

MR. RUSSERT: Yeah.

MR. MURPHY: And the Republican one’s a lot looser, and it could go—it could still get unraveled. But if you put a gun to my head in the hot seat, I’d have to say John S. McCain.

The Clintons are fighters and will not give up easily. However, Obama seems to be doing what Howard Dean hoped to do in 2004. Interestingly TC has learned that almost all of the New Hampshire Dean team from 2004 are working for Obama.

On the Republican side it is clear that the party establishment can't stand Huckabee and will do all it can to defeat him. At the same time it isn't clear (as Murphy's comment reveals) that they will embrace McCain. In fact the Republican process looks decidedly scrambled. Ordinarily the nomination process is quite Darwinian forcing out candidates early. However, if the early primaries and caucuses go in different directions (already Iowa has supported Huckabee while Wyoming went for Romney), TC holds out hope (albeit faint) that the race might go all the way to the Republican Convention in Minneapolis in the first week of September thus forcing the eventual winner to start the fall campaign exhausted from the nomination battle.

Reviews from Iowa

Ezra Klein is one of the best of the young bloggers TC enjoys reading. I think his reviews of the candidate's speeches to their supporters after the results were posted in Iowa are worth reading. You can find them here: Barack Obama, John Edwards and Hillary Clinton.

Like Klein I hope Edwards stays in the race.

Iowa - not just a win for Obama

Here is the information I was looking for this morning post-Iowa:

With 98 percent of precincts reporting, Iowa Democratic Party officials reported 234,000 caucus attendees, compared to 124,000 in 2004. With 93 percent of GOP precincts reporting, 112,349 Republicans had participated in their caucus, up from the 87,666 who participated in 2000.

I also like this comment on Kos about the turnout:

In 2000, the last time there was a caucus in both parties, Republicans turned out 87,000 voters, while Democrats produced 59,000. There are around 600,000 registered Democrats in Iowa, and about 550,000 Republicans, but when you consider that on caucus nights, Republicans just need to show up and point to a name, while Democrats are committing to two hours of public wrangling, it's not a surprise that more Republicans show up to be "first in the nation."

Except for yesterday.

When the Des Moines Register poll was predicated on a turnout of 200,000, I was scornful. And they were wrong -- but only because they were too conservative.

Last night, the Republicans produced around 115,000 voters -- an impressive 30% increase.

But the Democrats turned out 236,000. That's an increase of roughly one whole helluva lot.

Hillary is still well-financed and organized enough to carry on to February 5th so the contest will continue, but the key outcome is in the numbers just cited. I think they are a clear indicator of the year as a whole.

Iowa - Beginning and End?

The American election process begins today with the Iowa caucuses and it has been drowning in media attention. However, one key fact that defines why it is important is almost never mentioned: being the first to win has become the be all and end all partly because of the political media culture that above all celebrates success. Winning the first state in recent elections has caused instant growth in support even for transparently weak candidates such as John Kerry because the fact of success becomes the central theme of media coverage and the bandwagon effect thereby created becomes self-reinforcing.

I would prefer to see the contest this time last at least until February 5th (when there will be primaries or caucuses in 20 states) simply because that day there will be a large enough number of primaries in enough big states for the process to do a better job of finding the best candidate. It has a chance to do so on the Republican side but only because all the candidates are seen one way or another as being seriously flawed by some key Republican constituency. The Republican contest is less important to TC as in his view it is simply going to chose a loser. The incumbent party is responsible for a major war, which will be longer than Vietnam (as a major conflict if we date large scale U.S. involvement as being from early 1965 to the 1973 Paris Peace Accords) within three years. It is also a party likely to bear responsibility for an impending recession.

On the Democratic side I see John Edwards as the most realistic of the candidates in recognizing the need to deal with a highly class divided and unequal society, but it looks more and more to me like Barack Obama could sweep all before him. He has been rightly attacked by the likes of Paul Krugman for a campaign that leans too far to the right on crucial issues such as health care, although I think his early opposition to the war in Iraq is a key offset to that. But his real political cachet is an argument that he is the candidate of hope. There is no doubt that in the past political campaigns in the U.S. (Reagan 84) and in Canada (1993 Chrétien Liberal) that offered a persuasive optimistic message, no matter how loose or vague the actual commitments in the fine print, were enormously successful.

Obama appears to have the combination of qualities for that kind of political success. However, anti-black racism remains a fact of American life. While I don't think it sufficient to deny him victory in November, it is likely that his lead on election day in November will be smaller than the closing polls suggest because of this factor. However that won't matter today. His problem for Iowa is that it is a process dominated in the past by the highly active party regulars. Obama needs to transform his ability to appeal to the typically non-involved into votes at evening meetings. The premise of the Des Moines Register poll of a few days ago that gave him a significant lead was that he could do this.

2008 Federal Election

There was a new poll of federal party preferences released today by Probe Research in Manitoba. It produced numbers that are almost a replica of the 2006 Election:

C – 44% (compared to 42.8% in 2006)

Liberal – 26% (identical to 2006)

NDP – 24% (compared to 25.4% in 2006)

You get the picture. These numbers would likely produce the same 8 C, 3 L, 3 NDP result in Manitoba as in 2006.

The national polls fluctuate depending on news but I suspect there is a core of truth in the Probe poll that applies more broadly. Most likely the campaign will shape the outcome in which case the underlying reality, which appears to have changed little since January 2006 despite the ups and downs of government and opposition in 2007, will matter little.

As this CP report summarizes:

When the year began, Harper's Conservatives were sitting at 34 per cent in public opinion approval, according to Harris-Decima, just ahead of the Liberals at 31 per cent. The NDP was at 15 and the Green party at eight per cent.

As 2007 draws to a close, Harris-Decima's rolling three-week averages had the Tories hovering around 34 per cent, the Liberals at 29, the NDP at 15 and the Greens polling 11 per cent.

In Quebec, Bloc Québécois support has dropped to around 36 per cent from 43 at the start of the year.

However, the government does look to be finishing 2008 on a down. The most recent Ipsos and Angus Reid polls report weaker numbers that appear to be a result of the Brian Mulroney-Karlheinz Schreiber show. According to my forecast model, the Ipsos poll would give us a Liberal victory with 140 seats while Reid’s numbers would put the Conservatives ahead but with just 117 seats. Blending the two sets of numbers gives us a narrow Liberal win of 127 seats to 118 for the Conservatives and 16 for the NDP.

The holiday season has driven the Schreiber affair from the headlines so its influence may not persist. More ominously, for Harper and the rest of us, the government is now acknowledging that it faces serious economic pressures. However, such early warnings will not exempt it from the inevitable political fallout (which may well still be some months down the road).

On the opposition side I suspect (but can’t prove) that Dion is consulting closely with Jean Chrétien. If so he is keeping it very secret. He has appointed John Rae and David Smith, two of Chrétien’s key political operatives, to key posts. The media interpretation emphasized their 2006 leadership connections, but I think the links to the former PM are more important. This has the potential to be a real source of strength for a leader whose weak grasp of the English language remains his greatest liability.

All the polls have shown growth for the Greens compared to 2006. Even though I think this support is overstated, it is important. The Angus Reid poll numbers actually produce one Green seat (but it is not Central Nova). I feel certain that in some way, shape or form, Elizabeth May will endorse the Liberals towards the end of the next campaign and this could potentially have a significant impact (if every Green vote had gone Liberal in 2006 Paul Martin would still be Prime Minister). Most prognoses ignore her role in this regard despite the fact that she has a deal with Dion and has defended him in the media. It is completely untypical of a party leader (bizarre is one word that comes to mind) because the potential to help the Liberals is equally the potential to hurt the Greens. Its abnormality is precisely what makes it difficult to weigh its impact. Although it is generally overlooked now, it might make a great deal of difference in a campaign.

Also somewhat overlooked is that the NDP’s numbers have generally held up (despite their weakness in the last Ipsos survey). They are averaging about a point below their 2006 showing (recent Liberal and Conservative fortunes have been worse). The growth of the Greens has hurt all parties including the NDP. Although Layton continues to vastly outpoll Dion and May in the leadership stakes, TC regards leadership polling numbers as a relatively weak predictor of electoral success. The NDP enters the electoral season, however, with no obvious weaknesses: it does have good leadership, strong organization and adequate finances. It needs issue leverage to make progress (and avoid losses) but it isn’t obvious how that could be accomplished.

Bloc numbers are down somewhat but not enough yet to make a significant difference in the outcome in Quebec despite Stephen Harper’s fawning attentions.

In the U.S. the underlying poll numbers make it clear a political sea change is underway. The future in Canada is not nearly so obvious.