Gallup Commits Malpractice

Community (This content is not subject to review by Daily Kos staff prior to publication.)

Monday October 08, 2012 · 8:39 PM PDT

2012/10/08 · 20:39

Gallup did something unprecedented today, and while they may have had good intentions, it was egregious nonetheless. As most of you know, they took the extraordinary step of breaking their 7-day tracking poll into two 3-day periods, and lopped off the seventh day.

I think they did this for a couple of reasons. First, their 7-day window is simply too long to reflect rapidly changing events - and the debate was surely that. While their 7-day tracker showed Obama with a +2 lead as of Sunday (Saturday to the prior Saturday), their data from the 3-days immediately following the debates was even. So they announced at 10 am this morning that the three days since the debates were even, but the three days prior to the debates were Obama +5. From what we saw from Pew, PPP, Ipsos, and Rasmussen, this is probably right on the mark. It is fairly obvious that Romney had a good day Thursday, a monster day Friday, and then Obama recovered somewhat on Saturday and much more on Sunday. I would bet that any poll spread evenly from Thursday-Sunday would show a tie in the likely voter model (The bulk of the Pew sample was Thursday, Friday, and Saturday).

So our friends at Gallup already had an issue where their poll did not reflect a rapidly changing landscape, so they made changes in their methodology to reflect this. That I have no problem with so long as they disclose it - which they did. They are not responsible for a lazy media that does not report it accurately.

So where is the malpractice. Yesterday was a very good day for the President. We saw it in Rasmussen where it is easier to extrapolate daily results due to their tighter 3-day band. But since Gallup now gave us their two most recent 3-day band, we could do the same thing. Obama had a monster day in Gallup on Sunday. If the average of the 3-day pre-debate sample for Obama were +5, and the average of the three days immediately following the debate were even, Obama would have to have something like a +19 day yesterday to get a +5 seven day sample. Now Gallup they may feel that the pro-obama monster Sunday number to be an outlier, especially if they are also in the field with the USA Today poll, but to not disclose the big Obama day and essentially bury it in the 7-day number that still includes three pre-debate days is unconscionable.

Please note that I do not have any issue with Gallup's numbers - I think they are right on the mark. When you consider that they are still in a registered voter model and the likely voter screen probably nets 3 - 4 additional points for Romney, they are consistent with pew over the same 3-day period. My point is that they took great pains to highlight the Romney debate bounce, and then did nothing to show that the bounce was likely receding. In fact, they obfuscated it!

My advice to Gallup is that if events on the ground warrant breaking the 7-day poll into two parts, they continue to run to multiple tally's until the period following the event catches up to the 7-day number. That is, when they released the 7-day number at 1:00 today, they should have also released the 4-day subset of post debate days. Tomorrow, they should release the 7-day number and the 5-day post debate subset, and continue this until the mini-tracker aligns with the full tracker.

Raison d'être and the Romney Campaign's "Tell"

dagman

Community (This content is not subject to review by Daily Kos staff prior to publication.)

Sunday October 07, 2012 · 8:22 PM PDT

2012/10/07 · 20:22

"Raison d'être" is a French term that means "reason for existence". A "Tell" is a poker term for the subconscious actions or mannerisms that give away ones hand.

Throughout the campaign, the Romney ticket has focused largely on the state of the economy as defined by the official unemployment rate published by the Bureau of Labor Statistics (BLS). That is, the economy is bad because unemployment is high. They could have focused on sluggish GDP, on the types of jobs created, or a host of other factors - but they didn't. They have painted themselves in a box. This probably looked like a winning strategy in the spring as the economy appeared to slow, but over the summer, things began to tick up. We saw it in such indicators as auto sales, consumer confidence, and the stock market - and are now seeing it in tax receipts. For the past few months, it seemed as if there were positive signs all around us, even the non-governmental employment surveys looked to be picking this up. ADP (the private payroll survey) was showing 50,000 - 75,000 more jobs per month than the government survey, and Gallup was showing an unemployment rate at 7.9% even while the government survey (BLS) was at 8.1%. It was only a matter of time until we saw a pick-up in the government numbers - the only thing we didn't know was when.

Now for the Tell. Politico had a story of the concern on the Romney plane Friday morning after the Jobs numbers. On a day they should have been euphoric, there was concern. Mitt's campaign was on life support before the debate. There was no energy, the polls were horrible, and big donors were defecting. By Thursday morning, his campaign was in the best shape it had been since before the Republican Convention. So why the grave concern? The good jobs report came 2 months too early for them. Yes, 7.8% is still to high, but there is a growing consensus that things are getting better, and getting better at an accelerating pace - they had just lost their Raison d'être and they knew it!

The Unemployment Rate Did Not Artificially Drop This Month

dagman

Community (This content is not subject to review by Daily Kos staff prior to publication.)

Friday October 05, 2012 · 4:33 PM PDT

2012/10/05 · 16:33

The issue is not that the official unemployment rate artificially dropped this month, it is that it remained artificially too high in prior months. There are two separate data points that support this.

1) Private ADP Report

If we look at the private ADP report historically, we find quite a dispersion in month to month readings when compared to the government BLS report. But over longer periods of time, the two reports are fairly consistent. The ADP jobs report has been significantly stronger than the BLS report for a good many months, and what we saw today (and earlier in the month in terms of revisions) was some of the catch-up from the BLS. It is important to note that ADP is the largest payroll provider in the country with a very diversified mix of customers (Size, Industry, and Geography). Their survey in the US is derived from a sampling of a subset of more than 344,000 business customers that employ more than 21 million workers. Given that they provide services to large companies (more than 1000 employees), mid-size companies (50-999 employees) and small companies (1-49), they are likely to pick up employment changes very quickly!

2) Various Employment Related Economic Indicators

Over time, we typically see a positive correlation between employment, and economic statistics such as New Auto Sales and Consumer Confidence. The reasons are obvious. Those without jobs tend not to be confident and have a particularly hard time qualifying for a car loan. Both of these indicators have been rising in recent months, much more so than the tepid BLS job growth of 100,000 - 120,000 would indicate.

Romney supporters such as Jack Welch understand this. Whatever you think of his motives, he is no idiot. That is why Jack is on the record as of last night predicting the rate would be 7.9%. The problem is, Romney's entire campaign is premised on the need to make a change because the President's economic policies are not working. They have no choice but to go all birther on the BLS report, because if most americans believe the employment situation is materially improving, it is game over!

Clinton, Obama and the Housing Bubble

dagman

Community (This content is not subject to review by Daily Kos staff prior to publication.)

Wednesday February 06, 2008 · 2:38 PM PST

2008/02/06 · 14:38

We have seen endless deconstruction over why certain states go for Obama, and others go for Clinton. Some seem reasonable and some seem downright silly. However, I have seen no acknowledgement of the apparent relationship between the housing bubble and vote totals. Voting is now complete in 6 of the 7 states where the housing bubble is concentrated, and all have gone for Clinton - by large margins.

dagman

Community (This content is not subject to review by Daily Kos staff prior to publication.)

Tuesday January 29, 2008 · 8:10 PM PST

2008/01/29 · 20:10

Exit polling showed that HRC once again won voters who decided in the last day - just like New Hampshire. In that race, the press was dancing over her grave in the 3 days before the primary.

We had the same thing after South Carolina, especially with the Kennedy endorsement. I am a big Kennedy fan and will be the first to tell you that he has been unfairly vilified by the media for more than a decade. The conservative media loathes him, and the MSM to be "balanced" always presents him in a pejorative light. But he endorses Obama, and every thing changes. He is now the liberal icon - the elder statesman of the party. Even David Brooks goes gaga over Ted Kennedy.

Voters are not stupid. They have seen Ted Kennedy unfairly pilloried for 10 years. He endorses Obama and he is transformed into the second coming of Abraham Lincoln. I suspect voters see this for what it is - the press piling on and trying to influence the outcome. Obama is dynamic and gets great press. Unfortunately for him, the press drives voters to Hillary when they cross the line and start shilling for him.

Full disclosure - I am a HRC supporter, but that is just the way I see it.

Bush will Fall

dagman

Community (This content is not subject to review by Daily Kos staff prior to publication.)

Tuesday September 06, 2005 · 9:03 PM PDT

2005/09/06 · 21:03

Bush will be forced to resign over this. The death toll will be staggering, and the stories gruesome. But the real damage will come when all the FEMA incompetence is exposed.

Nixon and his advisors knew it was over months before the resignation, and the same thing is happening with Bush. The straw that will break him is the use of FEMA (and by extension Homeland Security) as a patronage system.

As with Nixon, someone from his own party will tell him it is over. Who will it be?

Gays for Bush cost us Ohio

dagman

Community (This content is not subject to review by Daily Kos staff prior to publication.)

Thursday November 04, 2004 · 6:44 PM PST

2004/11/04 · 18:44

I was a big supporter of Gay rights (even if it costs elections) until this morning when I saw the detailed exit polling. According to the exit polling, between 20% - 25% of gays supported Bush.

The Democratic Party supports abortion rights and pays a political price, but virtually all pro choice women vote democratic. We support affirmative action at a political cost, but 90%+ of African Americans consistently vote democratic. We support gay rights, yet according to CNN exit polling, almost 25% of gays voted for Bush. 75,000 votes from Bush to Kerry delivers us Ohio and the election. And I am not homophobic - I am coming to the same conclusion regarding unions - their support was even worse.

No Fillibusters

dagman

Community (This content is not subject to review by Daily Kos staff prior to publication.)

Wednesday November 03, 2004 · 7:20 PM PST

2004/11/03 · 19:20

Posted in an earlier thread - but I think relevant.

The Republicans must be allowed to implement all of their policies. I work with all Repuplicans (mostly fiscal conservative types), and they took solice in the fillibuster proof Senate, knowing that the Fundies won't get much of what they want. They sort of think it will be like a divided government. When I mentioned the prospect of no Fillibusters, they became visibly concerned. Only by allowing the Repugs to implement all of their policies will you drive a wedge through the 3 discreet groups that comprise the Rebublican base 1) Fundies, 2) Limbaugh Losers, 3) True Conservatives Read More

Stewart was right about the Media Whores

dagman

Community (This content is not subject to review by Daily Kos staff prior to publication.)

Monday October 18, 2004 · 6:42 PM PDT

2004/10/18 · 18:42

Many of us think the SLCM media has it out for the Democrats, or that they are in love with Bush. This is wrong. As Stewart said, "they are simply in it for the Theatre". Take the "Bush pulling away" meme than ran as the headline on the MSNBC website this afternoon. The problem is it wasn't true and they knew it - but it made a big headline. The article talked about the four weekend polls w/ Gallup up 8, Newsweek up 6, WAPO up 4, and only Zogby showing the race tied.

If you believe the LV models, it was a credible story. The problem is that the data didn't support the meme, so they ignored what they didn't like. They chose to ignore the Time poll out on the same day as Gallop that showed a dead heat. Its not like this was a minor poll that might be overlooked.

3 polls for Bush and 2 showing a Toss-up didn't support the meme, so they chose to ignore the poll.

Never let the facts get in the way of a good story.

My letter to Howard Fineman re Mary Cheney

dagman

Community (This content is not subject to review by Daily Kos staff prior to publication.)

Thursday October 14, 2004 · 9:02 PM PDT

2004/10/14 · 21:02

Mr. Fineman,

I was disappointed with your column today in Newsweek online. It pains me to think that "the press" is as bigoted as you imply - given their reaction to John Kerry's remark regarding Mary Cheney's daughter. There was NOTHING inappropriate about John Kerry's comment because there is NOTHING wrong with Mary being a lesbian. Secondly, it is appropriate for Kerry to bring it up because Cheney has done so himself within the context of a political setting. I am sure you remember the town hall campaign meeting with Lynn where he discussed it extensively.

This whole episode reminds me of a time in my childhood when my Aunt came to visit us in Florida with her new "friend" Marci. I was told that "we just shouldn't talk about it - it is ok, but a lot of people might not understand". You might think not a lot has changed in 30 years, but you would be wrong. You see, my Aunt was a 45 year old white woman, and Marci was her African American boyfriend. Today, I look back on that whole episode and laugh - as the most conservative justice on the Supreme Court has an Inter-racial marriage. I would not be surprised to see an openly gay man or Lesbian as Chief Justice 30 years from now. I expect to look back on this whole episode and laugh.

Something is up in Central Florida

dagman

Community (This content is not subject to review by Daily Kos staff prior to publication.)

Sunday October 03, 2004 · 3:55 PM PDT

2004/10/03 · 15:55

I live in the Tampa Bay area, and I have never seen more political ads for President than I have seen today. The morning talk shows were heaviest, but the NFL games have also seen a large number of ads. These time slots (especially the NFL) are not the MSNBC cheapies. Both Parties are burning a lot of cash today and this is not sustainable 30 days out. (Oops, here is another Bush ad as I type.) While Kerry has had a bunch of spots, it appears to me that Bush is outspending by at least 2 to 1.

Now that Gallop supports Newsweek, I think we know why the blizzard of ads. Both campaigns internal polling are probably showing a dramatic and unprecedented shift to Kerry. I suspect Rove & co. feel they must stop the bleeding now - while Kerry probably thinks they can continue the mo. I think I am seeing the first signs of desperation from the Repugs.

Its about turnout, stupid!

dagman

Community (This content is not subject to review by Daily Kos staff prior to publication.)

Saturday September 18, 2004 · 12:49 PM PDT

2004/09/18 · 12:49

When I deconstruct the available poll data, I find them to be very consistent - contrary to what the MSM seems to be saying. The internals seem very consistent - Bush getting in the low 90% of Republicans, Kerry getting in the high 80% of Democrats, and independents roughly split - with perhaps a slight lean to Bush.

This means it is all about turnout, and a polls accuracy depends on the validity of the turnout assumptions. The internals of the conflicting polls seem to say the following: 1) If voters turnout in the same ratio's as in 2000, we have a statistical dead heat, 2) if the Republicans can match the Democratic turnout (which would be their best turnout in recent history), Bush has a 3% - 4% lead, and 3) if the Republicans can out draw the Democrats by the amount they are typically outdrawn themselves, Bush is ahead by 8%- 10%. The real story the media should be reporting is "what will voter turnout look like"? I am beginning to wonder if anyone in the media actually performs research and analysis, or do they simply parrot the work of others.

If we are to believe the Gallup and CBS polls, we would need to believe their turnout assumptions. Have they provided any empirical evidence that we have had a dramatic shift in registrations over the last 4 years? For the Gallup ratio's to hold, we are going to have to have one of the following two things happen: 1) 4 million democratic voters have changed to Republican this election cycle, or 2) of all the new voters registered in the past 4 years, there are roughly 8 -9 million more new Republicans than new Democrats. Either of these or some combination is highly unlikely. The real story the press should be focusing on from these polls is the variation of voter registration and turnout assumptions, and which are most likely to be accurate.

Personally, I feel that the Republicans will come close to closing the turnout advantage gap traditionally enjoyed by Democrats, and therefore Bush probably holds a 3 - 4 point edge.

Newer
Older

RSS