Saturday, September 27, 2008

Con Poll Alert

First off, if you are reading this to get a review of what happened Friday night, you’re in the wrong place. I presume that everyone interested enough to read about the debate after the fact, will have had enough interest to watch the debate for themselves, or at least chase down the details from a few of the many places which did offer detailed accounts. Wizbang, for example, has five different threads about what was said, as well as the tone and pace of the event. What I am writing here is a heads-up on what to expect this week, or more accurately, what not to expect.

For me, the most significant observation during last night’s debate came when John McCain said ”I'm afraid Senator Obama does not understand the difference between a tactic and a strategy". Besides the obvious reference at the time to military operations, it was a significant observation as to how Barack Obama misunderstood the strategic purpose of the debate. Chew on that, and I will come back to it in a later piece.

During this week to come, there will be even more frothy expectation of glorious poll results by those addicted to bumper stickers and headlines for their news. But unless the pollsters twiddle with the party-affiliation numbers again, they will not be likely to change much. First off, that’s the historical model. By the way, the National Council on Public Polls says you should pretty much ignore those “instant polls” which come out right after a debate. Why? Here’s why:

Keep in mind that the instant post-debate poll: (1) measures only top-of-head reaction to the debates, (2) does not measure the debate's effect on candidate preference, and (3) applies only to those viewers who were contacted and participated. Remember that who won the debate may have little or no influence on candidate preference.”


Bet no one at Fox, CBS, CNN, or any of the other stations and networks bothered to note that, but as I said in an earlier post, the media is in the business of selling a story, not giving you facts.

Friday, September 26, 2008

The Fox News/Opinion Dynamic Polls: Independence Days

A few readers asked me to look into the Fox News polls, since Fox is commonly thought to lean right (they do not, actually, they simply do not engage in the orgy of Obambast that is commonly seen at other networks), yet their most recent poll has Obama leading McCain by six points. So, I had a look-see into the FN/OD polls.

Before I go into that, I suggest the reader observe some of the other polls released this week. RealClearPolitics shows Obama up by 5 in Rasmussen, up by 6 in Fox, up by 7 in “Hotline/FD Tracking”, down by 2 in Battleground, and tied in Gallup.

So there is a fairly broad range of opinion reflected there, and as I have said before, you should never take the headline for the whole story. I also think it’s best to look at movement in support within the context of a single poll; this aggregation stuff is popular for the 3rd-grade intellects at CNN, but it’s simply not valid for any empirical analysis.

The “Fox News” poll is really a poll performed for Fox news by a company called Opinion Dynamics. Just something to remember, when considering the bias source and degree. Looking at the record, it turns out that there have been only four FN/OD polls which provide useful information for the Obama/McCain comparison; one released September 24, one released September 10, one released August 21, and one released July 24 (which included data from April 30 and June 19 polls).



Looking at these, I found the following lines of support, going from April 30 to June 19 to July 24 to August 21 to September 10 to September 24:

Obama, support from Democrats:
67%, 81%, 75%, 78%, 79%, 84%

Note the sharp jump in June, the fall-off and recovery and how it has steadily increased to its 84% high


Obama, support from Republicans:
13%, 11%, 5%, 8%, 5%, 6%

Note the steady drop of support, although it’s interesting to see a little recovery in the last poll. Too soon to say if that’s significant or an outlier.


Obama, support from Independents:
37%, 30%, 34%, 31%, 31%, 36%

This is the most interesting category, since we see Obama drop support from independents, then recover it, although again the recovery is so recent and singular that it could be an outlier rather than a new trend.


Now, with those in mind, look at FN/OD’s overall support for Obama, with weighting noted for each (Dem-Rep-Ind)

April 30: 43% (44-30-21)
June 19: 45% (42-35-16)
July 24: 41% (42-33-19)
August 21: 42% (42-35-20)
September 10: 42% (41-34-21)
September 24: 45% (41-34-21)

Note how Obama increased his overall numbers for June 19, even with the weighting going against him.

There are two significant demographic movements within the FN/OD poll; the sharp jump in democrat support after April, and the sharp improvement in support from independents in the last poll.

Now, look at McCain the same way:

McCain, support from Republicans:
81%, 81%, 86%, 82%, 88%, 86%

McCain has had pretty steady support from republicans, note though that he’s had trouble getting it to increase since July.


McCain, support from Democrats:
22%, 10%, 10%, 8%, 9%, 5%

Not surprising, but it looks like McCain lost a lot of the Hillary supporters about June.


McCain, support from independents:
47%, 38%, 32%, 30%, 46%, 31%

As the race became more partisan, McCain appears to have lost independents, but what was going on during the first week of September, when he gained 16% from the previous poll? If it’s not an outlier, the September 10 poll may show an opportunity for McCain.


Now, with those in mind, look at FN/OD’s overall support for McCain, with weighting noted for each (Dem-Rep-Ind)

April 30: 46% (44-30-21)
June 19: 41% (42-35-16)
July 24: 40% (42-33-19)
August 21: 39% (42-35-20)
September 10: 45% (41-34-21)
September 24: 39% (41-34-21)


There’s some interesting movement in the FN/OD poll. It includes what may be outliers, but could also be significant trend indicators. What else is interesting, is how unstable support from independents seems to be, a factor which could put a lot of states in play.

For comparison, I went back to recheck Gallup’s demographic group support, and while they posted only between August 24 and September 21, the five weeks cited showed the following supports from independents:

Barack Obama: 29%, 23%, 29%, 27%, 24%
John McCain: 31%, 29%, 28%, 32%, 38%

Those numbers do not shift much, and with one small exception they have all favored McCain. It is impossible on the available information to say whether the Gallup or FN/OD poll is more accurate in gauging the mood of independents, but from what I can see, it would appear that this year the independents are likely to decide a number of battleground states.

Finally, it is worth noting how much of FN/OD’s respondent pool remains undecided. For each poll, look at the portion of undecided voters:

UNDECIDED
April 30: D 11%, R 06%, I 16%, overall 11%
June 19: D 09%, R 08%, I 32%, overall 14%
July 24: D 15%, R 09%, I 34%, overall 19%
August 21: D 14%, R 10%, I 39%, overall 19%
September 10: D 12%, R 07%, I 23%, overall 13%
September 24: D 11%, R 08%, I 33%, overall 16%

There is a double-digit percentage of ‘undecideds’ reported not only overall for every poll, but for both democrats and independents. FN/OD reports that effectively one in three independents have not decided between Obama and McCain.

Thursday, September 25, 2008

Protecting The Economy

The news has been scary lately. I worried about it for a while, but the grey cells kicked in after a bit and reminded me that the source of most of the noise was from people in the business of trying to scare folks. A lot of folks forget that whether it’s broadcast news, cable, or even an internet outlet, they are looking for ratings and the most popular way to do that is do a variation of Chicken Little. The media is – basically – dishonest in its priorities and focus, and always has been. This does not mean that you should not pay attention to the media, but you should keep in mind that they always have an ulterior motive, one which often skews how the present information.

So how big a deal is the present financial mess? Big enough that Democrats and Republicans are seriously talking about spending $700,000,000,000 of your money and mine to face it, with a minimum of narcissistic posturing. Big enough, that all the other markets in the world are affected by what happens here. Big enough that Barack Obama and John McCain actually say pretty much the same thing about what needs to be done. Big enough, that John McCain has suspended his campaign for a while to put his full attention on resolving the crisis. Big enough that even people with decades of experience in Economics are warning that we’d better be serious in how we fix this one.

Unfortunately, a lot of folks, some with advanced degrees, even including some who are in charge of fixing the mess, do not understand how we got here, and therefore they are pretty much clueless about what we do to straighten things out. Of course, I am no Nobel Prize-winner (in either the authentic or Goric sense), so my opinion is subject to a healthy skepticism as well, but I’d like to try to put this in some proper context. What has happened is essentially the combination of three significant causes, each of which would be serious on its own, but which together have caused a crisis of confidence in the United States economy.

The first point is the housing market correction. What basically happened there, was a series of goofs. First goof, the idea that everyone should be able to buy a house, regardless of whether they can pay for it or not. The plain fact is that some folks are lousy with budgeting their money, and some just do not have the fortunate situation that they can afford a big, new, expensive house. In years gone by, such people either saved up until they had enough, worked out how to pay for the house they wanted, or just made do on what they could afford, but about fifteen years ago, suddenly it’s “unfair” to pay attention to how much someone really makes, or even to pay too close attention to how they have handled their money and debts in the past. So, the rules got changed to make it easier for folks to buy a house, who have no business doing so, and to make it easier for folks to get a loan they cannot possibly pay off under ordinary conditions. The question of what would happen to the risk caused by such changes was frankly ignored; everyone just assumed someone else would take care of it. Look, I’m an accountant which means I know numbers but I’m not very tactful or diplomatic sometimes. The new laws were passed by politicians, who are very good at saying things in a way that most people like, but all too often they do not have a clue about the real financial effects of what they are doing. But of course we’re not done yet.
Making it possible for a lot more people to buy houses, meant of course that a lot more people did buy houses, and that old Supply and Demand law kicked in; more people to buy homes meant the price shot up, way way up in some places. That, of course, made it difficult for people with lower incomes to buy houses, including that group which should not have been buying them in the first place, since they could not afford the lower, original price. This time the mortgage people jumped in, offering all sorts of ways to deal with the payments, most notably the Adjustable Rate Mortgage (or ARM), the balloon-payment loan (where you pay a large amount later on the term), and the interest-only loan (where you can pay the principal amount at any time, even waiting until you sell your home). The problem with most of these devices, however, is that they were pretty much all built on the assumption that house prices would always keep rising, and rising faster than the inflation rate, something so absurd that one wonders how so many people bought into it. But of course, we are talking about salesmen and marketing people, whose role in life is so often to take care of their commissions, and let someone else clean up the messes they leave behind. As I said, I am an accountant, and I have a long and unpleasant history of conflicts with marketing people, who not only assume the best-case scenario is the one which will play out, they get angry and abusive if you even mention that there is a worst-case scenario which is bound to happen sooner or later. For years now, accountants have been warning folks that everything has to be paid for, that tricks with money do not change the way things work, that sooner or later the bill comes due, and we’d better be ready for that day.

That day is here.

One of the companies being bailed out is AIG, the giant insurance company. I noticed that earlier this year, AIG got into trouble for what are euphemistically called “irregularities” in accounting. Let’s be clear, irregularity in accounting is not like constipation irregularity, it’s like an irregular heartbeat, serious stuff. Basically, AIG started losing money and tried to hide it. Hide it, I note, without doing anything to find and address the cause. Freaking brilliant. And now the FBI and SEC are looking at more than two dozen companies for possible accounting fraud in connection with sub-prime mortgage lending.

So here’s the second point – John McCain was right when he said the economy’s fundamentals are sound. What happened was a mix of ignoring the accountants, trying to force feel-good policies onto banks without accepting that risk never goes away, and a roller-coaster media which feeds off fear and panic like the monster in Stephen King’s story, “It”. The accountants have been screaming warnings since before Sarbanes-Oxley was enacted (and I have a feeling that Sections 302 and 404 have not been properly enacted at many of these banks, which will trigger certain provisions of Section 1104), and the politicians have only lately begun to recognize that their ‘make everyone happy’ mortgage plan is somewhat less than ideal, but on the whole most businesses will survive the mess with little long-term damage.

So why do we need a bailout? Well, to start with the remedy is not really a “bailout” for most companies. The remedy is actually pretty simple, but it won’t be easy or pleasant:

[] The Housing market will continue to correct – the blunt truth is that houses are overvalued in some parts of the country, and some folks who cannot pay their mortgages will lose their homes to foreclosure. This is always painful, but in the overwhelming majority of foreclosures, the buyer was told about the risks and went ahead anyway. Where possible, loans will be reviewed and adjustments made, ideally to ordinary fixed-rate mortgages, but to do this there needs to be bank participation with willingness to accept reasonable risk. Congress, through Mr. Paulson’s office and requested authority, must decide which high-risk securities to take on itself.

[] The credit policies of banks and lenders must be internally examined for strength of infrastructure. At a minimum, this will cause a number of those companies to significantly delay or restate earnings, in order to clear bad debt. There must be protection against financial failure of the credit system, which includes not only the risks of bankruptcy and default of notes, but also protection of nominal credit access and liquidity instruments. The present fear has reached the point where even money market accounts have been impacted.

[] Despite the desire to look like Robin Hood, any money provisioned for this crisis must be applied to purchase of high-risk commodities and instruments (in a manner similar to the RTC actions of the 1980s), and to provide lines of credit for banks to use in making normal loans and conventional mortgages. The objective is to make money available in the short term which will be repaid as much as possible, not given away and never seen again.

[] Perhaps because it has come up so quickly and appears to be so serious, there has been remarkably and mercifully few attempts to play politics with the crisis. With that said, someone needs to sit the media down and teach them the fundamentals of economics, finance, and accounting. Education about the situation and the reasons for the actions taken would be helpful in preventing panic and advancing an effective resolution of the crisis. Katie Couric in particular should be prohibited from discussing the issue at all.

[] President Bush was correct last night, when he observed that if this deal is not approved there is a risk of panic about the economy as a whole. Public confidence is the most significant factor in addressing the mortgage lender situation. This is the time to restore calm, to work together to handle the issues, and to make clear that the resolution is being built. The answer is present, simple, and time-critical, and we must not waste time, energy, or focus.

UPDATE - I don't know if I believe everything it says, but THIS is a must-read.

Wednesday, September 24, 2008

Obama's Weight Problem

The ABC News/Washington Post released a poll today, which claims that Barack Obama is leading John McCain for President by 9 points. If that is so, just why is Obama so angry and short-tempered these days? A number of possibilities come to mind, but the most likely answer for me is that he knows those numbers are bunk. As Kim Priestap noted, the people running the poll weighted the responses by a 38-28 margin, with 29% listed as independent(question 901). But when they ask whether folks lean more towards the republican or democrat parties, it becomes a 54-38-07 weight (question 904). This is critical to the poll’s meaning, because the 52-43 Obama lead (Question 3) was extrapolated from the “net leaned vote”; in other words 54% of the response used was from democrats, versus 38% from republicans, and only 7% from independents.

The ABC/WP poll also did not break down support by party ideology, depriving the curious a simple way to see what folks really though within internal demographics. They also neglected to break down support by any other demographics, something which further degrades any claim to validity in the poll. I used the internal supports from the Gallup and Pew polls to reverse the calculations and found an interesting revision which shows what I consider a clearer picture, but for here I will just say that the ABC News/Washington Post poll shows what kind of problem Barack Obama really faces; the reality does not match the poll projections. Obama has a serious weight problem, and I don’t mean his cholesterol levels.

Every opinion poll taken this year has been weighted to show more democrats than republicans. There is some historical justification for that, but only to granting the democrats about 3 more percent of any respondent group. The weighting in polls produced by media sponsors, however, has at times been much heavier in democrat proportions, the ABC/WP poll being an egregious example. There is simply no rational basis for presuming that one party would be overwhelming in its representation. As an example, in 1984 Ronald Reagan absolutely crushed Walter Mondale, yet the republicans only represented 35% of the voters in that election. There have been minor fluctuations, but for more than three decades, party affiliation has been largely stable and predictable. There is simply no rational basis for claiming that either major party would have more than 39% of voters, or less than 32%.

Barack Obama is certainly capable of winning this election. He has a ton of money, the media is a team of Obama cheerleaders, and the public mood is desperate. But the polls released to the public do not reflect the actual level of Obama’s or McCain’s support, nor do they show an accurate pattern of support growth over time, or properly relay the time and source of momentum shifts and changes in support. When the polls are reverse-engineered and nominal weighting restored, it becomes clear that Barack Obama is depending on a three-tier plan for the election:

1. Barack Obama must collect at least 90% support from democrats nationwide; or
2. Democrats must make up at least 43% of the voters with present aligned support levels; or
3. Obama must collect at least 55% of independents’ support with present aligned support levels.

At this time, there is no evidence that any of the three conditions exist in fact. This is a serious problem for Obama with just the election a little over a month away.

For John McCain, the following conditions may result in victory, using the Obama plan as a template:

1. John McCain must collect at least 92% support from republicans nationwide; or
2. Republicans must make up at least 37% of the voters with present aligned support levels; or
3. McCain must collect at least 55% of independents’ support with present aligned support levels.

From the internal support reported by the polls, Barack Obama is presently 6 points below the 90% support level, while John McCain is presently 3 points below the 92% support level. Democrats historically represent 38-39% (4 below what Obama needs) of the voter pool, while republicans represent between 35-36% (1 below what McCain needs) of the voter pool. And depending on the poll, Obama enjoys between 42 and 48 percent support among independents (7 to 13 below his need), while McCain enjoys between 45 and 56 percent support among independents (he may need as much as 10 percent, or he may have what he needs right now).

There are several significant events between now and the election. The debates between Obama and McCain are yet to come, and the fallout from the financial crisis could well affect voter opinion. But right now, Obama has a weight problem, in that the election proportion of democrats in the actual election is not going to nearly match what the media is pretending it will be. While some believe the spin on these polls is meant to dismay McCain supporters and encourage Obama supporters, in the actual case this distortion could well come back and hurt Obama’s campaign.

Monday, September 22, 2008

There Is No Alternate Universe

People throw out a lot of strange ideas at times. Some of that is because people can believe some very silly things, and some of that I blame on television and comic books. But it really gets strange when we see fantasy and illusion play out in politics, in the guise of fact. CBS tried to smear President Bush with documents they admitted were fakes, the claimed inexperience of a Vice-Presidential candidate is loudly harangued, while the weaker resume of the other major party’s Presidential nominee is ignored, and biased opinion polls are touted as objective sources of news. Truly Alice-in-Wonderland stuff, except that this is the real world, and we dare not risk trusting the Mad Hatter.

Opinion polling is fun in many ways, useful in certain aspects, but in the end should not be trusted as a guide or counselor for course decisions any more than one might trust a Ouija board. The bias is often missed, even though it is obvious once you know where to look. I have shown before that every poll is biased to some degree, and I have repeatedly made clear what you should expect from a valid poll:

• public access to internal data
• a consistent, publicly reported methodology
• weighting according to Census norms or reasonable objective standard
• archive data available for comparison to current polls

Alas, there are areas where even the best of opinion polls cannot meet the standard of true objectivity. The most common stumbling block comes when a poll weights its political party affiliation. Polling groups understand that a truly random system of contacting poll respondents will produce results which are, to some degree, at odds with the true opinion of the public as a whole. To correct for this, respondents are asked certain questions to determine key demographic data, and their responses are categorized according to those demographic keys and the overall results weighted so that the response pool model is in line with demographic norms. The demographic standard used is almost always based on the most recent US Census data, for reasons that this data is considered the most unbiased and reliable demographic data available. So, Census data is used for gender, race, age, education, employment, and economic strata norm determination. This makes a lot of sense, and I applaud the pollsters for that standard and diligence in its application. The same polls who are so careful to avoid bias in most demographic norms, however, get completely squirrelly when it comes to party identification.

Political party identification is a stronger factor in candidate support than any other cited demographic category. It is no shock, after all, that democrats overwhelmingly support the democrats’ nominee, while republicans overwhelmingly support the republicans’ nominee. The problem comes in, when the poll weights the response pool to match a desired party identification standard. I noticed four years ago when I looked at the opinion polls, that the polling groups changed their party identification weights, some every week! When I contacted the polls about this practice, most just ignored me, but a few did respond. I found their explanations troublesome, however. Pew, Gallup, and Rasmussen, for example, choose their party weighting by examining the self-reported party affiliation of respondents from a prior period, usually a month (the ones who change each week generally use a rolling average of respondents from the period reviewed). Unfortunately, the reasoning behind such weighting is directly contrary to the whole purpose of weighting in the first place. To see what I mean, let’s apply that logic to other demographic groups.

South Texas was hit hard by Hurricane Ike, and it is very unlikely that any polling group had much success calling anyone down here in the past week; those who had phone service restored had a lot to do, and no time for answering polls. But even as phone and power service was restored, this was done according to critical needs (like hospitals) and a plan to get the most service restored to the most people as quickly as possible, which in practical terms means that urban areas and recent construction would get priority. This is because of the concentration of population, and the relative ease of repairing lines which would be less likely to have serious damage to the lines and transformers (older neighborhoods tend to have trees and other growth obstructing lines, so that in a major storm older neighborhoods are much more likely to have substantial damage to the transmission lines). Consequently, for the next month, an overwhelming majority of poll respondents in South and East Texas will be urban areas and new neighborhoods, which would heavily skew the demographics of the polling, if those respondents were considered “normal” for the area’s demographics. Or consider another example, where more mid-week polls are taken. People who work or who are taking care of children would be less likely to be available for polls, which would skew the apparent demographics towards the youngest and oldest voters, and those whose lifestyles matched the tactics used to reach respondents, like visiting malls or urban centers. Polling groups know these areas are heavily skewed in their demographic types, with heavy oversampling of certain groups, and so the weighting is adjusted to match Census norms, precisely because anything else would invalidate the poll results.

Those are just two obvious examples of why using poll results for one period to establish the demographic weights for another period would be clearly invalid; it’s simply circular reasoning and any errors (and there are always errors to some degree) would be greatly magnified rather than corrected. What’s worse, the only validity for any poll whatsoever is movement within a poll handled by consistent and transparent methods – if you shift party identification weights, you invalidate all conclusions. Anyone who has annoyed their lab professor understands that controls exist for a reason, and fiddling with the weights between polls is simply unscientific.

Some of the pollsters I have spoken with, argue that they shift their weights because no useful and objective source exists for party identification weighting. That, however, is not true. The exit polling from prior elections is a very valid source. It comes not from “adults”, “registered voters”, or even “likely voters”, but from people who actually voted. And what’s more, we can look at the last few elections to see if there is any substantive change. I looked at national elections for the past ten years, and discovered an interesting pattern:

In 2006, 38% of the voters were democrats, 36% were republicans, and 26% were independents;
In 2004, 37% were democrats, 37% were republicans, and 26% were independents;
In 2002, 39% were democrats, 38% were republicans, and 23% were independents;
In 2000, 39% were democrats, 35% were republicans, and 27% were independents; and
In 1998, 39% were democrats, 33% were republicans, and 28% were independents.

That sure looks like a consistent pattern to me. On average for the past ten years, democrats have averaged 38.4%, republicans 35.8%, and independents 26.0%. If we kick out top and bottom outliers, it becomes 38.7% democrats, 36.0% republicans, and 26.3% independents. Those numbers have a solid empirical history and a thoroughly objective source behind them, yet the polling groups do not use them. You might well wonder why.

Polling groups exist for one of three purposes. They either serve the needs of a client (private polling companies), they do academic research (which serves the faculty running the polls), or they are done for public release (including those by colleges like Marist and Quinnipiac). Each group has a very specific, and different, reason for existing. You may want to consider that both Barack Obama and John McCain have hired private polling firms, and consider why they each spend that money, as indeed every major political candidate does. And if political allegiance is so flighty, then why do so few of us know anyone who has supported one party, then another, in the same election? To read the reports from the polling groups, there are a lot of folks who loved Obama, then McCain, then Obama again, or vice versa, yet I have not met even one such person. Instead, I have seen democrats who stay democrats, I have seen republicans who remain republicans, and I have seen folks making up their mind as the campaign progresses. There are times where you might see a new jump in support, as previously undecided voters choose someone to support, but that happens for a reason – the “bounce” that supposedly comes just because a party has a convention, well, that’s pretty bogus if you think it through. The democrats may have been more excited about Obama after the democratic convention, but I do not believe for a minute that it made many new democrats. I am quite sure that the GOP convention excited republicans, but again I have to say I do not recall hearing about a bunch of new republicans after the convention. So, when the polls increase democrats’ weighting after the one convention and increase the republicans’ weighting after the other, it’s frankly dishonest because neither party saw much change in affiliation. What’s more, does anyone really think that there were fewer republicans after the DNC, or fewer democrats after the RNC?

Please.

Party affiliation is not some ephemeral quality, which winks in or out of existence because of a speech by one candidate, the headlines of a single day, or the personal desire of a polling group to have a hot story for the newspapers and TV stations. Party affiliation is something developed over a period of time, and while it does change over time, it does so gradually, taking into account the evidence of history as well as events of the moment. There were tens of millions of republicans after Bob Dole’s loss in 1996, just as there were tens of millions of democrats after Walter Mondale’s thrashing by Reagan in 1984. Support for a candidate is a thing apart from party identification. When it does change, it is spurred by a significant event of meaning with regard to the values and ideals of the party This is not merely a belief, but the evidence of American electoral history and the employment of common sense.