Subscribe via RSS Feed

Category: Dave Brockington

Classic George Will. Seriously.

[ 22 ] November 6, 2012 |

Looking through my 1976 copy of The Changing American Voter today, what should I find inside?  This vintage 1976 clipping of a George Will column, warning us sternly about the dangers of allowing voter registration by . . . mail.  Because it would mean a “substantial increase in bureaucracy, and a substantial increase in the opportunity for fraudulent voting.”




A Rambling Prediction of Sorts

[ 12 ] November 6, 2012 |

303-235.  Figure Obama’s PV around 50.5%.  I expect him to have slightly over a 2% win in the PV.  No clue nor care about his margin in MA.  Obama loses IN, NE-2, NC, and possibly / probably FL from 2008.

Democrats keep the Senate with 53 or 54.  We all know the House is out of the question, but I’d anticipate a fuzzily modest Democrat pick up, 5-10 seats.

I woke here to excellent numbers nationally, and specifically, movement towards Obama in FL (and a PPP poll showing a tie in NC).  However, yesterday’s (5 November in the US) FL numbers were from PPP, which has had a slight D lean all cycle.  The over night numbers in FL are from Gravis, Angus Reid, and IPSOS, the latter two having strong R leans.  Notable, those two suggest a swing towards Obama from +3% to +5% based on their own previous releases.  I know Silver models trend into his model, which is why he’s showing Obama as a very, very slight favorite in FL.  I’m going to discount that a bit, especially in light of the vote suppression efforts we’ve been hearing out of FL the past couple of days.

FL won’t be called for a while.  If it is called early in either direction, that’s a clear signal.  Plus, the recount procedure in OH is arduous.  If OH is within the “margin of litigation”, it could be a nightmare.  I loved reading that line in the linked NYT article while seated in the BBC Radio Devon studio this morning waiting to go on, and was able to use it.  Excellent line.

What if this is all wrong?  As most here read Wang and  Silver, we know that the polling, specifically the state level polling but latterly the national polling as well, has to be systematically and comprehensively biased in Obama’s favor.  There are reasons why this could be the case.  From the right, Ted Frank has a comprehensive, wishful thinking list of all the reasons the current polling data might be biased, and systematically so in one of the two possible directions, but to Frank’s credit, he’s looking at evidence, not relying on Morris-esque faith.

Regardless, something other than an Obama victory in this election would require a systematic polling error of significant magnitude.  Electoral Vote has around 900 state level polls in its database from this cycle last I checked a few days ago.  It would represent a large pile of error in one direction for this election to be incorrectly called at this point.

But it has happened before, UK 1992, when the polls underestimated the Conservative vote by a not middling 9%.

I catch the midnight (GMT) train to London tonight, as I’m flying to the west coast tomorrow to present a paper this weekend.  I’ll be looking at NH first — it’s a small state that will compile its numbers quickly.  That should be a strong signal as to how valid the overall polling numbers have been this year.

UPDATE: as I was writing this, Wang at Princeton reports that his modal estimation has Obama with 332.

Elected Police Commissioners — Why?

[ 28 ] October 31, 2012 |

In filling out my ballot yesterday and posting it off to Oregon, one of the decisions I had to make was for Clackamas County Sheriff.  This decision was nearly as easy as President, seeing as how the incumbent sheriff is running unopposed.

On November 15th, English and Welsh voters get their first chance to vote for such a position, which they’re calling the Police and Crime Commissioner.  This has generated criticism from most points on the political spectrum, and both regarding the fundamental need to its botched implementation.  The criticisms include a violation of the separation of powers, the politicisation of the police, that it will be a low information election both in terms of the office and the date the government chose to hold it, and that turnout will be perilously low to the point where the election itself will be illegitimate.

The Electoral Reform Society are particularly critical of the implementation and, according to their analysis, 18.5% predicted turnout.  I agree with much of what they have to say.  However, they’re rather fuzzy on how they arrived at the 18.5% estimate, and even then, judging the legitimacy of an election based on turnout is an ultimately arbitrary game.

I’ve been doing the usual bit of local media for the BBC during an American election cycle.  This year has brought something new — a seven day run on BBC Radio Devon called ‘pause for thought’.  Instead of the normal back and forth of an interview, it’s a set piece of about two minutes where I have clear air time.  The hardest part of it has been waking every morning to be in the studio around 6am.  Otherwise, it’s like writing blog posts, but pitched to an entirely different sort of audience, and read live on air.

I addressed a couple of the critiques mentioned above in this morning’s piece.  Specifically, while it’s a low information election, the partisan label of the candidates (those not running as independents) does serve as a voting cue and communicates some information.  Furthermore, local media have a role to play.  Radio Devon is trying to raise awareness on the Devon & Cornwall Constabulary election, and pieces like this in the Seattle Times (on the current election for King County Sheriff) are not uncommon in US media.

Second, using turnout as a metric for legitimacy is never going to fly due to its inherent arbitrary nature.  I get the normative problem: public policy in a democracy should represent the general will of the population (however measured and defined), but ultimately, and rationally, instead represents those who vote.  When the composition of the electorate is systematically different from that of the general population, policy will likewise deviate.

However, again, where to draw the line and claim an election illegitimate?  In 2005, Tony Blair was re-elected with 35.5% of the vote on a turnout of around 62%, resulting in a durable parliamentary majority that governed for five years.  Yet, when examined closer, 78% of the population either explicitly voted for someone else or failed to participate.  Labour’s re-election in 2005 was predicated on the support of 22% of the population.

Furthermore, while the position here is laudable for devolving accountability and some policy to a more local setting, it remains largely bereft of policy responsibility, so I’m not sure how far one can take the turnout critique.  Indeed, if the Seattle Times piece linked above is any measure, the election for King County Sheriff is about administrative and personality qualities, not policy.

I’m not terribly sure how to address the critique that these offices somehow violate a separation of powers.  The entire concept is, at best, muddy in the British constitution.  This isn’t a violation in the US context.  We might legitimately argue whether or not these positions should be selected by vote or on pure merit (as we might likewise debate voting for judicial positions), but I don’t see this as a separation of powers issue, much to the chagrin of an ex-student of mine.

This leaves the politicisation question.  On one level, this is a trade-off — adding the partisan label, useful as a cue for voters, by definition politicises the office.  However, claiming non partisan offices to be de-politicised is laughable.  Poring over the list of non partisan offices in Clackamas County and the State of Oregon yesterday, one cue I used to make decisions were endorsements.  Every candidate for “non partisan” positions have endorsements that cluster in readily identifiable, partisan clusters.  While removing an office from the electoral purview would seem to aid in its de-politicisation, that itself is superficial.

The theme of my piece this morning was that expanding democracy is a normative good, especially in the UK.  While extending democracy in England and Wales to the office of Police and Crime commissioners is open to criticism, the most damning are about implementation.  While spreading democracy is good, a much better place to start expanding democracy might have been giving local government real policy power (additionally, with few exceptions, there is virtually no relationship between local government as its understood in the UK and policing).  At present, only about 25% of the local council budget comes from local taxation, and the council’s ability to vary this is tightly restricted by central government.  These positions don’t really matter much at all, and, unsurprisingly, turnout to local elections is unsurprisingly modest.

Notes on a Couple of Polls: Pew and RAND

[ 74 ] October 9, 2012 |

First, let’s discuss the new Pew release, and attempt to do so without freaking out.  It presents grim reading for at least two reasons: the four point Romney lead, and that Pew has had a consistent Democratic lean, relative to the aggregate averages, this entire cycle.  Indeed, Silver notes

The Pew poll, however, may well be the single best polling result that Mr. Romney has seen all year. It comes from a strong polling firm, and had a reasonably large sample size. Just as important is the trendline. Pew’s polls have been Democratic-leaning relative to the consensus this year; its last poll, for instance, had Mr. Obama 8 points ahead among likely voters. So this represents a very sharp reversal.

As it’s important to not fall into the trap of focusing on confirmatory evidence to the exclusion of equally valid contrary evidence, it’s likewise not healthy to overly concentrate on one negative national poll to the exclusion of the weight of extant evidence.  In other words, I didn’t start doing back flips following the September Pew release showing an eight point Obama lead, and I’m not marching to the liquor store to buy out their stock of whisky because of this one.

There are legitimate, and less legitimate reasons to be weary of this one poll; John Sides does a solid job of pointing out some examples of both.  One reason attracting attention is the underlying partisan identification of the sample.  The most recent, October Pew sample had 31% Democrats to 36% Republicans.  Therefore, at best, Republicans were over-sampled accidentally; at worse, purposefully, or so goes the narrative.

A comparison to the September (12-16) Pew release is illustrative here.  The underlying partisan composition of that sample was 39% D to 29% R.  Does this mean that there has been a swing of 15 points in the underlying partisan composition of the American electorate from a ten point Democratic lead to a five point Republican lead in three weeks?  Of course not.  There are (at least) two dynamics at work here.  First, partisan identification is not a static attribute.  It’s a dynamic attitudinal measure that can be affected by context, and following the debate, it’s possible to conceive that claiming a Republican allegiance is marginally more attractive than it was prior to the debate.  Second, polls vary, even from the same polling houses, for random reasons.  Again, Sides: “Poll results vary for random reasons—that is, because of sampling error.”  Indeed, considering both the extreme (in comparison to other polls) top lines, and equally extreme shift in both the top line and underlying partisan composition of the sample, I’d suspect that both samples were unlucky.  In other words, while October is possibly an over-statement of Romney’s support in October, likewise September was possibly an over statement of Obama’s support.

There’s a second comparison to make: the LV N of the September sample was 2,192.  October, 1,112.  This does not make the former automatically better than the latter, but assuming everything else in their sampling methodology remained the same, the error bands in October are larger.  Meaning, we can be less confident in the validity of the four point lead in October than we could in the eight point lead for Obama in September.  The MoE of October for the LV figures is +/- 3.4%, meaning that there is a 95% chance that the “true” figure for Romney’s support lies within a band of 45.6% to 52.4%.  While this simplifies matters, and I feel that MoE can be overstated, it is illustrative of the random error that naturally exists even in a perfectly drawn probability sample.

An intriguing alternative to the Pew release is the ongoing panel survey at RAND.  Wang discussed this survey yesterday, suggesting that the data show that the Romney debate bounce resulted from inspiring the base and not from converting undecideds or siphoning directly off of Obama’s support.  This is notable because of empirical evidence that suggests viewers largely watch the debate to cheer their team and have their own underlying decision confirmed, yet is surprising because Romney gave a very non-base performance highlighted by a rapid shift to the center.

Before looking at this a little closer, there are some strengths, and weaknesses, to the RAND methodology to discuss.  The notable strength of this survey is that it is drawn on a panel, not a cross section.  They’re “interviewing” the same 3500 respondents every week (500 each day so the daily result is a composite rolling average).  In terms of research design, this puts us in a much stronger position to make causal inferences.  If we consider the debate performance a temporal “treatment”, we’re on safer ground suggesting that “the debate caused X” than we would be comparing a bunch of cross sections.

There are several weaknesses specific to using a panel, however, but in my estimation these weaknesses pale in comparison to the inherent strength of the design.  First, any initial errors in drawing the sample are locked in.  This can be, and is, partially offset through weighting, but it is something to consider.  Second, there’s the issue of sample attrition.  Finally, something I just considered today, it’s possible that respondents might feel internalised pressure to stick with their initial decisions.  An example might be a hypothetical respondent consistently answering with “Obama” from early September to the present, but internally is starting to lean Romney, yet doesn’t want to publicly admit this, as he or she does not want to appear fickle or superficial or inconsistent (or choose any of a number of possible awkward adjectives here).

There’s also a minor weakness, in my assessment, of their LV model.  This said, there are at minimum minor weaknesses in any LV model, so this should not be interpreted as a critique of this instrument in isolation.  One of the three items this survey asks is “What is the percent chance that you will vote in the Presidential election?”.  This is self reported, which suffers from questions of both validity and reliability.  Social desirability gets to the theoretical heart of the matter in both.  People will tend to over-estimate the chances that they will vote, because it’s the socially desirable thing to do, and before an election a respondent will feel this both externally and internally.  The former, they’ll feel that whoever is reading and coding their responses will look down upon them for claiming only a 50% chance of voting, so will inflate the number.  The latter, sure, they believe that they’re going to vote, it’s the right thing to do after all.

Thus, these figures are going to be inflated.  However, in terms of reliability, social desirability effects different groups of people differently, in systematically predictable ways.  Individually, those of socio-economic categories more predisposed to participate to begin with will be more influenced by their own perceptions of the socially desirable response (both internally and externally).  Basically, Romney supporters should be more susceptible to this phenomenon than Obama supporters.  This should present a silver lining to most of us here: we should discount the self reported figures for Romney supporters more than for Obama supporters.  Likewise, electoral context with higher levels of turnout over time will likewise be more susceptible to social desirability bias.

That said, they weight for this as well based on national participation figures from 2008, which might explain why the current estimates show only a 1.6% gap between Romney and Obama supporters probability of voting.  I’d expect this estimate to be larger.

Below are the rather precise daily estimates from RAND since the October 3 debate.  Inferring from these numbers, Romney’s bounce was in the neighborhood of 1.15% in his own support, 2% in the spread, and this has rather quickly stabilized.  Of course, this is out of line with what most are reporting.  That said, the October 8 rolling average only represents the updating of 2500 out of the 3500 overall sample, so it’s possible that Obama’s numbers could still deteriorate further.  Additionally, the debate doesn’t appear to have had a big impact on the number of undecided respondents: on October 4 undecideds were 6.15%, yesterday 5.69%.

Obama Romney Obama +
October 4 49.87 43.98 5.89
October 5 49.18 44.75 4.43
October 6 48.95 45.10 3.85
October 7 48.93 45.19 3.74
October 8 49.07 45.24 3.83


Examining these daily numbers present a picture different from the Pew estimates, and indeed from most of what we’ve been seeing in the past 72 hours.  RAND estimate a 3.83% Obama lead.  RCP’s aggregator: Obama 0.5%.  Pollster’s aggregator: Obama 2%.  Gallup’s daily tracking: Obama 5% (RV, it should be noted).  Rasmussen today: a tie.  Sam Wang’s meta-margin is Obama +2.3%.  This tells us two things: Pew’s an outlier, and RAND is at the high end the other way.

I’m Not Concerned

[ 229 ] October 4, 2012 |

In keeping with the spirit of LGM on this Fourth day of October 2012, I have a few observations about yesterday’s debacle debate.

As Scott suggested moments ago, and as Silver wrote yesterday, challengers typically do well in the first debate.  However, in this case, while it might make it more of a ballgame, it should not make a real difference.  Silver offers a lot of sound evidence-based points, as he does, and three are especially pertinent here:

The challenger’s gains have come mainly from undecided voters rather than from the incumbent himself.

No candidate who trailed by as much Romney did heading into the first debate went on to win the election.

There has not been any tendency for the challenger to gain over the remaining weeks of the election.

The average gain by a challenger after the first debate, between 1976 and 2008,  has been 2.3% according to Silver.  If we narrow the historical range to between 1996 and 2008, the average shift has been 1.4%.  Furthermore, if current polling data are to be believed, there are relatively fewer undecideds remaining at this point in the campaign than in previous years.  Hence, while the CBS poll of undecided voters has Romney “winning” 46%, Obama 22%, and 32% reckoning it was a draw, I’m relatively sanguine.  Assuming a) that this event was the crucial decision rule for this population, and b) the 32% break 50/50, 62% of undecideds were swayed to Romney, while 38% to Obama.  Even rounding up to a conservative 2:1 break between Romney and Obama amongst the undecideds, it’s not enough to shift the outcome.

The current RCP national average has Obama 49.1%, Romney 46%.  That’s a 3.1% Obama lead (which is one of the more conservative esitmates available).  This average suggests 4.9% remain uncedided.  If they break 2 to 1 for Romney, (3.3% to Romney, 1.6% to Obama) that leaves a final result of Obama 50.7% to Romney 49.3%.  Romney narrows his gap from 3.1% overall to 1.4%: a shift of 1.7%.  This is within the averages above: an average shift of 1.4% from 1996 to 2008, to average of 2.3% from 1976-2008.

The third point is also telling: this was Romney’s last chance.  Yes, Obama’s conservative strategy basically gave him this game, but at the same time, Obama did not screw up beyond his passivity.  While ideally the Democrats use Romney’s ill-conceived line about money and hearts against him early and often, I don’t think that the Romney campaign took anything away from this that they can use against Obama.

But let’s get even more conservative about this.  The current (October 4) map at ElectoralVote has four “battleground” states where Obama has a 1% to 5% lead (again based on the most recent poll, so the usual caveats about any single poll apply).  These are CO (9), IA (6), VA (13), and FL (29).  Let’s assume Obama retains the others where his lead is > 5%, but loses these four on the basis of last night.  This assumes a shift of at least 2% in FL and VA, over 3% in CO, and over 4% in IA.  Obama still wins the EC 275-263.  The current EC map at RCP, not including toss-up states, has Obama on 269 votes in the bag (toss ups include CO, IA, VA and FL I list above, as well as NV, MO, and NC).  We can safely assume that Obama will not win MO, but under a worst case scenario, he’ll still pull out one of the remaining states.

Hence, I don’t share the breathless overreaction noted by Scott’s insta-outsourcing of  Kuttner last night: “Tonight’s debate did serious damage. There are two more to come. If Obama does not pick up his game, he is a one-term president.”  Dude.  Chill.  A more realistic, sober appraisal is offered by Sargent:

What remains to be seen is whether he conveyed convincingly and affirmatively enough that his plan would engineer a faster rebound and would strengthen the middle class, at a time when voters are already concluding the groundwork has been laid for recovery. That’s what he really needed to do to alter the basic trajectory of a contest that Obama is currently winning. It was unquestionably a good night for Romney, but it’s unclear whether it will produce the big change he needed.

Frankly, I’m far more concerned that my wife just took three of five ‘words with friends’ games from me (following a ten game undefeated run) than I am about the debate last night.  The difference is, as I start the next five games, I have a better chance of reversing this tally than Romney does of winning the election based on the debates.

Determinants of State Level Support for the ACA, Enhanced Edition

[ 9 ] October 3, 2012 |

Three months ago I mashed together a little data set in order to examine variance in state level public support for the ACA, and wrote about it here.  This thing has magically morphed into a conference paper, so I’ve been kicking it around when time allows.  As I present this as of yet unwritten gem in a little over a month, it’s time to lock the model in and move on to the writing bit.

I’ll shamelessly quote from July’s post on variable description:

The two key variables in the data set are state level support for the ACA and the percentage uninsured in the state.  My source for the former is from a paper written by Richard Gonzales, a doctoral candidate at Harvard’s Department of Health Care Policy, which was discussed here at The Incidental Economist.  Gonzales estimates state level support from national Gallup data over a six month period (September 2009 to March 2010).  As this is an estimate, it does introduce an additional layer of uncertainty into the model, but it’s the best data I could find.  The point estimates and error bands appear sound in terms of face validity; the highest levels of support are found in New York, Hawaii, and Vermont, with the lowest in Oklahoma and Wyoming.  The public opinion data are old, but Monkey Cage suggests (and we all pretty much agree on) public opinion on the ACA has been relatively stable.  Percentage uninsured by state level is courtesy of Gallup, which are available here.

A bi-variate relationship was the exact opposite of what we would expect: as percentage uninsured increased, support for the ACA declined.  Believing that wasn’t an accurate story, I added percentage point change in uninsured from 2008-10, hypothesizing that a greater increase in uninsured is associated with an increase in support.  I also added both the PVI of the state and the percentage voting for Obama in 2008.  As the latter two are highly correlated (.98), I ran discrete models with each in turn.  Discussion on these variables can be found in July’s post linked at the top.

Since then, I sourced and included a bunch of variables that I had hoped would be revelatory, including per-capita GDP, average unemployment from 2008-10, the rate of Medicaid recipients expressed as a percentage, the poverty rate (again, %), and the percentages black and Latino.  Furthermore, on September 17 I added the ‘average poll rating’ from Silver’s 538 at the NYT.  Again, that is highly correlated with both the PVI and the percentage Obama vote in 2008, requiring three separate models.

Some brief explanation first: Partisan X is the respective partisan lean variable for each of the three models in turn (1. 538 polling average; 2. PVI; 3. Obama % in 2008).  Significance levels are sparsely needed; for the uninitiated, three stars represents a significance level < .001, + represents a significance level at or below .10.  Effectively what this means in these cases is that the odds of finding that estimate by random chance (as opposed to observing a real relationship) is less than one in a thousand (***), or one in ten (+).  One might argue that as I have the universe of data at hand, as these are not probability samples, significance levels are largely meaningless.  I’d argue this very point if it didn’t go off on some irrelevant tangent, so we’ll stick with convention for the time being.

Apologies for the table formatting; my staff of interns have the day off today.

M1 (538 poll avg) M2 (PVI) M3 (Dem % 2008)
Partisan X .73*** .82*** .77***
GDP per capita -0.02 -0.02 -0.001
Average uninsured 08-10 -0.28+ -0.14 -0.18
%change uninsured 08-10 -0.29 -0.19 -0.38
Average unemp. 08-10 0.41+ .17 .34+
Medicaid Rate (%) -1.59 -2.77 2.53
Poverty Rate (%) .04 -0.02 .14
% Black .22*** .25*** .25***
% Latino .08 .09+ .05
R2 .89 .90 .93


So, what do we have here?  The model fit is strong across the board: between 89 and 93 percent of the variance in public opinion support for the ACA is explained.  The model including the % vote for Obama in 2008 is the strongest, but it’s such a marginal difference I don’t want to read too much into this.

Only two variables are consistently significant: the partisan measure, and percentage black.  Before discussing the others, a further comment needs to be made regarding conventional significance levels.  The industry standard, however arbitrary, is .05; meaning, if there’s a 5% or less chance that the observed relationship is a product of random chance, then we accept the relationship as being significant.  It is permissible to stretch this to .10 but only in cases where we have an a priori theoretical reason to not only expect a relationship, but also to expect the direction of the relationship.  While I have average uninsured in bold and with a significance marker, I have a difficult time imagining a theoretical expectation that would predict the direction of this estimate: higher levels of uninsured people being associated with lower levels of support for the ACA, so I’m going to overlook this one for now (along with overlooking the bivariate figure I posted in July).  As an aside, it is possible that predominantly red states have atypically huge pockets of poverty and uninsured citizens, but the model should account for this through the other variables.  So what we have remaining is higher levels of unemployment being related with higher levels of support for the ACA, as well as a relatively stable estimate for Latino support, yet only significant in one model, both of which make sense.

I tried several methods of isolating the swing states (of which I included NV, NM, CO, OH, WI, IA, NH, VA, NC and FL).  I ran models with those states only, which was a universal dud, but then an N of nine isn’t going to afford any real statistical power.  I also ran interactive terms in the model isolating whether or not these states as a group had a unique estimate for uninsured %, unemployment, Medicaid and poverty rate, and generated nothing.

What lessons can we take away from this little exercise?  I have several ideas.  Politically, it’s evident why health care has not been a major issue in the campaign.  First, obviously, Romney would find it difficult to run against something that he supported in Massachusetts (but, of course, this is Mitt Romney we’re discussing.  He can be anything to anybody at any time).  Second, less obvious, it’s not going to make a difference.  Romney states are largely predisposed to oppose, while Obama states are in favor.  The ten swing states identified above have a much tighter distribution (45%-53%) than the nation writ large (32%-63%), but adopting a strident anti-ACA strategy has risks.  Iowa and Florida show 45% support in this dataset, so these would be good targets to attack the ACA, but you risk losing support in Nevada and New Mexico (both 53%), and the other six (all at 49% or 50% in favor).  It appears to me that there’s more to lose than there is to gain by going after the ACA in the swing states.

Academically, this paper might have something to add to the literature, which is pretty barren on public opinion and health care reform.  Tesler (2011) finds that race is a factor in determining support; given that President Obama is regarded as an African American, support for the ACA is mediated by racial attitudes amongst the white population. Relying on data no more current than 2004, Gelman, Lee, and Ghitza (2010) find opposition to be concentrated among those with higher incomes, and those over the age of 65.

That aside, I’m more convinced now than I was in July that judgment of the ACA as policy is mediated by both a pre-existing partisan lens as well as by race.  Obama was never going to win Republicans over to it, regardless of how hard he tried and how much he gave away at the beginning, because they were never going to support it anyway.  Electorally, it can be used to rally the base, but that’s it.

Of course, one must also suspend any concern that they might have with the ecological fallacy to make some of the inferences I make above . . .

Election Prediction

[ 27 ] October 3, 2012 |

This is the time of the season where many if not most of us pore over the various state level election tracking sites available.  Electoral Vote was a big one in 2004, and remains my favorite for daily poll releases; 538 in 2008 and again now under the New York Times in 2012.  One I wasn’t familiar with, inexplicably, until yesterday is Princeton Election Consortium.  I’m assessing the distinctiveness of this one when I have time, which means I haven’t accomplished more than a superficial perusal.  Their current analysis is similar to Silver’s: there appears to be a slight erosion in support for Obama in national polls that is not replicated in state level polls.  The model itself seems far less sophisticated in terms of variables, but this is not necessarily a bad thing in a purely predictive exercise.

It’s worth a look.

h/t Jeffrey Dudas

UPDATE: link fixed.  I hope.

George Galloway and Libel Law in the US and UK

[ 56 ] October 2, 2012 |

George Galloway, former Labour (he was kicked out of the party in 2003) current “Respect Party” MP for Bradford West, and global lefty gadfly, is suing the National Union of Students for libel.  Yes, he fights what he and others consider to be “the good fight”, but his fight is one that I disagree with as often as I agree, his methods can be interpreted as inflammatory, and arguably he is a discredit to the progressive cause at least as often as he helps.

I have three thoughts on this.  First, suing the umbrella organisation for British university students is short sighted at best.  These people should be a source of support for his various causes for a variety of reasons (e.g. having come of university age under the Tory – Lib Dem coalition, English and Welsh students paying £9000 per year simply to hear people like me talk).  In one move, he is alienating an entire support base.  But then, alienating his support, or the population in general, has not been a barrier to action for Galloway.

At issue is the NUS has banned him for being a “rape denier”, a description that Galloway finds as a defamatory characterisation of his widely known critique that the Swedish allegations against Julian Assange “don’t constitute rape” and were at worst “bad sexual etiquette.”  It’s not only the NUS who have a problem with this bizarre defense of Assange.  His party leader denounced them (and then was sacked or stepped down, I don’t recall), he lost his gig as a columnist for a Scottish publication devoted to politics, and Rape Crisis found them to be “offensive and deeply concerning”.

Taking the NUS to court over libel is bonkers, especially after his office desperately attempted to convince the NUS to not take the step of banning him.  Defending Assange on rape allegations when all the facts of the two cases are not in the public domain is reactionary, short sighted, and ludicrous.  Taking the national body that represents seven million British university students to court is not politically adept, but again ludicrous.

My second thought concerns a comparison of libel law in both the United States and the United Kingdom.  I preface this with the obvious: I’m not a lawyer, although I’ve had the occasion to employ several on two continents in the recent past, the present, and into the future.  The key difference between the two is the default status of the allegedly libellous statement, and the onus of argument.  In the United Kingdom (technically here England and Wales), the statement is regarded as false unless those making the statement can prove it to be true.  In this case, the NUS as defendant would have to prove that George Galloway is a “rape denier”.

In England and Wales, a private individual need only establish negligence on the part of the defendant to be rewarded compensatory damages.  Galloway, one of the more unprivate of individuals, needs to show that the defendant knew that the statement was false, resulting from actual malice.  In order to be rewarded punitive damages, both private and public individuals need to demonstrate actual malice.

The United States is far more forgiving on defamation law, for which we in part have current interpretation of the First Amendment to thank.  The burden of proof is (largely) on the plaintiff, and both constitutional and state level statutory law allow for many “outs” for a defendant in a libel case.  By my understanding, this dates back to New York v Sullivan (1964).  To use an example, the infamous parody of Jerry Falwell published in Hustler magazine is not protected in the United Kingdom, and I’d guess damages would have been rewarded to the point where said publication ceased to exist.  In the United States, of course, this resulted in Hustler Magazine v Falwell (1988), which protects parody and the publication of the obviously ludicrous.

My third thought regards, well, me specifically and LGM in general.  Which law holds should, say, Mr. Galloway object to my referring to him as a gadfly and decide to take me to court for libel?  My guess is that English and Welsh law would hold, as I’m making this claim against a British subject, on my laptop located in my house in England.  This will be tweeted to the University of Plymouth’s feed (it was my dean’s idea to tweet my academic / comparative stuff to the university feed, not to refer to Galloway as a gadfly).  However, the “publication” of LGM is based in the United States.  I know that the SPEECH Act (2010) would protect me in American courts as referring to Galloway as a gadfly is protected speech, and British libel law is not consistent with the protections afforded under the First Amendment.

One might imagine that my amateur understanding of variance in libel law has a permanent address in the back of my mind considering that I work, and live most of the year, in Britain, though I would be stunned if anything published in LGM would be considered of a high enough profile to warrant action.  At least I feel pretty safe in my characterisation of Galloway as an “unprivate” individual, considering the clip above from Celebrity Big Brother in 2006.

Voter Fraud, Sincere Edition: Holy Crap, it Exists

[ 45 ] October 1, 2012 |

When I wrote about this a couple of weeks ago, I argued that voter fraud “simply doesn’t exist”.  Mea maxima culpa.  Unsurprisingly, the right has, yet again, beat the opposition to the punch, and have done so quite cleverly.  They have framed the issue of Voter ID legislation in easily digestible language that is difficult to refute with equal parsimony, knowing full well that the impact will be distributed asymmetrically across SES categories.  Furthermore, by deploying an army of “volunteers”, organisations such as True the Vote, motivated solely by a concern for the crumbling integrity of American elections, have succeeded in harassing legitimate voters predominantly in precincts that vote Democratic.

Simultaneous to playing defense, the right is also playing offense.  The RNC and various state Republican parties had hired Strategic Allied Consulting to lead a registration drive in Florida, Colorado, North Carolina, Nevada, and Virginia. Over the past week or so, allegations of fraud in these registration efforts have surfaced in 10 Florida counties and in Colorado, leading the RNC to sack the firm from its registration drive.

The numbers are small; adding the instances discussed in the NYT article amount to little more than 100.  As I argued before, the incentives required to induce somebody to vote once, let alone often, is high enough on an individual basis that to swing even a relatively minor election requires considerable investment.  However, this article also lists two occasions where those registering voters on behalf of SAC dispose of Democratic registrations.  This might be a larger problem.  One notably dim witted employee of SAC lacked the presence of mind required to forgo honesty about the process:

In Colorado, a young woman employed by Strategic Allied was shown on a video outside a store in Colorado Springs recently telling a potential voter that she wanted to register only Republicans and that she worked for the county clerk’s office.

The owner of Strategic Allied Consulting, a Nathan Sproul, has been suspected of systematic fraud in the past.  In 2004, he was investigated by the Justice Department and the Attorneys General of Oregon, Arizona, and Nevada for “widespread” voter fraud.  He was previously Executive Director of the Arizona Republican Party.  It is inconceivable that neither the RNC nor the various state parties were unaware of his history when they hired him to do the very job that triggered investigations in the past.  SAC tops the Florida State Republican Party expenditure list for 2012.  They had to know what they were buying.

Balloon Juice sums it up rather nicely, complete with puppies and kittens:

I’ll admit my first thought was that animal shelters and rescues groups keep carefully updated “Do Not Adopt” lists of individuals known to be hoarders, abusers, and/or generally unfit to have pets. You’d think political organizations would have an equivalent “Do Not Hire” list for people previously convicted of voter fraud and other chicanery… unless, of course, that’s exactly the kind of behavior the GOP/RNC/Romney campaign is hiring Sproul to commit?

Righteous defender of Democratic integrity Sproul was also hired by the Romney campaign in June as a consultant.

The story here isn’t that there’s voter fraud in Florida or other places, which requires the perfect storm to have an effect on an electoral outcome beyond insignificant.  Nor is it the ongoing destruction of Democratic registration forms, which if systematic and methodical, could have a larger impact, but still negligible.  The story, of course, is that while the right are deeply suspicious that the left will stop at nothing to “win” an election, including fraud, to the point that they’ve passed voter ID legislation in several states and are out in some force harassing voters in Democratic precincts, no evidence of systematic conspiratorial fraud has surfaced tied to the left, organisations affiliated with the left, Democratic campaigns, or the Democratic Party.  The closest the right has come to identifying anything remotely systematic was ACORN, which, as Brad Friedman notes early and often, is not comparable:

ACORN, the non-partisan, four-decade old community organizing group (which has since been forced into bankruptcy as a result of the years-long GOP effort to mischaracterize them and their work) there is no evidence, to our knowledge, that any of its tens of thousands of registration workers ever screened out potential registrants from one party or another before allowing them to register, as seen in CO.

Neither is there evidence that any of their workers ever changed party affiliations on registration forms, as is being alleged tonight in Palm Beach County, or destroyed Democratic forms, as has been alleged over the years, as noted by Republican Rep. Cannon.

. . .

Of course, there is no real comparison to ACORN. Unlike Sproul’s outfits, the non-partisan community organizing group was never hired by the Democratic Party to do voter registration work. Moreover, it was ACORN themselves who discovered fraud by a handful of its more than ten thousand workers and notified officials of the fraud and the names of those who had defrauded them.

As perhaps best described by former Republican Rep. Chris Cannon of Utah, during a 2009 voter suppression hearing: “The difference between ACORN and Sproul is that ACORN doesn’t throw away or change registration documents after they have been filled out.”

Sproul’s proclivities were noted by a Republican during a Congressional hearing, yet he was still hired by the Romney campaign, the RNC, and several state parties to continue his questionable practices all the while decrying voter fraud as an evil that could very well undermine the republic itself.

Man, you just gotta love these guys.

UPDATE: in response to a couple commenters, a distinction should be made between voter fraud, and voter registration fraud.  TPM have an article here.  That said, I wonder if this distinction isn’t borderline semantic, and invite discussion.  It could be argued that the end effect is what matters; a fraudulent registration leads to the possibility of a fraudulent vote, and more critically, the suppression of Democratic registration forms eliminates those votes from the electorate, the impact, while unmeasurable, is certainly more significant than fraudlent voting in the first place.  However, ultimately, the result is the same in the aggregate: a fraudlent vote adds one illegitimate vote to the tally of a candidate, a fraudulent (non) registration subtracts one legitimate vote from the tally of a candidate.


Social Desirability and Response Validity in Current Polling

[ 22 ] September 28, 2012 |

Several days ago in these very pages, discussion ensued regarding the latest conservative attempt to rewrite reality through re-weighting polls to one guy’s liking.  Of course, polling is not an exact science, but it is a science, and the latest wingnut delusion has no grounding in theory or empirical evidence.  Like any science, survey research continuously attempts to improve upon the validity and reliability of its measures and findings.  While I’m not at all concerned about some nefarious (and successful) attempt by the MSM and that paragon of power, the Democratic Party, to turn otherwise professional and reputable polling houses into duplicitous shills.

However, I have been somewhat interested (note, not concerned) if there might be something else going on that causes the polls to over estimate support for Obama.  Social desirability bias is something I’ve published on in the past (direct link to the paper here).  While that article suggests a contextual effect that causes variance in social desirability across countries (regarding accurately reported turnout in survey research), relevant here is what is colloquially known as the Bradley Effect.  It’s possible (though I consider it unlikely in the specific context of the 2012 Presidential election) that this helps explain Obama’s consistent polling advantage in an election where many if not most structural conditions suggest an incumbent defeat.

It’s difficult with the data I have available to examine this hypothesis to any satisfaction, but that’s not going to stop me from trying.

To begin with, we have the current state of the polls.

Obama Romney Advantage
RCP 28 Sep. 48.6 44.6 4.0
538 28 Sep. 52.2 46.5 5.7
538 6 Nov. 51.5 47.4 4.1


RCP’s running average has Obama up 4 points, Nate Silver’s “nowcast” model up 5.7, and his current prediction for election day 4.1 points.

If social desirability is at work here, a poll respondent will state that she or he supports the President because internally, our not entirely sincere respondent is seeking the socially desirable response, and not supporting the black guy might be racist.  However, this is done knowing that they will ultimately support the white guy.  Practically, this would mean that Obama’s support in these polls is inflated.

I’m approaching this from several directions.  First, I’ve averaged the final month of polls for Presidential elections going back to 1976 (an arbitrary cut off) to examine how accurate the polls were in predicting the final outcome between two white men, with 2008 to serve as a benchmark for 2012.  Shift represents how wide of the mark the final polls were, to the benefit or detriment of the incumbent party.

Poll Result Shift
2008 D 7.6 D 7.3 0.3
2004 R 1.5 R 2.4 0.9
2000 R 3.0 D 0.5 3.5
1996 D 11.0 D 8.0 -3.0
1992 D 12.0 D 6.0 -6.0
1988 R 12.0 R 7.0 -5.0
1984 R 18.5 R 18.0 -0.5
1980 R 4.0 R +10.0 -6.0
1976 D 2.0 D 2.0 0.0


Social desirability response bias in an election can take on many forms, not just race.  To wit, the 1992 general election in the United Kingdom is a good case study (one I lectured on here at Plymouth about six or seven years ago, shame I have no clue where those lecture notes now reside) as the polls largely predicted a narrow Labour outright victory or a hung parliament with Labour having the plurality of seats, yet the Conservatives under John Major easily won by 7.5%.  This is called the Shy Tory Factor on this island, which is simply another manifestation of social desirability.  But in 2012, I’m primarily considering race, and comparing 2008 to past elections does not support the hypothesis that this might be a problem for Obama’s numbers in 2012.

I also considered several of the primary elections in January, 2008.  This was the beginning of a primary where Obama was a somewhat unknown junior senator only four years into his Congressional career, going up against the assumed nominee.  For these data, I average all polls from the last week of the campaign in the given state (there were 20 in New Hampshire alone).

Poll Result Shift
NH Obama 35 37 2
NH Clinton 30 39 9
NV Obama 33.25 45 11.75
NV Clinton 37 51 14
SC Obama 41 55 14
SC Clinton 26 27 1


This evidence is more ambiguous than the examination of previous general elections.  Both Clinton and Obama received shifts in their favor, which isn’t surprising considering the undecideds presumably made a decision of some sort once voting.  However, in both New Hampshire and Nevada, the shift was stronger towards Clinton than Obama: a 7 and 2.25 point advantage respectively.  Both are dwarfed by Obama’s advantage in South Carolina.

These are the wrong data to be analysing this with, of course; ideally we’d have individual level data.  While not individual level data, the following figure, by Greenwald and Albertson, offers a more holistic view of the 2008 primaries.

The above shows that among 32 states where data were available, the “Bradley effect” was only evident in three states, yet 12 states demonstrated what has been termed (erroneously, in my opinion) the “reverse Bradley effect”: states where Obama’s support in the primaries was under, not over, estimated (see South Carolina above).  I consider this an erroneous classification because where the theoretical explanation for the Bradley effect hinges on social desirability, the reverse has been hypothesized as a function of systematic sample bias, through either the under-representation of African Americans in polling samples, or the cell-phone effect.  However, some have hypothesized that “black voters might have been reluctant to declare to pollsters their support for Obama”, and the link above does discuss that

After the Super Tuesday elections of February 5, 2008, political science researchers from the University of Washington found trends suggesting the possibility that with regard to Obama, the effect’s presence or absence may be dependent on the percentage of the electorate that is black. The researchers noted that to that point in the election season, opinion polls taken just prior to an election tended to overestimate Obama in states with a black population below eight percent, to track him within the polls’ margins of error in states with a black population between ten and twenty percent, and to underestimate him in states with a black population exceeding twenty-five percent. The first finding suggested the possibility of the Bradley effect, while the last finding suggested the possibility of a “reverse” Bradley effect in which black voters might have been reluctant to declare to pollsters their support for Obama or are under polled.

There are numerous possible explanations for the “reverse” effect, including faulty likely voter models, under sampling of blacks, sampling bias due to cell phones, to name a few.  There might be some sort of contextual effect at work here, but to ascribe it to behavioural motivations (rather than factors exogenous to the individual, such as sampling bias) such as blacks being shy to state their support for Obama fails the face validity test to me.

Ultimately, given the wide array of mediocre data presented here, I am not concerned about social desirability biasing the estimates of support for Obama in any significant, substantive manner.  However, much as I’d like to, I wouldn’t say that the Republican conspiracy is more likely, if only because that is so creatively ludicrous I initially thought it was an Onion piece.

The Voter Fraud Fantasy, Continued

[ 61 ] September 17, 2012 |

What really fascinates me about the voter fraud myth is how fervently the zealots believe this shit.  None of their phantom busloads of fraudulent voters transported in from out of state have been verified.  Few if any of the fraudulent votes that they identify appear so much as questionable to elections administrators.  It simply doesn’t exist, no matter how much they try to will it into being.

At the same time, they never question just why in hell somebody would fraudulently vote.  As a student of voter behavior in general with a sub specialty in turnout, I can list at length how the costs associated with voting far outweigh whatever tangible benefits one might accrue.  One has to navigate registration, polling locations, the hours the poling location is open, let alone being in possession of both the internal and external efficacy such that you’re confident in your decision, and that decision will make one iota of difference.  As one who studies and teaches this, it’s amazing to me that so many do vote to begin with.

Why bother to do it twice?  If one is unemployed and has nothing to do with their day, perhaps $20, $50, $100 will get me onto a bus from Chicago (and we all know what type of person lives in Chicago) for a day out up to Wisconsin.  Then you have to ask, for whom is it worth to spend that much money.  The sheer number of fraudulent votes necessary to tip an election such as the Wisconsin recall is not one here, another there, a busload up yonder.  It would have taken 171,106 votes added to the Democratic tally to swing it — to a recount.  Best add another 50,000 to ensure a recount doesn’t happen, as we wouldn’t want this fraud coming to light.  That’s 220,000 voters.  At $20 per fraudulent voter, “labor” costs alone are $4,400,000.  If it costs $100 for a person to give up their day, break the law, and experience both Wisconsin and the interstates, a cool $22,000,000 is required to replace a Republican with a Democrat in Wisconsin.

Then there’s transport.  It would require 2,445 sorties of a Blue Bird All American school bus to transport our wave of nefarious democracy from Chicago to Wisconsin, if packed to their capacity of 90.  I have no idea what it costs to rent, fuel, and provide a driver for one of these, but this cost is not insubstantial.  And don’t forget lunch; in the classic tradition of GOTV, we should feed our anti-democratic legions.

Let’s review.  We need a population of 220,000 people willing to give up their day, knowingly commit a crime, for some modest financial remuneration.  We require the infrastructure to transport them.  And food.

Basically, we need a hell of a lot of money to pull this off.  The entire recall itself, both sides, raised $63 million.  Our project would add a significant pile of cash to this total.

The activists have consumed large quantities of the kool aid.  They believe that these monsters exist.  But those financing the project really know what they’re doing.  It’s more pernicious than preventing granny from voting because she lacks state photo ID.  They’re specifically targeting minority precincts or any other conveniently concentrated demographic that tends to vote Democrat, acting as election “observers”:

In Houston, the group targeted the Congressional district represented by Sheila Jackson Lee, a Democrat who is black. Ms. Engelbrecht said the group settled on Ms. Lee’s district because thousands of addresses there housed six or more registered voters, which it took as an indication of inaccurate registrations. The methodology, which the group still uses, could disproportionately affect lower income families.

“The first day of early voting, at many of the 37 locations, primarily in minority neighborhoods, dozens of poll watchers showed up sent by King Street Patriots,”

It must frighten these people that six or ten adults live at the same address, let alone that they’re registered to vote at the same address.  This is an economic reality that is incomprehensible to the King Street Patriots.  Yet, here’s the reality:

“They had one particular case I remember very well,” said Douglas Ray, the Harris County assistant attorney who represents the election registrar. “They had identified an address where eight or 10 people were registered to vote. There was no building there.” Mr. Ray found out that the building had been torn down and that the people simply moved.

And then there’s those damned college students:

On Election Day, poll watchers appeared to have slowed voting to a crawl at Lawrence University in Appleton, where some students were attempting to register and vote on the same day.

Charlene Peterson, the city clerk in Appleton, said three election observers, including one from True the Vote, were so disruptive that she gave them two warnings.

“They were making challenges of certain kinds and just kind of in physical contact with some of the poll workers, leaning over them, checking and looking,” said John Lepinski, a poll watcher and former Democratic Party chairman for Outagamie County.

He said that as a result of the scrutiny, the line to register moved slowly. Finally, he said, some students gave up and left.

This transcends lofty concerns about good government.  This is blatant, targeted voter harassment.

And it will have an effect.  In stark terms, this is nothing more than an additional cost that must be incurred to vote.  I can readily understand how it dissuades people from voting to have an election observer, already distrusting your right to vote because you’re the other, breathing down your neck and that of the polling place workers.  Whereas we can roughly measure the effect that a lack of photo ID might have in terms of a reduction of the potential electorate, and we can infer from demographics the asymmetric impact that this has on the two parties, the sort of intimidation discussed in the NYT article linked above will be extremely difficult to account for from a social science perspective.

We’ll never really know what impact that this will have on the election.

Justice for the 96

[ 60 ] September 13, 2012 |

The Hillsborough Independent Panel’s report was released yesterday, following nearly three years of work re-examining the Hillsborough disaster of April 15, 1989.  A good summary of the findings are here.  The NYT has a story here.  For those unaware, Hillsborough (the ground of Sheffield Wednesday FC) was the neutral venue for an FA Cup semi final between Liverpool and Nottingham Forest.  96 Liverpool supporters lost their lives [*] immediately before and in the opening minutes of the match in a crush at one end of the ground — the match was cancelled only six minutes in.  The official narrative blamed the Liverpool fans themselves; drunk, violent, ticketless trying to force their way into the ground.  This myth was helped along by the print media, most notoriously The Sun.  Indeed, Boris Johnson, present Conservative mayor of London and all around moppy clown, oversaw if not wrote an unsigned editorial which reiterated blaming the fans, specifically “the part played in the disaster by drunken fans at the back of the crowd who mindlessly tried to fight their way into the ground” while editor of The Spectator as recently as 2004.  Today (literally), he is “very, very sorry“.  As is David Cameron, who exonerated the fans role yesterday in Parliament.

The true cause was not the fans or their behavior, but a combination of incredibly amateur crowd control on the part of the South Yorkshire Police and the remarkably decrepit state of stadia serving as the venues for the most popular spectator sport in Britain, against a sociological backdrop that stereotyped soccer fans as lower class hooligans.  The latter in part resulted in the perimeter fences then standard at grounds in the UK.  Indeed, the Leppings Lane end of Hillsborough had been subdivided into five “pens” (yes, they were called pens).  Hillsborough was supposedly one of the better grounds in England at the time, hence being a frequent location for a semi final, but it was in dreadful condition (as was a majority of the grounds around the country up and down the pyramid).  Slightly less than four years prior, 56 fans died in a fire at Valley Parade, home ground of Bradford City, and 66 people died in a crush (on exit) at Ibrox in 1971.  I can’t think of anything remotely similar in US major league sports during my lifetime (the year I was born, Bob Gibson had an ERA of 1.12, and Denny McLain won 31 games).  The best I can come up with on a cursory search is when some bleachers collapsed at the Baker Bowl in Philadelphia in 1903, killing 12.

The incompetence of the police, both in planning and during the match itself, deserves the majority of the blame.  While the tragedy was unfolding, a sizable portion of their presence was employed in making a barrier across the pitch to prevent the Liverpool supporters now on the pitch from rushing the Forest supporters at the other end (because that was precisely on their mind after they got over the relief at simply being alive).  These expressions of incompetence suggest why, in the aftermath, the South Yorkshire Police systematically covered up their responsibility.  In 116 cases, written statements by officers on site had been effectively cleansed, “to remove or alter comments unfavourable” to the police, in preparation for the official inquest.  This conspiracy extended to portions of the media and the government of the day.  It wasn’t the police, it wasn’t the infrastructure, it was those working class hooligans from radical Liverpool.

I have several friends who are Liverpool supporters, including a good friend of mine who lives down the street from my house here in England.  The findings of the report released yesterday have been common knowledge for 23 years.  But, it’s never been official knowledge until now; the official narrative was quite different.  A lot of people have been queuing up to apologise in the past 24 hours, deservedly so, including the FA for hosting the semi final at a ground lacking a safety certificate.  It looks as though the South Yorkshire police will refer this to the independent commission that investigates the police.

Anticipate a criminal investigation to the cover-up in the near future.

[*] 94 died at the ground, two more later in hospital.  I knew that.  Having just read this, I spotted my error I figured I’d correct it before a reader does.

Page 10 of 28« First...89101112...20...Last »