GrebnerBlog

This blog is intended to house essays discussing in-house research conducted by Mark Grebner.

The original “Technical Politics” was hosted by MichiganLiberal.com, and consisted of a series of essays written roughly from 2000 to 2012. At some point, the domain name was lost, and the essays became unavailable. I will post a few of the old essays here, looking for those that still seem relevant and useful. Some of the writing is probably lost forever.

Essay #1 – Recruiting Petition Circulators by Mail

As a result of the COVID-19 mess, circulating of nominating petitions became difficult, and various candidates found that reaching the legally required number was impossible. One solution I tried was to mail out blank petitions with an explanatory letter to people who might be convinced at least to sign and return the petition themselves, and might ever gather a few additional signatures from family members or others.

I convinced several candidates to try the method, which was more successful than I expected. Carol Koenig, running for Ingham County Circuit Court, captured the filled out petitions which were returned to her, and allowed me to compile the data. The following results are not intended to advance the frontiers of political science, but only to serve as a rough-and-ready recounting of what we learned by the process.

First, it should be explained that my firm, Practical Political Consulting, was hired to provide 5000 voter names and addresses for the mailing, which was sent First Class and contained a stamped First Class return envelope. We were hoping for possibly 5% response, but actually received petitions from 787 of the 4990 pieces mailed out – nearly 16%.

Koenig was not particularly well known in the community, but the circumstances were unusual, with media reporting of legal cases filed by candidates struggling with similar problems meeting the signature requirement. In short, the tactic may not be equally successful under less extreme circumstances. Still, the patterns we found would probably arise from similar efforts in the future.

The file selected for Koenig’s mailing was not a random cross-section of Ingham County voters, but was deliberately using a formula constructed based on gut instinct. We chose voters who each met several of the following criteria:

• attorneys
• Democrats
• primary voters
• signers of previous nominating petitions
• signers of previous initiative petitions
• circulators of previous initiative petitions
• financial contributors to political candidates

Posting the 787 returned petitions to the original file, I ran a simple set of cross-tabs to see which groups out-performed. The effects were not extremely strong, but the large universe made even small differences statistically reliable. Even among the very weakest groups, the response rate was about 8%, and only rarely did I see a sub-class reach 20%.

Without worrying about scientific nicety, this is what I found:

• Attorneys were slightly less likely to return signed petitions.
• Very consistent August primary voters were better than inconsistent August voters.
• People who requested absentee ballots in previous elections were better.
• People who had signed the Voters Not Politicians (anti-gerrymandering) petition were better.
• Having previously circulated a petition did not seem to matter.
• Having previously signed nominating petitions didn’t matter.
• The best age groups were surprisingly old. The very highest percentage return came from people in their eighties. Next best were people in their seventies and nineties.
• People under the age of 50 – keep in mind these had been selected because they generally seemed to fit the desired profile – were noticeably weaker.
• Gender didn’t make any difference.
• Partisan orientation didn’t make any obvious difference. (People who were clearly Republican were excluded from the mailing, but to the extent “soft Republicans” were included, they didn’t particularly to underperform.
• Geography didn’t make any difference. Urban, suburban, and rural all worked about equally well. Neither did socioeconomic status make any obvious difference.

 

Candidate polls work.  First, let’s talk about candidate polling.  Political polls are conducted all the time to assess which candidate is ahead in some race, and to predict who will win the election.  Each published poll has a “margin of error”, based on statistical theory, that states how far off its prediction is likely to be.  If you look back after an election, you almost always find that the polls were reasonably accurate in predicting which candidate would win, and by how much.

Ballot proposal polls don’t work.  It’s completely different with ballot proposals.  After an election, if you look at the final predictions in the polls, it’s common to find them missing by 10%, 20%, or even 30%.  A poll with a “margin of error” of 5% (which requires 400 interviews) should NEVER be off by 10%, not even one time in a thousand.  But missing by only 10% would be counted a success by most pollsters.

Let’s take a specific example:  the proposal in 2002 to repeal single-action straight-ticket voting in Michigan.  Various polls were published, mostly showing the proposal was strongly supported by the public.  The final poll of the campaign, published six days before the election by EPIC-MRA, showed the proposal with 77% support, compared to 21% opposed, and only 2% undecided.  The margin of error was approximately 5%, but on election day, the voters rejected the proposal by a 60% to 40% margin – meaning the poll was 37% off.   Of course, after the fact, we heard the usual gibberish about “late-deciding voters” and so on, but the truth was that the poll was completely worthless for predicting the outcome of the election.

When ballot proposal polls turn out to be completely wrong, a number of explanations are heard, none of which are reliable:

• “When the voters don’t know how to vote, they choose NO.”   (What about when a proposal that’s supposed to lose instead wins overwhelmingly, like 2006-4?)
• “It’s important to read the exact language from the ballot.”  (Firms that follow that practice have results that are just as bad.)
• “It’s hard to poll emotional issues – public opinion is apt to be volatile.”  (The biggest errors seem to occur on issues that have very little public awareness at all.)

Why are ballot proposal polls different?  There’s no theory that predicts that candidate and ballot question polls should be different, but since they are, we need to create a theory to explain it.  Over the past fifteen years, I’ve gradually worked out an explanation, made public predictions based on it, and been backed up by the election results.  This theory has nothing to do with statistics, and very little to do with conventional political science.  It’s simply based on what I’ve observed about polls that were either accurate or not, and the factors that seem to predict their accuracy.

First, we ought to ask why are candidate polls (reasonably) accurate?  The answer seems to be that when a voter talks to a pollster they make roughly the same choices they would if they were marking a ballot.  So if you ask a representative group of voters whether they plan to vote for McCain or Obama, you get more or less the “right” answers, so the laws of statistics determine how closely you can guess the actual election results.  That may seem obvious, but given all the practical problems of conducting polls (people who can’t be reached, who refuse to answer, who lie, who don’t end up voting, et cetera) it’s almost amazing that candidate polling works as well as it does.

A well-conducted candidate poll can furnish useful information about what will happen, whether spending additional money on a campaign would be a good investment, and even which issues are affecting voters’ decisions.  That’s what we mean when we say that candidate polling “works”.

In contrast, polling ballot proposals doesn’t work.  That is, even a well-conducted telephone survey of randomly chosen voters doesn’t provide accurate information about which side will win, or whether it’s worthwhile to invest more money, or what is really driving voters’ decisions.  As I said at Bill Ballenger’s “Pundit Summit” after the 2006 election, it would be much cheaper to buy three goats, have them ritually slaughtered, and then have the entrails professionally read – and the margin of error would probably be better.

The problem appears to be that when voters actually cast their ballots, their decision is mainly based on the actual language printed on the ballot, which they read under some time pressure and in secret.  The decisions that come out of that process simply aren’t the same as the ones they make when they are asked over the phone by a survey taker about the same issue.  It’s not clear which exactly what causes the responses to be different – maybe it’s different for different ballot proposals – but the differences are often very large and there doesn’t seem to be any way to “adjust” for them statistically.

A theory of ballot proposal testing.  Pollsters assume that voters have opinions about everything, and all you need to do is ask them, to find out what they are.  Ask them: “Would you support increasing fuel taxes by five cents per gallon in order to repair Michigan roads?” and they’ll tell you.  If their responses aren’t what you want, you can buy advertising and wage a public campaign to change their opinions.  On election day, the ballots are just an official method of recording and counting their opinions.

But on most issues, voters don’t have a ready, pre-formed opinion.  Voters are full of “attitudes” which can be brought to mind by a trigger, but those attitudes don’t precisely answer a question like “Would you support increasing fuel taxes by….”.  Depending on the circumstances, that kind of question could bring a complaint about taxes in general, about the bad condition of the roads, about the dysfunctional state government, about a particular official who recently made some public statement, or a host of other responses.   Being telephoned to answer a survey will cause a representative sample of voters to awaken and summarize their latent attitudes, which will result in a hard-and-fast percentage that say they would vote “YES” on such a proposal.  Then the survey firm will perform the necessary calculations to project the election result from those answers.  And those answers will have virtually no connection to what would really happen on election day.

Instead, it appears that the election day result depends on which of the voters’ latent attitudes are triggered at the moment each voter, having finished voting on all the questions higher on the ballot, finally reads the headline and 100-word description on the ballot.  There’s nobody to answer questions or even read the language.  Nobody can be impressed by how public-spirited the voter is, or how self-centered.  What images flash through the voter’s mind:  Potholes?  Lazy government workers?  A gas-pump’s electronic price display?  Whatever each voter thinks about at the instant before they mark their ballot is what determines how the votes are cast.

Why would the decisions made by a voter at the voting booth would be different from the decision made in response to being asked during a survey, since there’s no such problem with candidates?  Maybe reading a name on the ballot is so similar to hearing it over the phone that the same choices are made, which the more complicated ballot language results in differences.  It’s not theory that predicts candidate and proposal surveys will be different, but the evidence is overwhelming that they ARE different.  And the differences between polls and election results aren’t consistent enough to adjust them away by applying a simple rule or multiplying by a factor.

How can we accurately poll proposals?  The only solution we’ve found is to mimic the voting situation as closely as we can.  After a lot of trial-and-error, we’ve created a survey method that we’ve tested on a dozen different proposals, and we’ve always gotten acceptable results.

First, we create a “sample ballot” which includes much of the actual ballot the voter will face.  (We don’t include every single race, simply because the ballot varies in every precinct of the state, and perfect imitation doesn’t turn out to be necessary.)  Our ballot might have the candidates for Governor, Attorney General, Congress, State Senate, MSU Trustee, Supreme Court, Court of Appeals, and a series of ballot proposals.  (In other words, it would omit Secretary of State, UM Regent, Probate Court, et cetera.)

We send paid canvassers to randomly selected neighborhoods throughout the state, where they knock on the doors of registered voters and ask if they’d be willing to complete our survey.  We explain that it’s anonymous and they can mail it back in a provided envelope when they finish it.  Almost everyone approached typically agrees, and then they are pleasantly surprised when the canvasser hands them a shiny new dollar coin, “for your time”.  About 55% to 60% are typically returned.

The results.  Our “Straw Ballot” technique isn’t perfect;  it seems to be slightly less accurate than statistical theory predicts, possibly because of all the compromises that are necessary to put it into operation.  Instead of missing by 4% or 5%, we typically miss by 6% or even 8%.  But never by 20% or 40%.

In addition to conducting polls immediately before elections to predict their outcome, we have used “straw ballots” to test language before an issue is placed on the ballot, in order to determine whether a campaign is feasible or to help fine-tune the language, by circulating multiple versions that vary in some interesting way.  Those tests have been conducted for paying clients, and so are not available for disclosure, but their accuracy has been similar to our pre-election surveys.

Interestingly, our results for partisan candidates have not been particularly accurate – it appears ordinary telephone surveying is superior for that purpose.

There were five proposals on Michigan’s statewide ballot in 2006.  We’ve compiled all the polls cited in the Detroit Free Press and Detroit News during the final 60 days of the campaign, comparing them to PPC’s straw ballot results, and the actual vote totals.

Proposal             Polling Firm    Published     Yes    No

1 Trust fund      PPC                       —–           84.4   15.6
ACTUAL RESULT                81.0    19.0

2 Affirmative   Selzer                    Nov 5          39        49
Action              EPIC-MRA          Oct 27        40        44
EPIC-MRA           Nov 4        45        40
EPIC-MRA           Nov 7        41         46
PPC                        —–        64.7    35.3
ACTUAL RESULT              57.9    42.1

3 Dove                  EPIC-MRA            Oct 27      25          66
Hunting            PPC                          —–     30.4      69.6
ACTUAL RESULT             31.0      69.0

4. Eminent           Selzer                   Sept 4       43          44
Domain            Selzer                   Nov 5        44          46
PPC                       —–       87.2      12.8
ACTUAL RESULT             80.1       19.9
5. K-16                   Selzer                    Nov 5         43          45
Funds                EPIC-MRA           Oct 27        37          36
EPIC-MRA           Nov 7         39          45
EPIC-MRA           Nov 4         38          43
PPC                        —–        40.5       59.5
ACTUAL RESULT               37.7       62.3

I don’t have any illusion that anybody cares what I expect of Donald Trump’s presidency.  I don’t think I have any special knowledge.  If anybody reads this, it won’t give them tips on how to prevent the impending damage, or improve their own position.  In short, this post serves only one purpose: it allows me to put my markers down, so if I happen to be proven right, I’ll be able to point to it.  (Or, more literally, send out emails with the link, saying “I told you so!”)

Trump is so far from our national experience that it’s dangerous to make predictions based on similar candidates in the past – there just aren’t any good models.  As he has proven again and again, he’s willing to do things that nobody else has ever gotten away with, and that nobody thinks he can get away with either – until he does.

But we shouldn’t confuse his breaking our expectations with his ability to transcend physical or mathematical laws.  His businesses may have used bizarre tactics, but none of them have been particularly successful or profitable.  And he shows few signs (none, really) of maturing over time, so we can assume his future actions will be re-enactments of the things he believes worked for him in the past.  He’s not likely to try anything truly against type, such as honesty, dependability, or trustworthiness.

So here are a few predictions:

• Over the next six months or so, he’ll enjoy the honeymoon period every new administration is granted, as people who were on the fence give him the benefit of the doubt.  Because he starts out far less popular than any newly elected president in modern history, his honeymoon won’t take him to the soaring heights; there are just too many people who are already on to him.  But his favorable numbers, which are now 45-50% will rise to something like 60 or even 65%.  The opposition will become distraught as the Republican majorities appear to be enjoying a sort of public mandate for all the crazy and destructive things they’ll be working on.
• Trump will have no real problem with Congress.  Both houses are controlled by right-wing extremists, who will quickly assemble legislation which will shock even the people who think they’re prepared for what will come.  Abortion rights, voting rights, union rights, free speech, academic independence – they’ll all find themselves beleaguered and desperate for support.  Trump won’t be a problem for this agenda, not because he agrees with it – he doesn’t actually have any fixed political views at all.  But he will quickly discover the Republican leaders in Congress will give him any symbolic victory he wants, in exchange for his signing off on their (non-symbolic) power grab.
• This will all proceed for at least six months, with only an occasional hiccup.  But Trump’s continued incontinence, combined with Republican overreach, will begin to create problems for themselves.  Public support will peak, and then begin to slacken, as the Republicans are seen undermining popular programs and positions.  That doesn’t mean they won’t be successful, but only that each success will cost them.
• Pretty soon, we’ll see the first glimmerings of evidence that Trump can’t repeal the fundamental laws that govern public administration.  In particular, Trump has always shown that he’s completely incapable of genuinely delegating authority to autonomous subordinates.  If somebody isn’t family, and they begin to show independent judgment, Trump turns on them.  His businesses, like his presidential campaign, have always been hollow structures that simply echoed his personality.  He can’t trust outsiders, and he can’t keep his hands off the controls.  The result – in every century and every society – is an inefficient and corrupt mess.  The Federal government in 2017 is WAY too large for Trump’s personal deal-making to be effective.
• For a while, each cabinet department can coast on inertia.  But in one case after another, major problems will arise due to the incompetence, inexperience, and fanaticism of the people Trump has selected, both at the Secretary level and one or two levels below that.  At first, the public won’t perceive a pattern, and will give Trump some slack.  But as the honeymoon period wanes, public tolerance for wackiness and dysfunction will wear very thin.
• Sometime about a year from now, the first glimmerings of investigations and even prosecutions will emerge.  One result of putting outsiders with no experience in control is that hundreds of tempting financial opportunities will be left unguarded.  People who until now have been running scams of a few thousand dollars at a time will suddenly realize there are opportunities in the million and billion dollar range, if only they find the courage to grab for them.  Purchasing, hiring, grant evaluation – they total in the TRILLIONS of dollars.  As Ronald Reagan’s administration demonstrated, a few dozen petty crooks can create a cesspool that encompasses the entire Federal government and every department of the cabinet.
• Trump’s reaction to the relative acclaim and calm of the honeymoon period has already proven to be brittle and hyper-reactive, in keeping with his character.  As public support drops into the forties, and then the thirties, various investigators will feel free to pursue the infinite number of loose ends Trump always leaves.  Criminal investigations.  Civil suits.  Journalists.  Leakers.  Political opponents and opportunists.  Pretty soon, Trump will find himself surrounded by critics, and – once his popularity drops – he won’t have the firepower to keep them at bay.
• Eventually, this is going to cause a terrible problem for the Republican Party, much like the anguish we saw during both the campaign for the nomination and again in the general election.  Republican office-holders will realize they have to either be WITH Trump or AGAINST him.  They either have to assist those who are undermining him, or they have to circle the wagons and threaten the heretics with burning.  My guess is that – even as Trump loses public sympathy – the Republican hierarchy will keep making the decision to draw together in his support.  This would be very similar to the Republican experience in 1974 with Nixon, where a few took well-remembered positions of courageous opposition, while the overwhelming majority stood silently with him until the end.  (And were routed in the 1974 elections, as a result.)
• I see one large dark cloud on this scene, which might allow Trump to continue his scams for several additional years – conceivably even long enough to be reelected in 2020.  All he would need would be some sort of violent threat, to cause the public in its simpleminded patriotism, so rally around him.  Either a foreign war (See: Grenada) or domestic terrorist incident (See: September 11) could do the trick.  In the past, I would have taken for granted that such events would have been governed by the random forces of history.  But both Trump and his Russian friends have made clear that they’d be perfectly willing to fabricate whatever incident would allow Trump to continue his depredations.  The impact on world history (and less important, on the Republican Party) would be incalculable, and not in a good way.