politics – Page 2 – ViewfromthePearl.com

Problems with polling: Redux

I haven’t posted since April since I had little to contribute to what I saw as the two overarching goals of the last six months: electing Joe Biden and developing a COVID vaccine. I did my civic duty toward the first and had nothing to contribute to the second, and so it seemed a time to pause. Now, I feel my free speech is restored and for a moment at least there is some attention to one of my favorite topics – how we need to do research for campaigns differently. I have covered much of that previously but here is a redux on the problems with polling with some updating for 2020.

1. Samples are not random. If you ever took an intro stats course, it grounded most statistics in the need for a random sample. That means that everyone in the population of interest (e.g. people who voted November 3) has an equal probability of being included in the sample.

The margin of error presumes a random sample. The number of people required to give you an accurate picture of the array of views in a population depends on size of the sample, the breadth of the views in the population, and the randomness of the sample.

The intuitive example: Imagine a bowl of minestrone soup. If you take a small spoonful, you may miss the kidney beans. The larger the spoonful (or sample) the more likely you are to taste all the ingredients. The size of the spoon is important but not the size of the bowl. But if you are tasting cream of tomato soup, you know how it tastes with a smaller spoon. America is definitely more like minestrone than cream of tomato.

The problem with polling has little to do with the margin of error, which remains unchanged. The problem is that pollsters have not used random samples for a generation. The advent of caller ID and people’s annoying proclivity to decline to answer calls from unknown numbers (a proclivity I share), plus some changes in phone technology with fiber optics – including a proliferation of numbers that are not geographically grounded, and an explosion of polls and surveys (How was your last stay at a Hilton?), makes the act of sharing your opinion pretty unspecial.

Not to worry, we pollsters said. Samples can still be representative.

2. The problem with “representative” samples. A representative sample is one constructed to meet the demographics and partisanship of the population of interest (e.g. voters in a state) in order to measure the attitudes of that representative sample.

The researcher “corrects” the data through a variety of techniques, principally stratified samples and weighting. A stratified sample separates out particular groups and samples them separately. Examples include cluster samples, which stratify by geography, and age stratified samples, which use a separate sample for young people, who are hard to reach.

Professional pollsters usually sample from “modeled” files that tell how many likely voters are in each group and their likely partisanship. They upweight – or count the people they are short of extra. They may up-weight the conservative voters without college experience, for example, to keep both demographics and partisanship in line with the model for that state or population. Virtually every poll you see has weighted the data to presumptions of demographics and partisanship.

Back to the minestrone soup example: Samples are drawn and weighted according to the recipe developed before the poll is conducted. We presume the soup has a set quantity of kidney beans because that’s what the recipe says. But voters don’t follow the recipe – they add all kinds of spices on their own. Pollsters also get in a rut on who will vote – failing to stir the soup before tasting it.

Most of the time, though, the assumptions are right. The likely voters vote and the unlikely voters do not, and partisanship reflects the modeling done the year before. But disruptive events happen. In 1998 in Minnesota, most polls (including my own) were wrong because unlikely voters participated and turnout was unexpectedly high particularly in Anoka County, home of Jesse Ventura, who became Governor that year. That phenomenon is parallel to the Trump factor in 2016 and even more so in 2020. Unexpected people voted in unexpected numbers. If the polls are right in 2022, as they generally were in 2018, it is not because the problem is fixed but because conventional wisdom is right again, which would be a relief to more than pollsters, I expect.

3. What’s next. I hope part of what’s next is a different approach to research. If campaigns and their allies break down the core questions they want to answer, they will discover that there is a far bigger and more varied toolbox of research techniques available to them. The press could also find more interesting things to write about that help elucidate attitudes rather than predict behavior.

Analytics has a great deal more to offer. That is especially so if analytics practitioners became more interested in possibilities rather than merely assigning probabilities. Analytics has become too much like polling in resting on assumptions. Practitioners have shrunk their samples and traded in classical statistics for solely Bayesian models.

Please bear with me for another few sentences on that: classical statistics make fewer assumptions; Bayesian statistics measure against assumptions. When I was in grad school (back when Jimmy Carter was President – a Democrat from Georgia!), people made fun of Bayesian models saying it was like looking for a horse, finding a donkey, and concluding it was a mule. We will never collect or analyze data the way we did in the 1970s and 80s, but some things do come around again.

It would also be helpful if institutional players were less wedded to spread sheets that lined up races by the simple probability of winning and instead helped look for the unexpected threats and opportunities. In those years when everything is as expected, there are fewer of those. But upset wins are always constructed by what is different, special, unusual, and unexpected in the context of candidates and moment. Frankly, finding those is what always interested me most because that’s where change comes from.

More on all of this in the weeks and months ahead, and more on all the less wonky things I plan to think about Democrats, the south, shifting party alignments, economic messaging, and my new home state of Mississippi. I am glad to be writing again, now that I feel more matters in this world than just Joe Biden and vaccines.

Some questions for post-Labor Day Polls

I suspect we will see a spate of new polls fielding after Labor Day. I am hoping they ask some questions beyond the horse race that tell us more about what voters are thinking around the Democratic presidential contest. Here are some suggestions (in no particular order):

Candidate Qualities

Here are some qualities people might look for in the candidate they ultimately support for President. On a scale of 1 to 7, please tell me how important each one is to you, with a 1 meaning not important at all and a 7 meaning it is the most important quality. (READ AND RANDOMIZE)

Can beat Trump in November
Shows compassion for people
Knows what they want to do as President
Would bring the country together
Would make significant policy changes
Has a new approach to governing
Will protect individual rights and freedoms
Will promote economic opportunity
Has the wisdom of experience
Will advance equality and anti-racism

Which of these qualities – or some other quality – is most important to you of all?

Electability

We know voters care about whether a candidate can beat Trump but we don’t know what qualities make a candidate stronger in their views. How about a couple questions, like:

How important are each of these in telling you a candidate can defeat Trump in November, using a scale of 1 to 7 with a 1 meaning it is not important at all and a 7 meaning it the most important quality? (READ AND ROTATE)

Is there another quality that is important in telling you a candidate can win?

Tough and willing to fight
Has moderate issue positions
Popularity with Trump voters
Inspires young people
Relates to diverse communities
Leads Trump in the polls
Likeable and appealing

Thinking about your friends and neighbors, if the Democratic candidate is a woman, will that make them more or less likely to turn out and support that candidate in November, or won’t it make any difference to them?

Thinking about your friends and neighbors, if the Democratic candidate is over age 75, will that make them more or less likely to turn out and support that candidate in November, or won’t it make any difference to them?

Thinking about your friends and neighbors who are uncomfortable with Trump, do you think they are looking more for a return to the pre-Trump years or more for new policies that will bring change?

Issues

What issues are most important to you in the 2020 election? (Open-end, multiple response)

If we elect a Democratic president in 2020, which of the following should be their top priority in their first term: (READ AND ROTATE)

Climate change
Affordable health care
Access to post-secondary education
Infrastructure like roads and bridges
Higher wages
Immigration reform
Criminal justice reform
Other (specify)

When it comes to health care, which would do more to expand access to quality affordable care – (ROTATE) a public option in which voters can choose government-administered insurance OR Medicare for all in which everyone is in a government-administered insurance program (with response options for neither as well as don’t know)?

(IF CHOICE) Would that system be much better, somewhat better, somewhat worse, or much worse than the current system?

If there is a Democratic president, how likely is it that the proposal will become law in the next five years – very likely, somewhat likely, not very likely, or not at all likely?

The Horse Race Question

I have been concerned that asking about 20 people in phone polls flattens choices because people can only hold seven plus or minus two item in short term memory. Consider asking the horse race in groups of 5 to 7 candidates – preferably randomizing sets although it is also tempting to ask the top 7 together. Then add a question like:

You indicated candidates A, B, and C were your top choices within the groups I gave you. Which of these is your first choice among all the candidates?

Is that the candidate you would most like to see as President or the candidate you feel can best win? (Code for volunteered both)

Vote History

Most public polls are asking how likely people are to vote in the primary or caucus in their state. Consider asking whether they voted in the 2016 contest between Hillary Clinton and Bernie Sanders and for whom they voted.

The question allows analysis of how large a primary electorate you are polling and what the standing is among those most likely of all to participate, as they have done so historically. It will also say where the support for these two candidates is going.

Demographics

Basic demographics are fine, but also consider asking whether they live in a county that supported Clinton or Trump in 2016, as these voters may have different perspectives from each other.

# # #

Questions like these would say more about what voters are looking for in the next president (other than that he or she is not Trump). Crosstabs of questions like these by candidate preference might also provide more insight as to why voters are making the initial choices they are, and how the contest may evolve.

Liberal Elite Is Not a Hyphenated Term

There is much discussion lately about how some Democratic candidates may be “too liberal” to win. That term needs disambiguation.

First, “too liberal” may be about policy: if a candidate will raise voters’ taxes and spend money in ways they do not believe benefit them, they may be judged “too liberal.” Second, candidates may express liberal social values that voters believe are out of step with their own. Often, however, too liberal is a euphemism for a candidate who appears to disrespect the way people live their lives and the struggles they face. A candidate is “too liberal” if they appear too elite. Arguably the third of these – elitism – has been more of a problem for Democrats than the first two.

Some things to watch out for:

The Politics of Pandan

Shortly after the 2016 election, I had a lovely dinner at a French-Asian fusion restaurant with my friend Ed. The food was terrific and we ended the meal sharing chocolate encrusted Pandan Cheesecake, which cost about $12. It was delicious. Since we were not familiar with pandan, we asked the server about it, who extolled its subtle vanilla-like flavor, bright green color, and widespread use in Indonesian cuisine. When the server walked away, Ed turned to me and said, “That conversation we just had, that’s why we lost the election.”

Ed had a point.

Pandan cheesecake consumption reveals disposable income and foodie tastes. It may also show a tendency to waste money – a lot of people think it wasteful to pay $12 for an individual exotic dessert when you can get a whole frozen Sarah Lee cheesecake at Walmart here in Jackson for $4.98.

On its own, the waste may be excusable. Until pandan-eaters start to make fun of Sarah Lee aficionados on social media.

Donald Trump is not a pandan guy. His lack of elite tastes are an asset to him. Especially when the pandan-eaters make fun of his putting ketchup on his well-done steak, his swoop-over hair, and the ill-fitting tuxedo he wore at dinner with the Queen.

Just cut it out. Stick to how Trump’s policies hurt people and damage the country. Excise from conversation any notion that elite tastes rule – or should.

In a democracy, Ivy League grads are not better than those whose last degree was from their local community college. Those who vacation on Martha’s Vineyard are not better than those who go up to the lake. And pandan-eaters are not in charge of devotees of Sarah Lee.

Victims and Executioners

Adding to a confessional of my own tastes, I acknowledge that Albert Camus’ essays on being neither victims nor executioners are core to my world view. Camus wrote that, “In such a world of conflict, a world of victims and executioners, it is the job of thinking people, not to be on the side of the executioners.” (He wrote it in French. I don’t read French.)

Generally, liberals are not on the side of executioners, but there are more victims and more executioners than many acknowledge.

I have never polled the question of whether people believe they have been the victim of the arbitrary and unfair exercise of power, but I suspect just about everyone has felt a victim of that experience. There are the patterns of discrimination – systemic racism and sexism – but also the frustrations of dealing with bureaucracies and bosses, and the feeling of being unheard, misheard, or misunderstood by people who have power over you.

Some of the ways people are treated unfairly have policy remedies – I wish Democrats would discuss the overtime rules more than they do – but others are just there. People are not looking to elected leadership for redress of all the ways life is unfair. Still, it would behoove leaders to recognize and acknowledge that life is unfair for almost everyone.

A whole lot more people face executioners and warrant support than we often recognize. When we fail to acknowledge them and their struggles, we risk their choosing – when they have the chance – to become executioners themselves.

Zero Tolerance

I am generally uncomfortable with the idea of zero tolerance – it proscribes a world without ambiguity or exceptions – even when zero tolerance is for unambiguously bad behaviors like hate speech, drunk driving, and unwanted touching. Vaunting the idea of zero tolerance can take a judgment designed to protect a victim, and turn it into an act of execution.

I would like to think, for example, that all men will respect all women all of the time and never treat them as objects. After all, the Bible says (Matthew 5:28) “anyone who looks at a woman lustfully has already committed adultery with her in his heart.” Former President Jimmy Carter, a man who knows his Bible, acknowledged to Playboy Magazine that he had committed “lust in his heart many times.” It happens. Among all genders, binary and non-binary.

Racist images are common among white Americans. Almost everyone has unfairly treated someone as “the other.” Or made an assumption about skill or character based on appearance.

Oppose bad behaviors that make people victims. Declare that hate speech, drunk driving, and unwanted touching are wrong. But it is a political and arguably a moral error to decide that those who have practiced a bad behavior should be the objects of zero tolerance. Argue, persuade, point out the error of their ways – and prosecute them when a crime is committed. Don’t make people the victims of zero tolerance in the name of tolerance. Only the very few who are entirely guilt-free will respond to that.

# # #

So what does all this have to do with Medicare for All?

I don’t know if people are willing to risk losing something they have for something that may be better. It requires a leap of faith. I know they will not trust someone whom they perceive as elitist or who fails to recognize that people are nervous about the exercise of power, which in their experience is rarely benign.

The perception that Joe Biden is more electable than Elizabeth Warren is in part because he does not have the image of being a pandan-eater. And he certainly knows that life is unfair.

Elizabeth Warren, whose biography to a point is less elite than his, can gain on trust and empathy over the next five months. If she does make those gains, the policy nuances of Medicare for All versus a public option seem unlikely to be dispositive in a November match-up that includes either of these candidates.

Is Joe Biden the most electable Democrat? Maybe – but maybe not…

Joe Biden’s lead in the Democratic primary field apparently rests on the three legged stool of long-standing familiarity, appreciation particularly among older African American voters as President Obama’s trusted second, and a perception that he is more electable than other Democrats.

Electability is not actually a testable proposition as the Democrats will have only one nominee who will win or lose. Still, the concept of electability is important to voters and thus worth examining.

Biden’s perceived electability may rest in part on his having been part of a winning national ticket. But as Natalie Jackson pointed out in the Huffington Post, of the nine Vice Presidents who ran without first acceding to the office on the death of the President, only three have won. George H.W. Bush was the most recent Vice President to win, then Richard Nixon, although he lost before he won. Before Nixon, you have to go back to Martin Van Buren to find an example of a Vice President who won election without becoming an incumbent first (Vice Presidents Elected President).

Blue collar appeal – or the avoidance of elitism and its imagery – will certainly be important in the 2020 election, given the states needed for an Electoral College win. Biden has a reputation for appealing to blue collar voters, although the evidence seems largely anecdotal. He grew up in a blue collar family but such roots are not unique in the field. They are shared by at least Elizabeth Warren and Bernie Sanders. Others grew up in families that were far from wealthy. Several are the children of school teachers or of a single parent who struggled financially. It is not a field of elites.

Here are some other considerations where Biden may not reach parity with others in the field:

A Referendum or a Choice: One of the most fundamental framings of an election is whether it is a referendum on the incumbent or a more lateral choice between the contenders. The Democratic nominee will be the challenger to an unpopular incumbent. Democrats should be advantaged if the election is a referendum on Trump.

Biden may be the best positioned to make the campaign a referendum because he is more moderate, has lower negatives (at least now), and is a comparatively comfortable choice for many.

There is, however, a risk that the electorate is not in the mood for such a referendum election. They may want to know what is next on raising wages, lowering health care costs, making America safer, healing divisions, and changing the way we do politics.

Biden may have answers on all of those issues, but his long record can be a hindrance in saying that he will bring more positive change than the country experienced in the Obama administration. As popular as President Obama is, the demand for change may exceed the nostalgia for him.

A Different Electorate. If the electorate were similar to the 2016 electorate, a return to the Obama years might be enough. The 2018 election saw enough dissatisfaction with Trump, and enough change in turnout patterns with higher Democratic and lower Republican participation, that it is tempting to simply try to carry the 2012 – or 2018 – election forward.

The electorate seems unlikely to be a mere extension of the last two elections, however. The eligible electorate will have more voters of color, especially voters of Hispanic origin, and it will be older, absent a shift in turnout that increases the number of younger people who participate. (Pew: An Early Look at the 2020 Electorate).

The 2020 electorate promises to be larger than the 2016 electorate. The Trump campaign and its allies are already spending millions online to find people who have not voted before or voted irregularly whose participation they can compel. Emotion on the Democratic side likely means higher turnout organically – and presumably Democratic efforts to expand the electorate will eventually match Republican efforts.

Still, an electorate that is more Democratic than 2016 likely depends on turnout, particularly among younger voters who have the most room for turnout growth. It may be that an anti-Trump message will compel turnout, but it would also be useful to have messaging and a candidate who can optimize that young voter participation.

Biden does not appear the best candidate to motivate younger voters. Younger candidates, particularly Pete Buttigieg and Julian Castro are speaking to generational change, while Jay Inslee is making climate change – an issue particularly compelling to young voters – central to his campaign, and Elizabeth Warren is speaking to student debt and child care issues, which resonate with younger voters.

Indeed, on the message level, several other candidates seem to have stronger youth appeal than Vice President Biden.

Anti-Partisan Voters. Swing voters by definition are not locked into a party – that is part of what makes them swing voters – and the voters who supported both Obama and Trump fall into this category. Biden seems to be betting that they will like bi-partisanship, and speaks to his historic civility with Republicans.

Many swing (and third party) voters, do not like either political party and are more interested in non-partisanship than bi-partisanship. They are looking less for politicians to “reach across the aisle” (a term they often do not understand) than for leaders to be separate from either party. That is a tougher case for Biden who has lived party politics, than for those like Governors and Mayors who have some separation from the hyper-partisanship of Washington, or those who are newer to politics than Biden is.

Swing voters tend to be younger than average and are disproportionately women. Biden may not be ideally positioned to motivate either young voters or women voters who are mistrustful of politicians of either party. There are white, blue collar, older men who voted for President Obama and for Trump, but not very many and they are unlikely to be the core swing vote, as opposed to younger women without college experience – across ethnic lines – who are low or moderate propensity voters.

# # #

The first test of electability is whether a candidate will be nominated. Biden has the lead right now in most polls, although the size of the lead depends on assumptions about the shape of the electorate. Biden is stronger among older Democrats than younger Democrats, and seems especially strong among older African American voters.

Biden seems for many Democrats to be the safe choice at a time when defeating Trump is Priority One. The problem is that playing it safe may not be the best winning strategy.

The first test for some of the contenders with apparent advantages among younger voters, is whether they can motivate their participation in the early primaries and caucuses. If so, they have more room to overtake Biden. In showing their capacity to attract younger voters, they may also take away his electability argument.

Women and Electability – Part 2

In Part 1 of this post, I argued that there is no solid reason to consider women less electable than men in the 2020 Presidential contest. Women candidates do need to grapple, however, with four areas that may create misconceptions of their potential. First is implicit bias, second is the nature of leadership archetypes (and negative stereotypes), and third is the management of the strong value among women voters of caregiving and the “Caregiver” archetype. Finally, there is the differing nature of media coverage of women candidates. None of these are barriers but they are considerations for women candidates and those observing them.

1. Association and Implicit Bias. One reason men may be currently considered more electable is simply how often they have been elected. Older men look more like the panoply of former Presidents than women do, even though men have given up both the wig of our first president and the mutton chops of many. People are more used to seeing men in leadership positions and so they associate men and leadership qualities.

Both academic and popular research shows that people have stronger associations with men and leadership qualities than with women and leadership qualities. Such “implicit bias” is not necessarily unconscious and it does not necessarily project behavior. In fact, there is some academic literature lately that suggests it does not predict behavior. Still, there is such bias.

To measure yours, the American Association of University Women has provided a test on line. It is anonymous and instructive: AAUW Implicit Association Test of Gender Bias. Implicit bias is not a barrier because at a time when people may want change – and perhaps big, structural change, as one candidate promises, such associations may not matter, or perhaps even underline the change a woman might bring.

2. Archetypes and Negative Stereotypes. At a deeper level than the associations shown in implicit bias tests, there is leadership imagery that is sometimes more male than female. Jungian psychology introduced the idea that we share unconscious ideas, often gender-associated. Thirty years ago, Robert L. Moore and Douglas Gillette wrote King, Warrior, Magician, Lover about the archetypes of mature men (as opposed to other archetypes like “the Trickster,” who survives challenges through trickery and deceit, which may remind you of someone).

The use of archetypes for communications and branding is recounted in Margaret Mark and Carol Pearson’s classic book, The Hero and the Outlaw: Building Extraordinary Brands through the Power of Archetypes. Their work helped establish Nike as “the hero” brand, and Apple as “the magician.”

Some archetypes are more gender-laden than others. The King is certainly a gender-based term (and the Queen has different associations), as is the “Everyman” or “Regular Joe” that their book discusses. There are, however, women Warriors, Magicians, and Sages, which are among the highly desirable leadership archetypes, in mythology, in history and in popular culture.

The shadow (a Jungian term too) of the Queen archetype bears watching. There we find negative stereotypes like the manipulative “Queen Bee” who destroys other women while the men work for her; the Queen with her clique of Mean Girls so well-profiled by the movie of that name; the Bossy Beyatch (to use the more acceptable colloquialism), and her sister the “Angry Woman,” with the latter two carrying qualities less acceptable for women than for men.

Another archetype, the Innocent, is not inherently negative but not what voters want in a President. Children and some women are perceived as The Innocent and we do not want a President who is untutored in the ways of the world.

The “Good Girl” or “Daddy’s Girl” archetype is more likeable but follows status quo authority a little too much rather than bringing change, and generally follows men rather than aligning equally with other women.

The “Victim” archetype is also not a desirable President. The victim feels powerless and blames others for their predicament. There is a fine line for women leaders in talking about discrimination against women and sounding like they believe women – perhaps including them – are victims. The President of the United States should show compassion for victims but should never be a victim.

3. The Caregiver. One of the biggest challenges for women candidates is integrating one of the most powerful, positive – and generally female – archetypes: The Caregiver. The Caregiver is compassionate, generous, thoughtful and kind. It is reputedly part of the branding of Campbell’s Soup, Johnson & Johnson, and McDonald’s – with billions and billions sold.

Women identify with caregiving. In a survey I conducted for a client many years ago, nearly 70 percent of women voters said caregiving was one of their most important values. The importance of caregiving is presumably why George W. Bush modified his declaration of being a conservative with the word “compassionate.”

Women candidates – despite their self-evident ambition and aggression – generally have advantages on compassion, as they do on issues associated with it like health care. Failing to display Caregiver qualities can alienate other women, who value caregiving in themselves and in leaders. The challenge is in nurturing the caregiver, which almost all successful women candidates do, without appearing to be the Innocent or the “Good Girl.”

Part of the answer for women candidates in balancing strength and compassion is to define who and what they fight for: in the mythological world from which archetypes derive, the male fights to be the Alpha male; to win the competition for its own sake. The woman or female fights to protect her cubs (if she is a lioness or a bear), or her children, family or community. The behavior may look the same but the motivation is different.

Note that if she fights for victims, then you have to see yourself as a victim to believe she fights for you – and most people do not see themselves that way – and even fewer want to be a victim. The strong Caregiver fights for what she loves to make it stronger. The Caregiver is not patronizing.

4. Media Bias. Others have written about how the media cover men and women differently and how some men candidates do the same (Suzanna Danuta Walters Washington Post Op Ed). I do think the coverage is more balanced than it has been in the past and some in the press clearly make a conscious effort to diversify their sources. Still, the reality remains that most of the press corps covering the presidential campaign and reporting it on television are men, and, indeed, white men. The media need to understand and give women candidates’ credit for messaging and strategies that incorporate gender differences – women candidates and their strategies are not supposed to be just like men’s. I hope the press talk to more women – and many more people of color – who live outside the bubble of punditry about what they hear the candidates saying, and what they are listening for. The perspective is likely to be different, but also more reflective of the majority of voters.

# # #

It is still 230 days until the Iowa Caucuses when the first votes are cast. The electorate in Iowa will very likely be larger and younger than eight years ago, and in other states it will be larger, younger, and more ethnically and racially diverse than in the past. In every state, the majority of the primary electorate will be female.

In what may be a historically large primary and general election, pollsters don’t quite know who to talk to – and not everyone wants to talk to pollsters. The best anyone can do at this point is to reach out to the extent they can, and be aware that the dynamics of gender, race, ethnicity – and generation – are changing the electorate in ways that may be difficult to predict. The picture may look similar in seven months. Or it may be very different, indeed.

Buying a Ticket Out of Iowa – Online

The Iowa caucuses help frame who is a contender for the nomination, which is especially important in such a large Democratic field. Historically, there are “three tickets out of Iowa.” Only once in recent political history has anyone become President without a top-three finish in the Iowa caucuses (although a whole lot of precedent-breaking is going on, including early California voting right after the Iowa caucus).

One element of precedent-breaking that is certain: online communications matter more. Yes, Iowa is an older electorate and people expect a strong field organization. But it is also heavily wired and online engagement is starting high and growing (although the number one search name nationally in the last 30 days was not a candidate but Nipsey Hussle – I checked.)

Here’s one approach to researching and defining your target online:

1. How many voters do you need? First, decide the size of your initial target. There are complexities in that basic calculus. Democrats have not always released raw totals as opposed to delegate percentages so history is limited and the number of caucus attendees per delegate varies across the state. Additionally, 10 percent of delegates this year will be chosen by a mobile phone caucus held before the in-person caucus event and campaigns will ultimately need to have separate goals for each stage.

As a starting point, (unless your campaign is Sanders or Biden, both of which have higher bars), I recommend finding 75K voters – reasonably distributed across the state – who become committed to your candidacy. The highest Democratic caucus turnout was 239K in 2008. Even with the mobile caucus 300K this year would be a stretch, as it represents nearly half the registered Democrats. Thus, 75K should produce at least a 25 percent popular finish and threshold everywhere; it is also competitive with Sanders who won less than half 2016’s 171K participants, not all of whom will either stick with him or return to the caucuses.

2. How to find your vote. Every campaign will individually target repeat caucus attendees but well under 100K participated in both 2008 and 2016. So talk to the repeats and monitor their choices – at least until they stop answering their phone (although you should have their emails by then).

The next layer is people who will turn out in the caucuses because they are excited about one of the current candidates – that is the factor that ultimately expands the caucus universe; people who are caucusing not out of habit, civic duty, or party loyalty but because they really support someone. Finding your own unique base also means less immediate competition for those voters and time to engage and mobilize them. That’s where online strategies help – in finding the people with whom your candidacy resonates enough to draw their participation and who are not prior caucus attendees (whom everyone will target in field, mail and online).

3. Defining your target online. As soon as interest from outside the regular caucus universe is in the thousands – and preferably close to 5,000 – your internet team can do “look alike” modeling to find people whose internet behavior is similar to theirs. Those who have opted in to your emails, attended an event, contributed money, or simply visited your web site tell the campaign in online terms, rather than simply demographics, whom it is attracting. The most likely next set are people who behave like them in terms of their online habits. Exclude from that modeling regular caucus attendees so you are finding more people who are attracted specifically to your campaign.

Add to the look alike model message-driven targets through affinity targets or search terms. If your candidate is a veteran, people who search for veterans benefits on line can receive an ad about your candidate. Or if your candidate has been a leader on climate change, those who search on that issue should hear about it. If your candidate just announced a student loan policy initiative, perhaps it is time to buy “student loan” as a search term. Such a candidate could use search to drive voters to a web site – and use look alike modeling to deliver ad content.

As voters engage, they will help refine your model and, if they opt in, they become part of your target.

4. Adapt your message for online. The expression of the candidate’s message and narrative is different online than it would be in television or a speech. A 30 second television ad buys emotional impact, especially since those who watch television will see it 20 times. Online ads are more about engaging a conversation – piquing curiosity first rather than creating a dramatic moment. Klobuchar’s recent video promotes her name – and also that she is smart, funny, practical and a mother. That is not her full message but it is an introduction.

Internet engagement is slower than television impact. Start now to build your narrative and the process will tell you a lot.

Your narrative will likely include three elements of message by the end: (1) that yours is the right candidate to take on Trump – because of their fighting personality, because of elements of the contrast, or because a lot of the same voters like them (although careful there). (2) they have an optimistic notion of what the future looks like; while true that we are all going to die, that doesn’t get people to want to talk to you more; and (3) personal intangibles that meet the moment – voters can’t tell you what these are yet but internet testing and modeling may help you figure it out.

Internet response and the results of your canvassing data stream can replace many traditional functions of polling because they can tell you who is attracted to a candidacy and some well-placed questions in the canvas data stream (or in a brand lift survey) can tell you about why.

One question all this won’t answer is who is ahead right now. However, until we can say who the electorate is, that’s not very answerable or interesting. And 75K committed but geographically dispersed supporters put you in the running.

5. Do you need a pollster? You certainly need someone in your campaign whose job is to explore what voters may be thinking, listen to them, figure out how to reach out to them, and quantify their response. That is the pollster’s traditional role: to focus on voters rather than the news cycle or Twittersphere and thus to save campaigns from isolating themselves within their own bubble. Your campaign needs that regardless of how research and communication techniques evolve.

Good luck!

People Do Not Want To Be Polled

The core problem with polling is that people do not wish to be polled. Those who answer their phones when the caller is unknown to them are unusual and atypical. And even many who do answer do not choose to complete the poll.

This year’s telephone polling results were closer to the final election results than in 2016. Much of the improvement, however, was in the nature of the mid-term electorate and not because the polls themselves were better. The mid-term electorate was highly polarized, and rabid partisans are easier to poll than voters in the middle. Polls were still wrong when those in the middle did not break proportionately to the partisans.

Back in the 1980s, polling achieved representative samples of voters by calling phone numbers at random. The definition of random is that everyone in the universe of interest (people who will vote in the next election) has an equal chance of being polled. With the advent of cell phones, caller ID, and over-polling, samples have not been random for a while – not since the last century anyway.

Pollsters replaced random samples with representative ones. Political parties and commercial enterprises have “modeled” files – for every name on the voter file, there is information on the likely age, gender, race or ethnicity and, using statistics, the chances that individual will vote as a Democrat or Republican. If the sample matches the distribution of these measures on the file, then it is representative and the poll should be correct.

There are three problems (at least) with that methodology: (1) there may be demographics the pollster is not balancing that are important; pollsters got the 2016 election wrong in part because they included too few voters without college experience in samples and college and non-college voters were more different politically than they had been before. (2) rather than letting the research determine the demographics of the electorate, the pollster needs to make assumptions about who will turn out to make the sample representative – including how many Democrats and how many Republicans. When those assumptions are wrong so are the polls. This year, conventional wisdom was correct and so the polls looked better.

The third problem is perhaps the most difficult and follows from the first two: pollsters “weight” the data to their assumptions. If there are not enough voters under 30 in the sample (and they are harder to reach) then pollsters count the under 30 voters they did reach extra – up weighting the number of interviews with young people to what they “should” have been according to assumptions. Often, however, the sample of one group or the other wasn’t only too small, but was an inadequate representation in the first place – a skewed sample of young people is still skewed when you pretend it is bigger than it actually was.

The problems can be minimized by making more calls to reduce the need to up-weight the data. If 30 percent of some groups of voters complete interviews but only 10 percent of other groups, just make three times the number of calls to the hard to reach group. That is what my firm and others did this year. It is, however, an expensive proposition and still does not insure that the people who completed interviews are representative of those who did not.

Next Post: The Self-Selecting Internet