Archive for July, 2009

Why Media Polls Frequently Get It Wrong

Friday, July 24th, 2009



Not long ago, a local TV station asked me to look at a political poll it had just commissioned.   The poll showed a well-known incumbent unexpectedly in trouble and they wanted an outsider review to make sure they weren’t missing something.

While I have commented frequently on media polls, it was the first time in 30-years of consulting that a news outlet contacted me to review a poll before they ran with it.   Their “top lines” – the numbers from all voters surveyed – accurately showed the incumbent had problems.   But, I simply glanced at those numbers before immediately turning to the “crosstabs” – the grey pages at the back of the poll which few people actually read.

Crosstabs are, in fact, the most important part of the poll.  They are the statistical tables that break out the details of how various demographic groups responded each question.  Using that data and knowledge of traditional voting habits and patterns, I showed the reporter that while their “top lines” may accurately report what respondents told them, the top lines didn’t really reflect how voters would behave once they entered the voting booth.  Drawing on an understanding of past voter behavior and other data, I pointed out why certain demographic groups were unlikely to behave according to their poll answers once in the privacy of their voting booth.

They looked like prophets when they reported these anomalies in their poll and just days later the election played out exactly like they predicted.

Media poll data is frequently accurate, but the data are frequently reported wrong.   Why?   Because the media typically reports the “top lines” of the poll without performing the hard analysis of comparing voter responses with voter history and other “predictors” of voter behavior.

First, polls are snapshots in time.   A poll in September is not valid for predicting November.   Remember the famous “Dewey Beats Truman” Chicago Tribune headline that Harry Truman held up on Election Night, November 1948?

The news media was overly influenced by a September poll of Readers Digest subscribers – a group that was then disproportionately Republican.   It was also in an era when polling was still unsophisticated.  Yet, it convinced the media that Truman would lose.

It was conducted before Truman crisscrossed the country during his famous whistle-stop train campaign and brought “home” certain Democratic voters who the poll indicated were straying from their party.  Dewey, equally convinced by the poll, curtailed his fall campaign trips and relied primarily on radio speeches to reach voters.  Thus, Truman eked out a win, much to the “Trib’s” embarrassment.   

Even today, when polling methodologies are more refined, challenges remain in properly reading poll results,   For instance, it may surprise you, but some people lie to pollsters.   

Yet, in spite of that fact, most methodologically sound polls are fairly accurate if you realize that poll responses are simply one aspect of discerning likely voter behavior.  Even the best polls require sophisticated analysis to predict voter behavior accurately.

In the 1998 Georgia gubernatorial race between Guy Milner and Roy Barnes, the Milner campaign conducted a poll showing its candidate with a seven point lead in September.   Everyone was trumpeting this as a major news item.

Later that week, pollster John McLaughlin was in my party office and asked if he could look at the poll sitting on desk.   He, too, ignored the “top line” data at the front of the poll and flipped to the statistical tables at the back.  After two or three minutes of examination, he announced, “Milner is down by five points.”

How could a professionally conducted, methodologically sound poll be off by 12 percentage points?  It wasn’t.  The data was correct; the analysis was wrong.

“Look at the black voter percentages for Milner,” he responded.   “They are at 17%.   They won’t vote that high for a Republican.   By election day, their numbers will drop to the normal single digit level and if you extrapolate that across likely voters, Milner’s down five-to-seven points.”

Those were almost precisely the election results two months later (okay, in this case a September poll did predict November.  Also, Milner’s pollster probably understood the discrepancies, but used the top lines to influence media coverage of the campaign).  
To use polls accurately in campaigns, you must compare what voters tell pollsters against what they typically do on Election Day.  It doesn’t mean voters won’t behave differently than they normally do in certain elections, but the analysis must determine which factors exist that will cause certain demographic groups to depart from their normal voting patterns. A properly analyzed poll will find those factors when they exist.

This kind of information is rarely found in the aggregated numbers that the press typically reports, but can be easily found by comparing the poll’s crosstabs with voter history and other past behavioral patterns from defined voter groups – white men, black women, college-educated Democratic females, blue-collar Republican males, etc.

There may be valid reasons at any moment in time when Republican males would be upset with a Republican incumbent.  That was the case in the TV poll I reviewed.  However, history shows that demographic is among the least likely group to vote Democratic in  November.

They almost always “come home” on Election Day.   I pointed that out in my off-camera analysis of the poll and, sure enough, when they voted, their anger took a back seat to partisan loyalty.

Ditto with black voters.   As the Milner example showed, they, too, almost always come home in November, meaning a 90+% Democratic vote.

Barring a watershed election where partisan loyalties shift, normal elections hinge on decisions by swing and independent voters as well as partisan turnout – which party’s voters are most motivated to vote.   While few call 2008 a “watershed”, it did feature abnormal elements.  Data from the 2008 election shows America’s most habitual voters - older whites - stayed home, while the country’s least likely voters - the under 30 age group - came out in droves.  Why?  McCain never excited his base, while Obama electrified America’s youth.  Shrewd pollsters were picking up those signals as the campaign closed.

These are the anomalies that pollsters look for to help them arrive at accurate predictions and these factors often are overlooked in media polls - and poorly drawn campaign polls, too.  

So, the next time you see a media poll that misses the mark in an election, you’ll look like a Frank Luntz-ean genius at the Waffle House when you tell everyone “their prediction was way off base because the media ignored some anomalies in the crosstabs when they reported their top lines.”

While no one there will understand a word of what you just said, they will all nod in wonder at your insights.   It might even get you an election morning invitation to the local Optimist Club every four years to predict the presidential election outcome.

At least, it worked for me.



Friday, July 17th, 2009

Microtargeting – the Latest Tool for Reading the Voters’ Minds

Friday, July 17th, 2009

Among the greatest challenges any campaign faces is deciding how to invest its limited resources – particularly candidate time and campaign funds.Targeting in campaigns is as old as politics. The goal is to determine who supports, who won’t support and who can be persuaded to support the candidate so resources aren’t wasted persuading the persuaded or attempting to persuade the non-persuadables. Campaign resources are most effective when concentrated on the unpersuaded who can be persuaded.

In today’s world, you also want to know if they are early voters, absentee voters or Election Day voters so you know when to turn them out.

Historically, the basic predictor of voting behavior was primary election participation. If a voter participated in four or five recent GOP primaries, they were labeled Republican leaning voters. If they voted in the same number of Democratic primaries, they were considered Democrat leaners. If they voted in a couple of GOP and a couple of Democratic primary – or only voted in general elections - they were considered swing voters.

How many primaries they participated in determined whether they were hard or soft partisans.

Voter history is a reasonably good predictor of future behavior, but it isn’t perfect. As campaigns became more sophisticated, candidates began combining demographic information with voting history to develop more precise predictors of voter behavior.

The problem with these voting indicators is it gave campaigns little indication of which voters were affected by different campaign messages. Soon, campaigns started incorporating their polling data into the voter analysis, assuming that if 80% of a district’s white, college-educated, executive/managerial males were opposed to increased taxes, the campaign should target anti-tax messages to that group of voters.

Soon, however, campaigns began asking what was different about that 20% of white, college-educated male population that weren’t anti-tax. They wanted to know what messages could motivate anomalous groups to vote for their candidate.

To answer that question, political consultants developed a new tool called microtargeting. Borrowed from current business marketing analytics, microtargeting is an diagnostic tool to help campaigns find the messages that best move the herd. Equally important, microtargeting helps identify subgroups of voters that don’t run with the herd and determine what motivates them, too.

Presidential and statewide campaigns have been using microtargeting for several election cycles. George W. Bush used it very effectively against John Kerry in 2004. By 2008, the Dems caught up and surpassed the GOP in their microtargeting sophistication.

For instance, the Obama campaign blended microtargeting with highly effective email, text messaging, viral marketing and social networking tactics to generate its massive advantage in capturing and turning out young voters.

Microtargeting helps campaigns answer the most important strategic questions they face. Who should we be talking to? Why should we be talking to them? What should we say to them? What media is most effective in reaching them? When should we talk to them (are they early voters, Election Day voters, absentee voters, etc.)?

In addition, microtargeting can pinpoint voters who are already with you or will never vote for you, so you don’t waste resources talking to the “saved” and “the eternally damned.”

Microtargeting incorporates a wide range of data, such as the kind of car you drive, where you shop, magazines you read, what you buy, racial, religious, ethnic, economic data – all information campaigns can purchase from credit card companies. Basically, microtargeting taps into the data which credit card companies and retailers like Kroger and Publix with shopper cards use to learn more about you than your mother knows.

Every time you swipe that credit or shopper card, some computer somewhere records all your purchases and categorizes you based on what you buy. They take your data and lump it with other people who bought the same items you purchased and make certain assumptions about you and your fellow subgroup members.

Campaigns are now doing the same thing. A campaign needs an ability to digitally store, archive, analyze and quickly retrieve massive amounts of data.

Then, you need savvy data crunching and sharp-eyed analytic capability working together to find the subgroups that most campaigns don’t understand or know about.

But with the right approach, microtargeting can be effective in finding groups of voters that are outside the norm and who have a hot button that motivates them to support your campaign and vote. These analytic tools are particularly helpful when the election is expected to be close or when you’re running in a district where the partisan trends are normally against you.

It’s not something a small budget campaign typically can afford, though software developers are beginning to produce programs that can handle many of microtargeting’s analytic chores. For example, Caliper Corp. developed a product originally designed to help state and local legislatures with redistricting efforts.

Realizing the campaign applications for this technology, the company released Maptitude for Precinct and Election Management™, and later released Political Maptitude™. Users can load the programs and add their voter files, consumer data or any figures they wanted to see displayed as a multi-faceted map. Versions of Political Maptitude™ even allow users to calculate how long it takes to walk a precinct and the hot issues in each neighborhood.

The ultimate goal of microtargeting is to let candidates understand the key issues that are important in each household within the district. Using widely available data, they can almost read the voters’ minds. Then, using that information to craft campaign messages precisely for individual households, they can figuratively place the candidate at every kitchen table in the district.

With newly emerging software packages and more powerful PCs, microtargeting will begin to show up in local elections during the 2010 and 2012 cycles. If your campaign isn’t prepared to explore this option, your candidate may be at a significant disadvantage to campaigns that do.

Microtargeting is not a silver bullet that guarantees success. It is simply a more sophisticated tool to do what campaigns have always done.

Yet, when it becomes more widely available, microtargeting local races will become a valuable weapon for campaign targeting and messaging.

Rusty Paul is former Georgia GOP Chairman, state senator and long-time campaign consultant for state and local campaigns.  He did his masters work in Campaigns & Elections at Georgia State University.  iSquared Communications is a political consultant helping Republican candidates in state and local races. Email him at

Grassroots Political Campaigning - Substituting People for Cash

Friday, July 10th, 2009

Grassroots campaigning is as old as American democracy.  Grassroots political campaigning was the driving force in America’s first contested presidential race, the 1800 match between John Adams and Thomas Jefferson. 

Partisans on both sides started newspapers to attack their opponents, influence uncommitted voters and mobilize their supporters.   Since that era’s culture prohibited candidates from campaigning for office, each candidate’s supporters enlisted thousands of regular citizens to do voter contact work.

Campaigns had rallies, parades and other activities that involved the general public in the process.   American grassroots campaigning took another step in sophistication during the 1828 race between John Quincy Adams and Andrew Jackson – a campaign that literally lasted four years and had one of the highest voter participation rates in our history. 

More than 15,000 campaign workers stormed the White House on Jackson’s inauguration day because they felt they had more to do with Old Hickory’s win than he did.

In the Tammany Hall and big city machine era, grassroots campaigning evolved into “ward politics” as parties organized down to the street block level to ensure every voter turned out.

In an age of paid media, Internet communications, political consultants and high-tech campaign strategies, grassroots campaigning is becoming a lost art.   But, low budget campaigns – particularly municipal and legislative candidates - can tear a page from history and create a real advantage against even a well-financed opponent with a well-organized grassroots effort.

Essentially, using Twitter™, Facebook™, MySpace™ and other social networks for voter contact is simply applying new technology to old campaign tactics.

Campaign consultants typically hate grassroots campaigns.  They want candidates raising money to fund more mail and paid media.  However, some campaigns – particularly challengers – simply lack access to big bucks and we’ve had success using grassroots tactics in low budget campaigns.  

Recently, we used modified grassroots tactics in a statewide challenger’s race to take down a well-entrenched, well-financed incumbent.   It was the only successful statewide challenger race in that cycle.

Grassroots campaigns replace cash with people-power.

The first thing a grassroots campaign needs is a plan and a good plan today is better than a perfect plan tomorrow.  After all, time is a precious commodity in a grassroots effort – particularly when dollars are scarce.   

Each second of campaign time must focus on maximum voter contact and the plan outlines how it that occur.  It will outline very precise tasks and the timeline for completing each one.


Start by determining how many votes you need to win. If a district has 150,000 voters and it typically has a 60 percent turnout rate, you need 45,001 votes to win (150,000*60%/50%+1).   Your plan should analyze past district voting history at least on a precinct basis so you can focus you efforts on areas where you are most likely to get those 45,001 voters.   


Next, people are the most valuable resource in a grassroots campaign - particularly volunteers.   The more the better, but you don’t necessarily need a cast of thousands.   Successful grassroots campaigns in small districts sometimes involve a dedicated, focused candidate’s family.

Of course, family and friends make great volunteers as do properly motivated party activists.    So, to attract volunteers, a good grassroots campaign must understand what motivates people to get involved.  

Some people are motivated by ideology.  These are typically the party faithful.   Others are motivated by issues, such as the environment, guns, abortion and local issues affecting their neighborhood.  

Identify groups in the district who support your issue positions and work to mobilize them. For instance, if education is a big issue, local PTAs may offer a great volunteer pool.  If neighborhood protection or property taxes are key issues, seek out homeowner associations or local taxpayer groups.

Others are attracted to campaigns as a social outlet.  Many people like the excitement of campaigns or the opportunity to meet new people.   This is typically a good way to get young people involved.

Some people get involved because they aspire to a career themselves in government or politics. Just remember, if you win – these volunteers expect help from you with their own aspirations.

In developing a volunteer base, there are two ways to go – top down or bottom up.   Top down approaches involve going to an organization’s leadership to seek help.  This includes asking for lists, invitations to meetings and other ways to enmesh your campaign with the targeted organizations.

The primary downside is you become associated with these groups in the public mind, so pick your groups wisely.  Yet, a grassroots campaign’s primary goal is to assemble a coalition that can generate the 45,001 votes you need.

If you want to avoid being beholden to a group’s leadership, you can always go “bottom up.”   Bottom up approaches require much more work.  You must rely on a few key members to help build lists and it involves significant follow up. 

Bottom up approaches also help if the leadership isn’t particularly keen on your campaign.   You can bring pressure on a group’s leadership from the rank & file.   Also, bottom up approaches allow you more control over the information going to the membership.

The biggest mistake a grassroots campaign makes is failing to ask people for help and not getting a specific commitment.  Your requests for help must be very specific.  Can you be at the headquarters at 6 p.m.?  Can you put up these 100 signs at the following addresses?   Can you call these 50 people and ask them these questions.   

You must also match the job to the volunteer’s skills and interests.   If they volunteered to put up signs and you have them making fund raising calls, your volunteer may disappear.  That doesn’t mean volunteers can’t multi-task, but not every volunteer can – or will.

Another mistake is not asking for enough.  Always ask for more and people will feel relieved if they opt for less.  But worst is not getting everything a volunteer is willing to give.   Make the tasks as simple as possible and ensure volunteers have everything they need for the job.

Also, volunteers want access to the candidate and they need to feel their work is meaningful.  So, don’t assign tasks and leave.   Spend time working with the volunteers.  If they see you stuffing envelopes, going door-to-door and making voter contact calls, they will sense you have assigned them a meaningful job instead of scud work.

Before sending volunteers out for door-to-door work, check out the targeted neighborhood to ensure it is safe and appropriate for door-to-door activities.

Finally, verify that tasks are done.  Tasks you thought were finished, but that are uncompleted will kill a grassroots campaign.   If a volunteer isn’t performing, give the job to a proven performer.   As Reagan would say, trust but verify.

In an upcoming edition, we’ll cover the activities that make a grassroots campaign successful.

When Political Figures Fall Short, It’s a Long, Hard Fall.

Sunday, July 5th, 2009

No one is perfect.   Virtually everyone has done something they do not want to see on the front page of their local newspaper.  

So, South Carolina Gov. Mark Sanford’s troubles are simply the latest reminder that high-profile people get high-profile treatment when they don’t live up to expectations.  

First, let me acknowledge my relationship with Gov. Sanford.   For almost a decade, we have served on the Awakening conference’s board together.  In January, Ms. Sanford moderated an Awakening panel on which I participated and annually our families have joined over 400 individuals for the annual Awakening gathering on the Southeast coast.

Like most others who know him, I considered him to be the straightest of arrows.

Yet, when confronting embarrassing situations, Mark Sanford’s handling of his ordeal offers a solid blueprint for how NOT to do it.    As The State newspaper characterized his press conference, for “the next 18 minutes he thought out loud.”
In other words, he mistook a microphone for a confessional.  It’s not the first time that a guilt-ridden, embarrassed political leader has poured his/her emotions into a camera and microphone.  However, it’s not a good way to manage a difficult, personal and political crisis.

First, let’s look at what he did right.   He faced the cameras alone He accepted responsibility for his actions and apologized.   He didn’t try to shift accountability or blame.   He acknowledged the pain and embarrassment he caused the State of South Carolina, its citizens, his family and his staff.  He said he would do everything possible to make amends.

That’s about it on the positive side.   Let’s use this situation as a primer for priority setting and decision-making in the initial moments of a crisis.

When dealing with emotional, personal information, you must first prepare very carefully what you are going to say and stick to the script.   Sanford arrived with no prepared remarks and hence spent 18 minutes thinking out loud – something even the most savvy, articulate individuals cannot do well when facing excruciating admissions.  

Anything over 10 minutes is too long …and five minutes or less is better. At a press conference, you are almost required to accept questions, but no more than two or three.

The statement should provide enough information to explain what happened and, if possible, why.   If you can make amends, discuss briefly how you plan to do that.  The answers to questions – like your statement - should be brief and concise, but not abrupt.  

This is not a time to let your thoughts wonder or give answers that would arouse additional speculation.  Avoid details – particularly the sordid kind – and get off stage as soon as possible.   If you feel a need to expunge guilt or discuss details in depth, find a minister or therapist – not a reporter.

Meanwhile, think about the impact your words will have on those closest to you.   Your family must be foremost in your mind.  Your actions not only embarrass you, but they pain and mortify your kin.

Don’t make your spouse suffer further by asking that individual to appear at the press conference or, for that matter, any public appearance while the controversy rages.  Do everything to create a zone of privacy for the family that reduces exposure to further injury.

Ultimately, they have to face their public, too, but time helps.   Also, don’t do or say anything else that adds to their discomfort – particularly if children are involved.   Adolescent kids can be earth’s cruelest creatures, so don’t provide more ammunition for those who would be unkind.

At the same time, resolve your situation quickly as possible.   Affairs of the heart, for instance, aren’t easy to resolve, but this isn’t a time to play mental ping-pong.  Develop a course of action that gets you out of limbo.

Next, think about the people who put their trust in you – the voters.   Can you continue to serve effectively?   If you decide to stay in office, is it for personal reasons or do you truly have the interests of the voters in mind?

If criminality, malfeasance or indictments are involved, you are probably fighting a losing battle to hang on to office.

Regardless, talk to people whose judgment you trust – not just those who depend on your position for their jobs.   Smart politicians keep people around who will tell them the truth – no matter what.   At times like this, these people are worth their weight in gold.

Not every mistake will disqualify you from office. More and more, people today seem to separate personal problems, mistakes and foibles from criminality when determining whether a political figure can stay in office – but not always.

Get a consensus from those trusted outsiders and let that inform your decisions.

Finally, time does make a difference.  Ultimately, this too shall pass – though don’t expect even those who forgive to forget.

As the Bible says, we have all sinned and fallen short.  But when high-profile people fall, the fall isn’t short.   It’s a long, hard drop. And, the impact is painful.


Rusty Paul is former Georgia GOP Chairman, state senator and long-time campaign consultant for state and local campaigns.  He did his masters work in Campaigns & Elections at Georgia State University.  iSquared Communications is a political consultant helping Republican candidates in state and local races. Email him at