Persuasion Visually – There Are No Laws of Persuasion

Failure lurks behind every success.

Nokia dominated cellphones then lost their mojo and are struggling now to survive. In 2007 Nokia held 40% market share. Now it is 21% and still spiraling down. Share price or profitability show the same pattern. Worse still over the time period of this continuing disaster Nokia spent over $40 billion on research and development. They were trying.

There Are No Laws of Persuasion and If There Were Why Would Anyone Tell You?

Death as a Persuasion Box and Play with the Future

Let’s consider this report that details two experiments manipulating thoughts about death and how people make charitable contributions. Participants were randomly assigned to get a Death Prime (a news story about a plane crash) or a Control Prime (a story about a mathematician solving a puzzle). They were then entered into a lottery for a $1,000 prize with a condition: They had to donate a portion of their winnings to a charitable organization. Enter now the second manipulation. Participants read descriptions of two charities, one providing immediate assistance to people in need, the other aiming at building lasting improvements to impoverished communities, thus creating a choice between Present or Future action.

So. People are Primed to think about Death or Not, put in a lottery for one thousand dollars, then required to make a contribution to a Present or Future charity. Clearly, the researchers are looking for some kind of interaction between the Prime and the Time Orientation of the charity. Turns out thoughts of death lead to a stronger Future Orientation. Consider the results.

Consistent with our predictions, results showed that individuals in the control condition allocated more money to the present-oriented charity (M = $257.77, SD = $249.79) than to the future-oriented charity (M = $100.00, SD = $173.21), t(20.38) = 1.95, p = .065; in contrast, participants in the death-prime condition allocated significantly more money to the future-oriented charity (M = $235.71, SD = $223.98) than to the present-oriented charity (M = $40.00, SD = $80.97), t(17.35) = 3.01, p = .008. In addition, individuals in the death-prime condition allocated more money to the future-oriented charity than did individuals in the control condition, t(24.19) = 1.86, p = .076.

A figure helps illustrate these Medium to Large Windowpane effect sizes.

The researchers replicate this outcome in a second study that also includes moderator variables to understand the process behind Death Primes and Future Giving. And, the effect sizes are similar.

Now. Where’s the persuasion in this? See the money contribution as a proxy for attitude or evaluation and not a direct measure. See messages about death and choices about time as a persuasion Box and Play. Death is a powerful situational manipulation and when people think about Death in the Local, it creates an interesting Persuasion Box. In this instance with free money from a lottery and a charitable TACT, people primed with Death think more about the Future than the Present.

Realize the potential limits here. The TACT involved free money for charitable giving. Buying a new car at your local dealership may be a different enough Local to kill this Box and Play. But, certainly any Local that involves donation, volunteering, politeness, social cooperation and on and on with those slightly self-sacrificing social interactions, Death primes build a particular Box. Consider a slightly unexpected application for this research – life insurance. No wonder so many people buy long term life insurance. When thinking about death, they focus on the Future.

If you are aiming at a Present TACT, avoid Death like . . . death. Alternatively, if the Other Guy is moving toward a Present TACT and you want Her to move to a Future TACT, prime up death, right?

My analysis of this interesting research goes beyond their theory building work with decision making in ways the researchers may not consider or approve, so see that I’m going way beyond the information given. However, It. Could. Work.

Kimberly A. Wade-Benzoni, Leigh Plunkett Tost, Morela Hernandez, and Richard P. Larrick. (2012). It’s Only a Matter of Time: Death, Legacies, and Intergenerational Decisions. Psychological Science, first published on June 12, 2012.

doi:10.1177/0956797612443967

A Persuasion Quotaphor

Consider the master of strategy.

“A small jump is easier than a large one, but no one on that account, wishing to cross a wide ditch, would jump half of it first.” Clausewitz, On War, Book VIII, Chapter 4.

Everyone who thinks that they are making the world a better place (yeah, all you health and safety knuckleheads) needs to contemplate this quotation from Uncle Carl. Let’s save the world over full strength soda pop, sitting, distracted driving with devices, vitamin supplements, dining education with the CIA, and on and on. Each is a small jump of knowledge that merely lands you in a wide ditch of attempted change. The Rules!

Persuasion Is Strategic Or It Is Not.

PostModern Persuasion 2.0 at Acxiom . . . Darkly

They’re back.

It knows who you are. It knows where you live. It knows what you do. It peers deeper into American life than the F.B.I. or the I.R.S., or those prying digital eyes at Facebook and Google. If you are an American adult, the odds are that it knows things like your age, race, sex, weight, height, marital status, education level, politics, buying habits, household health worries, vacation dreams — and on and on.

Thus begins another tough examination of the PostModern Persuasion 2.0 masters, this time at Acxiom. Big Data. Check. Big Statistics. Check. Always on. Check. Anonymous. Veiled. Dangerous. Check, check, check.

Maybe Acxiom is the Queen of Tomorrow. We’re living through the worst recession since the Great Depression, yet all the corporate customers of Acxiom and Their ilk are rolling in profit. While it’s not the crazy good of the Clinton 1990s or even the middle Bush 2000s, it’s still very good. Maybe Acxiom is the reason?

Or maybe Acxiom is selling smoke and mirrors.

Start with that name. Why do all the New New Thing companies spell like rapper acts and lyrics? Sure, it’s poetry and attention getting, but if you are the Queen of Tomorrow you don’t want or need something so obvious, transparent, so grasping as a clever name. Just HSM Corp. Something plain. Navy blue suit.

Now, consider the product.

Acxiom has its own classification system, PersonicX, which assigns consumers to one of 70 detailed socioeconomic clusters and markets to them accordingly. In this situation, it pegs Mr. Hughes as a “savvy single” — meaning he’s in a cluster of mobile, upper-middle-class people who do their banking online, attend pro sports events, are sensitive to prices — and respond to free-shipping offers.

Okay, let’s glide by PersonicX without a giggle and consider the Savvy Single, one of the 70 detailed clusters. If you’ve read Marketing for Dummies you know this as audience segmentation, a concept that is as old as Satan. Slice the Other Guys into categories, then pitch to the type.

That’s Queen of Tomorrow persuasion? Persuade types? You’ve got individual data down to keyboard fingerprinting and the best you can do is sort an Other Guy into one of 70 types? This is not your Father’s Oldsmobile persuasion, this is your Ancestor’s Sandal persuasion.

Nothing in any of the Queen of Tomorrow stories ever hits on the big thing: real time WATTage. Kids, when Acxiom combines its data with a good measurement of your WATTage, then maybe we’ll see the Queen of Tomorrow. Add in continuous assessment and alteration, and you can see Her shadow. But she’s not hanging on Acxiom’s arm. Consider the financials.

For Acxiom, based in Little Rock, the setup is lucrative. It posted profit of $77.26 million in its latest fiscal year, on sales of $1.13 billion.

The US GDP is about $15 trillion. Acxiom did $1 billion of business and earned about 7% on that as profit. Look, if I’m the Queen of Tomorrow with Persuasion 2.0 and I can only get 1 billion bucks for my skill in a 15 trillion dollar US economy, it’s time to storm the Queen’s palace. And, if I only earn 7 cents on a 1 dollar of sales, whatever I know about persuasion is wildly overshadowed by what I don’t know about business.

Of course, if Acxiom is really the Queen of Tomorrow, then all these figures are deceptive. There’s another set of books in a back room that show a different bottom line and it’s hiding out in Swiss bank accounts or whatever they’re using nowadays to hide wealth the way everyone once used Swiss bank accounts. This means, then, that the Acxiom mavens are living like smart bank robbers who just pulled the biggest heist in history. Sure, they’ve got a fortune but they have to live like mugs to avoid attention and detection.

You see the problem with Persuasion 2.0. If indeed you are the Queen of Tomorrow with a persuasion engine, you’ve got to hide away all the fruits of your success. Even though you control the world, you have to live like you don’t.

I’m exaggerating for effect. You don’t have to be Queen of Tomorrow good to have a genuine, world class, top of the heap New New Persuasion. You just have to be a little bit better than the Other Guys and your competition, right? So, look again at the financials on Acxiom. It’s publicly traded. Here’s the stock price for Acxiom versus the industry average over the past 10 years when all this Queen of Tomorrow Big Data New New Thing took off.

Not exactly best of breed. Acxiom got hotter than the others during the good times and then got colder than the others during the bad times. That’s not the performance curve of a master. And, I can make it look even worse by comparing Acxiom to the S&P 500 index over the same time period.

The doggy index, most certainly your Father’s Oldsmobile of stock performance, does better than ACXM. How much of a difference is all this Persuasion 2.0 really making?

Forget the dark hype from the NYT article. The writer’s a Useful Idiot whether she’s a believer or a paid shill. This is great PR for Acxiom, getting the Big Fear from the NYT. Acxiom comes off as a first order Super Villain of Big Marketing. They just need a chartreuse costume for that Queen of Tomorrow performance. Something like this.

Here we are again, selling sand to Sauds, ice to Aleuts, and persuasion to persuaders. Nothing wrong with a little smoke and mirrors with your persuasion, but that’s not persuasion, of course, just smoke and mirrors. And, this is not unique with Acxiom or probably any other Persuasion 2.0 operation. The Count never adds up to the Change.

I’m a gullible guy despite my public performance as a persuasion maven. I believe in the Queen of Tomorrow and I will know her when I see Her. She won’t look like Julianne Hough on the red carpet at the Rock of Ages movie premiere to me . . . just to everyone else.

Summertime Blues Turn To Fear The Reaper

With all the caveats, exceptions, and nuances noted before, I tread carefully and observe the latest with Team Obama.

Again, this is reporting from the outside of a persuasion campaign so you don’t know the strategy, the TACTs, the timing, the traps, and on and on with the art you never see in great persuasion. This could be the persuasion play called The Leak. Or this could be a good, old fashioned Fear Appeal. I’ll go with what’s in front of my face while looking frantically for curtains.

As I’ve noted with health and safety persuasion, fear appeals is one of the more dangerous persuasion plays you can pull. A good fear appeal produces Small effects, a Windowpane around 45/55 (go look up Witte and Allen for a meta analysis). Most fear appeals are not well done and often take at least one step back for every two steps forward – just read all that awful Health Beliefs Model research. As Kim Witte’s work proves, when a fear appeal elicits an efficacious response – you can do something to kill the threat – everything works to the good. But, sometimes that threat merely produces pure fear and the Other Guys freeze in place ruminating over the fear and not about removing the threat.

Team Obama is clearly aiming their fear appeals toward an efficacious response. They ask in many of these emails for just $3 in contributions, for example. Virtually anyone can afford that. So, if you believe the threat and you believe that $3 kills the threat, then this fear appeal is effective.

Yet, at the same time that Team Obama is providing that efficacious response to their Other Guys, they are also looking weak, vulnerable, and scared with the fear appeal. How can a likeable incumbent President possibly get outspent? He’s had four years to pile up money and build more supporters, yet here he is, publicly worrying about losing because he doesn’t have enough supporters or money. And, if things are so bad that he’s expressing his fear in public, is a $3 donation as efficacious as he says it is? Shouldn’t it be $30 or $300 or $3000 to beat those Evil Rich Republican Bastards?

You see the problem with fear appeals. They can easily generate off-topic responding with both negative cognition and negative emotion. Fear appeals are like shotgun bursts that scatter shots across a wide range. You simply do not have as much control with a fear appeal compared to a wide variety of other persuasion plays. Jeepers, just look at every other CLARCCS Cue or any good Central Route Argument play.

Many persuasion plays hide in plain sight and the Other Guys never know they are out there. Thus, when those plays fail, you don’t compound the failure with public ridicule because no one saw the failure. Team Obama is playing their persuasion in plain sight, always a dangerous game. (Unless, of course, there’s a curtain I’m not seeing.)

Just a few weeks ago I warned about the Summertime Blues for Mr. Obama, but advised he still had the Fall to get his persuasion mojo going. Now, it looks like his blues are turning to fear that the reaper will take him. He needs to listen to Blue Oyster Cult (YouTube) and consider the possibilities.

P.S. But not More Cowbell.

Persuasion as Universal or China Goes Vogue

Advertising (understood as applied persuasion) is doing great in China.

Late last year, Cosmopolitan editors in China started splitting its monthly issue into two magazines because it was too thick to print. Elle now publishes twice a month because issues had grown to 700 pages. Vogue added four more issues each year to keep up with advertising demand. Hearst is even designing plastic and cloth bags for women to easily carry these heavy magazines home.

But given the vast cultural differences between East and West you know that the persuasion for advertising in East and West must be different, too. Like this.

Persuasion principles are universal for all faces and places, times and rhymes. You just change the wrapper.

Braking While Accelerating with Fear

We’ve seen the Audacity of Hope with the current CDC anti-smoking media campaign that will persuade at least 50,000 smokers to quit according to the public guarantees of the CDC director. I see this effort as the Audacity of Hope because there’s no science and only Wishful Thinking, which, if you think about it, is another name for the Audacity of Hope. No one could prove with anything other than scientific science that 50,000 people will quit smoking from a national ad campaign because 50,000 is within the measurement error of any national evaluation scheme. Past this propeller head, bean counting, I Understand Little Greek Letters Better Than You, there’s another reason the Fear Appeals campaign won’t work: Reactance!

Recent research provides insight into the trouble with Fear Appeals. While Fear Appeals may produce Small Windowpane change, along the way Fear Appeals produce Reactance. Stated in a driving metaphor, Fear Appeals hit the gas pedal and the brake pedal at the same time. You do move forward, but not as far as all the gas you burned because you also hit the brake along the way. The gas of Fear tends to produce the brake of Reactance, that sense of a perceived unfair restriction on thought or action. Let’s look at a simple experimental example.

Participants were randomly assigned to watch a series of 4 anti-smoking ads that used either Fear, Empathy, or Control messages. Fear messages emphasized threat, risk, and damage to you as the smoker. Empathy messages showed people suffering from the loss of smoking parents or friends and emphasize their pain, worry, and distress. Control messages simply raised specific problems with smoking with no emotional appeal. Thus, each person saw 4 examples of Fear or Empathy or Control messages. They then rated the ads for their impact on emotion (fear, empathy, reactance) and persuasion (attitude and perceived effectiveness).

Results demonstrated that Empathy messages produced more favorable persuasion outcomes than either Fear (Medium Windowpane, 35/65) or Control messages (Large Windowpanes, 25/75). More interestingly, the research provided evidence that the Empathy effect was stronger because Fear appeals generated more Reactance (again, about a Medium Windowpane). More interesting still, the results indicated that while Fear appeals generated higher Reactance, the Empathy messages generated lower Reactance. From this you can infer that Fear messages activate Reactance problems, while Empathy messages will tend to put Reactance to sleep.

This research from Lijiang Shen in Health Communication demonstrates the persuasion of emotional appeals, but provides considerable insight into the variable ways different emotions operate. Shen replicates the standard finding that Fear appeals do produce favorable persuasion and at the same, Small Windowpane that various meta analytic studies of this play have reported. Shen extends this knowledge with the inclusion of other emotional and rationale persuasion plays in the form of either Empathy or just plain Strong Arguments and, most importantly, the measurement of Reactance. All these comparisons reveal the limitations of Fear appeals.

Yes, Fear appeals can produce desired change, but with considerable costs and inefficiencies. It’s the proverbial two steps forward, one step back. You get some gains, but also have to take losses that your persuasion play produces as part of its proper application. Thus, Fear appeals are like a pill that does work but also comes with bad side effects. If you’ve got nothing else, then take the pill, but with persuasion you’ve got a lot of other plays you can make, including emotional plays, that work better and cheaper than Fear appeals.

All this nuance is lost in a health and medical community that is stuck with a primitive understanding of Reinforcement Theory. Whack the Other Guys with the Stick of Fear, baby. That’ll do it, baby. Or jam fearful, emotional responding into that failed, disconfirmed, and evergreen Health Beliefs Model, none of that Your Grandfather’s Oldsmobile Reinforcement Theory, baby. Fear marks the muggle, the innocent, the earnest, the sincere, the authentic, in other words, the failed persuader.

L. Shen. (2011). The effectiveness of empathy- versus fear-arousing antismoking PSAs. Health Communication, Jul-Aug;26(5):404-15. Epub 2011 Jun 29.

doi: 10.1080/10410236.2011.552480

Ask Intentions to Pop TACTs

You look through descriptions of 5 new brands of candy with different taste ratings, calorie counts, and fat content. After considering this information you answer one of two questions.

1. How likely or unlikely would you be to try the presented candy bars if they were available to you?

2. How positive or negative are you about the presented candy bars becoming available to you?

The first question addresses your Intention while the second asks your Attitude. Does one question drive future behavior better than the other? Consider the key results analysis from three different experiments.

Significantly more participants who answered an intention versus an attitude question chose their most preferred brand (76.4% vs. 56.8%, χ2(92) = 3.94, p = .047)

A separate chi-square analysis for the first choice decision indicates that participants who respond to an intention question are more likely to select their most preferred choice option than are participants in the control condition with an attitude question (80.5% vs. 54.8%, χ2(83) = 6.256, p = .012) or the control condition with no additional question (80.5% vs. 45.7%, χ2(87) = 11.174, p = .001).

The most preferred choice option was chosen more often in the intention question condition (82.1%) than in the attitude question condition (53.7%; χ2 (238) = 21.817, p < .001).

Across three different studies with three different samples of participants, we obtain an consistent near Medium effect size, a Windowpane of about 35/65. Remember that these are experiments so people are randomly assigned to the Intention or the Attitude question and that’s the key independent variable. Furthermore, the effect size is pretty obvious, practical, and apparent. And, these results are consistent with related persuasion concepts like Implementation Intentions.

Merely asking the Other Guy about her intention makes the intention-related TACT more likely to occur than when you ask a question about attitude. Kerckhove, Geuens, and Vermeir provide much more than this in their multi-study package in the Journal of Consumer Research. In particular they test the cognitive processes behind this effect and argue that activated intentions trigger the related TACT while attitudes and other attributes do not. Thus, cognitive activation not only makes memory more salient but also the behavior routines associated with that memory.

The Kerckhove team provides another great demonstration of a full cycle testing of a persuasion effect. They not only manipulate the intention effect on the TACT, but they demonstrate how this occurs. If you just enjoy reading a well done and focused research program, this is an outstanding example. And, if you just want a practical demonstration of How-To, this research is what you want. Contact the authors and get a copy.

Hey. After reading this post, how likely is it that you will read another post in the Persuasion Blog?

Anneleen Van Kerckhove, Maggie Geuens and Iris Vermeir. A Motivational Account of the Question-Behavior Effect. Journal of Consumer Research, epublication ahead of print.

doi: 10.1086/661936

Persuasion Joins Science to Journalism

Journalism, like all professions, seeks a New New Thing and the current New New Thing for journalism is writing the first draft of history with science. I quoted in an earlier post,

By enabling anyone to drill down into data sources and find information that is relevant to them, as well as to verify assertions and challenge commonly received assumptions, data journalism effectively represents the mass democratisation of resources, tools, techniques and methodologies that were previously used by specialists — whether investigative reporters, social scientists, statisticians, analysts or other experts. While currently quoting and linking to data sources is particular to data journalism, we are moving towards a world in which data is seamlessly integrated into the fabric of media. Data journalists have an important role in helping to lower the barriers to understanding and interrogating data, and increasing the data literacy of their readers on a mass scale.

That last sentence in particular scares the hell out of the scientist in me because Big Data without science is just jibber-jabbering with numbers. Big Data Journalism must also aspire to some kind of science as it lowers barriers and interrogates data while increasing data literacy of the masses. Consider a good and, at the moment, innocuous example with the New York Times and the FiveThirtyEight Blog by Nate Silver.

At first glance this post from Silver on Do Presidential Polls Break for Challengers? looks and feels a lot like something you’d read in Public Opinion Quarterly or American Political Science Review. Just take a second and scan over the post. Look at all the charts and the regression lines, the beta weights, and tables of numbers, colors, and blocks. Then just glaze over the words with the Brooks Effect and you can see the strong resemblance to a peer reviewed article in an academic journal. And, yet this is in the New York Times, not Public Opinion Quarterly.

Silver himself presents an exemplar of the New New Journalist doing the Big Data Journalism/Science. Read his wiki entry. UChicago economics, quirky skill as a sabermetrician, and qualifications as a psephologist. The guy has elite training in statistics and speaks Numerosity with equal fluency in Baseball or Election statistics. Plus, he’s a good writer. What more can you ask?

Let’s start with a punch almost no one sees about scientific writing: Peer Review.

The hallmark of scientific publication is getting through peer review. You submit your jibber-jabber with numbers to an editor selected from your field of interest who then sends your jibber-jabber with numbers to two to four researchers with proven competence in the field who provide a blind, anonymous review of your jibber-jabber with numbers. If your work survives that review, then the jibber-jabber with numbers becomes something different and joins the peer review literature.

Nothing in Nate Silver’s work, nothing in the New York Times daily publication, and nothing in Journalism – whether from ink stained wretches or Big Data Web 2.0 digirati – comes close to the peer review process. At best an editor and a copy editor read the work for its general fitness and style sheet compliance. If legal bells ring, perhaps a lawyer looks over the submission. But, at no time does journalism conduct a standard peer review over the work it publishes.

Silver’s particular post on Polls Breaking For Challengers may be great science. It looks pretty good to me as a guy with experience in peer review, but I’ve never written or reviewed for political science and thus am not a qualified reviewer for this kind of research. I have a lot of experience with campaigns, statistics, and read widely in politics, but I’m still not qualified to pass judgment on the scientific value of Silver’s work.

My point is that as good as this work appears to be, especially within the confines of your daily information fix over your favorite morning beverage is by the lights of science, nothing but jibber-jabber with numbers. Yet, I think that it qualifies as an A+ example of the New New Journalism with Big Data.

It’s a bit odd to compare Nate Silver’s work in the NYTimes with another writer at the Wall Street Journal with a blog, but who has the scientific bona fides, Dan Ariely. Here’s a recent post from Ariely.

Dear Dan,
What is the best way to get myself, and other citizens, to be more motivated to vote in elections for public office? I realize this is a loaded question, but there must be some behavioral component behind the motivation and initiative required to vote.
—Randy
The best way to get people to vote is to get them to the voting place for another reason (free ice cream, for example) and then to make the extra effort needed to vote as minimal as possible. This is not, of course, an answer to your question, because it tells you how to get people to vote, not how to get them motivated to vote.
For people to be motivated to vote, they would need to care about the outcome of the election and the people whom they are voting for—and that is a very big challenge.

No numbers! No regression lines with intercept and slope! No statistical significance! Ariely has the scientific credential, but he writes like Dear Abby in the WSJ blog. Compare that to Nate Silver, a guy with no doctoral degree, no graduate training, no academic work experience, no peer review publication, no listing as a peer review editorial board member. Yet, Silver writes in his blog the way Ariely writes for Psychological Science.

Please catch the point here. I think both Silver and Ariely are doing excellent work in everything they do. They break no Rules of Persuasion and in fact each follows Them in their own way. What I’m concerned about is the NYT and other journalism sources faking science as a survival strategy. They present people like Silver as doing science while letting scientists like Ariely do journalism.

In much the way preachers admonish that you cannot serve both God and Mammon, people need to realize that you cannot serve both journalism and science. You’re just persuading yourself when you try.

Making Twitter Relevant for Elections

Twitter has a new feature, Twindex, invented for the Presidential election. It tracks daily user sentiment toward Barack Obama and Mitt Romney based upon the tweets sent about either man. Here’s the chart for May 2012 just to illustrate.

If you look closely at Twindex you can see that Obama has a lot more Twitter followers than Romney, but the Twindex as it is called clearly adjust for the base rate here. Thus, the sentiment score for either man is scaled relative to the number of followers he’s got, otherwise Obama’s scores would always be more extreme than Romney’s. Of course, adjusting the base rate does not obscure the problem: one guy has 10 million+ followers while the other has less than 1 million.

This illustrates the problem with a biased sample of data. Sure, Twitter has over 400 million tweets a day, talk about Big Data! But, there’s a 10 to 1 difference on this particular issue, a Windowpane approaching 99/1, a Colossal effect size. How can you possibly de-bias a sample of data even with so many data points when clearly the sample is dominated on one dimension? The smart guys who do political polling are on this point.

Pollsters are very interested in social media analysis because of the volume of opinions expressed, says Frank Newport, editor in chief of the Gallup Poll. “It would be foolish not to examine the value of these millions of opinions. But there’s a big ‘but,’ ” he cautioned. Twitter users, 140 million worldwide, aren’t a representative sample of the general population — for instance, they tend to be younger and, obviously, more technologically savvy. “My concern is, who are these people?” Newport says. As a result, “I don’t think anyone has figured out whether it’s truly useful.” In other words, if a Twitter analysis can’t predict who is going to win the election, what’s the point?

Not only is Twitter Colossally filled with Obama followers compared to Romney, among both groups the Twitter users are probably different from the larger universe of Likely Voters. This Twitter Twindex is doubly biased, first, with users being different from nonusers, second, users being biased for a particular person, Obama.

Biased Data is never a Good Thing. Never. You will always get fooled in ways you cannot anticipate when you believe Biased Data. Consider only the Twitter example.

If you take just a moment to visit the Twitter verse, you quickly overwhelm yourself with Obama, Obama, Obama! Everywhere you turn you will find Twitter evidence of Obama, Obama, Obama! And even if none of it is manipulated, bought, or persuaded and all of it is the deep, abiding, and unchangeable sincerity of the twitterati, it all shouts Obama, Obama, Obama!

Anyone living in Twitter will believe Obama, Obama, Obama when even Mr. Obama and his Team are currently sweating blood, bullets, and bile with a race that should be ripe for the taking with a likeable incumbent, but yet shows a dead heat among likely voters. Living in Twitter will lull those voters into thinking this election is a Sure Thing for their man which will translate into less Enthusiasm meaning, less money, less volunteering, and, worst of all, less voting. When you’ve got the Sure Thing, fuggitaboutit.

If I’m running Team Romney, I’m planting stories about invincible Twitter and doing everything I can to put the Twitterverse to sleep on Election Day. Please believe what you tweet or retweet. Please. Here. Lie down by the shady tree of Twitter near the river of Facebook and enjoy the triumph of like minds living in the echo chamber. Twitter and Like all day and read the Twitter Election Prediction Tracker that shows Obama, Obama, Obama on Election Day . . . until the polls close.

If I’m running Team Obama, I’m trying to shut down Twitter or at least scare the living daylights out of the twitterati. None of these guys are getting the memo that this is a close election. Obama does not want to win the Virtual Election; he’s already done that as the Twitter follower numbers prove. And, he needs to convince these digital dimwits that they are only winning the Virtual Election. They’ll learn the difference when Romney takes the White House.

Past the political lessons, persuasion mavens, see the Social Media possibilities. Gain your TACTs on Web 2.0 with Obama Blue plays. If you haven’t figured this out already, Web 2.0 is the progressive playground and talk about easy, ripe, and luscious. Sure you can sell those t-shirts and mugs, but think about priming. Take progressive words and turn them to your products and services as Big Green does with True Green. The Twitterati run on Low WATT Cues or High WATT Blue Bias; they are not willing and able to think objectively in situations of uncertainty, ambiguity, choice, doubt, or controversy. And they have neatly collected themselves in one place through one channel all the time. If you can’t hit that pitch, get a new game.

P.S. Push this example out to other Big Data cases as with those Evil Marketing databases or even the Good Global Warming databases with billions of data points. They are Big, but they are all Biased. Much of the Marketing Big Data from the New New Thing of Social Media from Web 2.0 flows from all those young early adopters of a new thing; all the old guys and resistors are not to be found on Facebook or Groupon or Foursquare and on and on – that’s bias in the sample. Less obviously with those Big Data Global Warming That You Caused datasets, you need to read the fine print in the methods and results section to see how strange, unique, and unrepresentative the various batches of data are.