Showing posts with label math and numbers. Show all posts
Showing posts with label math and numbers. Show all posts

Saturday, December 3, 2016

The Monty Hall problem: Amy Santiago is right. You should switch doors.

Oh Amy, you're such a funky cat with your feisty stats, and I love you for that.

Let’s talk about the Monty Hall problem, as featured in the most recent episode of my favorite TV show of the moment, Brooklyn Nine-Nine.

In the episode, the precinct’s captain, Captain Holt, is in an argument with his husband over the Monty Hall problem, and one of the detectives, Amy Santiago (who, incidentally, is my fictitious dream woman, because she is wonderful and adorable and I am apparently a dude who crushes on TV characters), is trying to figure out a way to explain the correct answer.

The Monty Hall problem, named for the original Let’s Make a Deal host Monty Hall*, is as follows: You’re on a game show trying to win a car, and the host asks you to pick one of three doors. Behind one of the doors is the car; the other two contain a goat. After you pick a door, the host—who knows which is the car door and which are the goat doors—will open an unpicked door revealing a goat and ask if you want to stay with the door you picked or switch to the unopened door you didn’t pick. Should you switch doors?

The answer is indisputably yes, you should switch. By switching, you double your chances of winning the car—staying gives you a 1 in 3 chance of winning the car; switching improves your chances to 2 in 3. It may be counterintuitive, but you can test this out by playing repeated simulations of the problem, either in real life or online. The result, with enough simulations, will be the same: if you stay, you’ll win the car about 33 percent of the time; if you switch, you’ll win about 67 percent of the time.

Holt is unconvinced, thinking that, once it’s down to two doors, it’s simply a 50/50 chance and switching doesn’t affect anything. Amy, who’s the sort of person who gets excited over weekend math conferences called Funky Cats and Their Feisty Stats (again: fictitious dream woman like whoa), is determined to explain why switching is the correct strategy.

Unfortunately, we never get to hear Amy’s explanation because it turns out the spat wasn’t really over the math problem but rather a dearth of boning. So let me take a stab at it.

In the Monty Hall problem, there are exactly nine scenarios that can play out: you picking Door 1 and the car being behind Door 1, 2, or 3; you picking Door 2 and the car being behind Door 1, 2, or 3; and you picking Door 3, and the car being behind Door 1, 2, or 3. They're organized in this chart:

As the chart indicates, you win the car by staying in only three of the nine scenarios. You win the car by switching in six of the nine scenarios. Thus, staying yields only a 3 in 9 (i.e., 1 in 3) chance of winning; switching yields 6 in 9 (i.e., 2 in 3) chance of winning. Thus, you should switch—it doesn’t guarantee the car, but it significantly improves your chances.

Another, perhaps less intuitive, way of thinking about it: let’s say you pick Door 1. When the host asks you to switch, he’s asking if you’d like to change your pick from "Only Door 1" to "Either Door 2 or Door 3"—in other words, asking if you’d like to change your chances from "1 in 3" to "2 in 3." The chance of your door being a car doesn’t magically increase once the goat door is opened; it remains 1 in 3, so the alternative choice is 2 in 3.

It becomes a little bit clearer if, instead of three doors, we’re playing with a million doors. When you first pick a door, you have a 0.0001 percent chance of winning a car. When the host opens 999,998 doors revealing 999,998 goats, you’re left with your door and, say, Door 784,912. You intuitively know to switch—it’s highly unlikely that you picked the right door on your first try, and Door 784,912 just looks so appealing sitting there all by itself. In essence, the offer to switch is asking if you’d like to stick with your 0.0001 percent chance of winning, or if you’d like to take the 99.9999 chance that your initial pick was wrong. Even more simply, the offer to switch is asking, "Do you think the car is behind the one door you picked, or any of the other 999,999 doors?"

With a million doors, switching makes sense. And the math stays true with 500,000 doors. And 10,000 doors. And 50 doors. And indeed, all the way down to three doors.

If the whole doors thing makes this a bit opaque, consider this mathematically identical problem: let's say I'm thinking of a (whole) number between 1 and 1,000,000. Once you pick a number, I tell you that the number I'm thinking of is either the number you picked or 784,912, and I ask you if you want to switch to 784,912. Of course you'd switch—intuition dictates that it's highly unlikely you just happened to pick the correct number out of a million, and switching just makes sense. Clearly, once you're down to two numbers, switching does affect your chances significantly; you're going from having a 1 in 1,000,000 chance to a 999,999 in 1,000,000 chance. The concept remains the same if I'm thinking of a number between 1 and 1,000,000 (99.9999 percent chance of winning if you switch), 1 and 10 (90 percent chance), 1 and 5 (80 percent chance), and, yes, 1 and 3 (about 67 percent chance).

Finally, if all of that is still unconvincing, here's maybe the simplest explanation. If you stay with your door, the only way you can win the car is if you initially picked the correct door, which has a 1 in 3 chance of happening since there's only one correct door. If you switch, the only way to win is if you initially picked a wrong door, which has a 2 in 3 chance of happening since there are two wrong doors. Basically, you probably picked the wrong door initially, so staying means you're probably going to stay with the wrong door, while switching means you're probably switching to the correct door.

In conclusion, you should definitely switch doors because Amy is right. Also, Amy is totally dreamy, even when she's shame-eating hamburgers.

God, I need a girlfriend. Or Season 3 on DVD, either one's fine.

*It’s worth noting that, on Monty Hall’s Let’s Make a Deal, this situation as described never came up on the show, and thus, the name of the problem is kind of a misnomer. It's like having a problem about funny jokes, and calling it the Big Bang Theory problem. Bazinga! (For real, that show sucks.)

On the other hand, Monty Hall himself was able to come up with the right answer to the problem pretty quickly—a lot faster than many PhDs and other math experts, in fact—so maybe he deserves to have the problem named after him after all.

Sunday, February 8, 2015

Slovakia’s gay marriage referendum: an exercise in game theory and ballot design

A genuine Slovak portable ballot box.
(Photo: LiĊĦiak/Wikimedia Commons)
Voters in Slovakia rejected three referenda asking voters to ban same-sex marriage, adoption by same-sex couples, and compulsory sex education yesterday.

Referenda in Slovakia are only legally binding if 50 percent of eligible voters—in this case, 2,205,765 out of 4,411,529 voters—actually cast a ballot. So while more than 90 percent of people who voted on the ballot questions voted “yes,” voter turnout was less than 22 percent. With insufficient turnout, the referenda won’t take effect.

This was the strategy adopted by LGBT activists in Slovakia: rather than asking voters to vote “no,” they urged voters not to cast a ballot at all. It’s a sound strategy, because—assuming everybody who wants to reject the referenda is on the same page—voting “no” can only improve the chances of the referenda passing.

For convenience’s sake, let’s say there are exactly four million eligible voters in Slovakia. If 1,999,999 voters vote “yes” on a referendum, and nobody votes “no,” the referendum is rejected due to insufficient turnout.

But if 1,999,999 voters vote “yes” and one person votes “no,” the referendum passes.

In fact, the theoretical smallest number of “yes” votes required for passage in this example would be 1,000,001—assuming 999,999 people voted “no.”

In the case of this past vote, it looks like Slovaks who wanted to reject the referenda were on the same page and mostly stayed home instead of voting “no.” But there’s an interesting bit of game theory involved with deciding whether to stay home or to cast a “no” ballot:

Assuming you think your fellow referendum-rejecters are completely rational, it’s obviously best to stay home and not vote at all.

But if you think enough of your referendum-rejecters haven’t thought it through and will show up to vote “no”—enough to push turnout over 50 percent but not enough to make up 50 percent of those who do vote—then you need to show up to vote “no,” so that the “no” vote has a fighting chance of winning.

But then if you don’t necessarily think referendum-rejecters are stupidly going to vote “no,” but you think enough referendum-rejecters think that enough referendum-rejecters are stupid enough to vote “no” and will subsequently vote “no” themselves, then you need to vote “no” as well. Which is bananas, right?

And there’s another wrinkle to this, too. These three referenda are bundled together on the same ballot. If blank responses—that is, checking neither “yes” nor “no”—count toward the turnout tally, that may explain why those who voted “no” did so: it’s not stupidity, but rather, a split ballot. They may have very strongly wanted to vote “yes” on one question (say, banning compulsory sex education) but weren’t in favor of another (like banning same-sex adoption). Since they’re contributing to the turnout even if they leave the adoption ban question blank, they decide to vote “no.” Which is yet another factor in the game theory calculus for referendum-rejecters—are there enough ballot splitters to push turnout over 50 percent?

In this particular ballot, the questions deal with related topics, and there’s a good chance that “yes” voters will vote “yes” for all three, and “no” voters will do the same. But there’s plenty of opportunity for monkeyshines in this setup—what if the questions were phrased differently, like asking voters to ban same-sex marriage (so that the “no” vote is the pro-LGBT vote) in one question and approve same-sex adoption (so that the “yes” vote is the pro-LGBT vote) in another?

Or what if they bundled together unrelated questions that could scramble voting coalitions? I don’t know anything about Slovak politics, but if this referendum system were in place in the United States, it’d be hard to figure out the correct strategy if the ballot questions concerned, say, mandatory vaccination, gun control, and marijuana legalization.

In any case, it highlights a quirk of democracy—that the way you let people vote can play a huge role in how that vote ends up. An ideal voting method is one in which no game theory is required as a voter; you merely express how you feel without worrying about what other people are doing.

But that rarely happens. As a result, it’s worth remembering that who’s designing our ballots can be just as important as what’s on them.

Friday, May 13, 2011

Dove's hypocritical campaign for its “Go Sleeveless” deodorant creates insecurites

This post originally appeared on my old Joe Dellosa on Advertising blog.

► Dove is trying to invent a new, fake insecurity for women (unattractive underarms?) in order to shill its new Go Sleeveless Deodorant, completely blowing whatever credibility it had as a trustworthy brand that cares about the self-esteem of young women.

Last month, Dove (the beauty brand owned by Unilever, not the chocolate brand owned by Mars) made a stir with its Dove Go Sleeveless Deodorant. The stir was not a positive one.

The deodorant—whose name is officially stylized as “Dove go sleeveless Deodorant,” with lowercase letters to emphasize casualness of going sleeveless or something—purports to beautify women’s underarms. In addition to its duties reducing underarm odor, Dove promises that Go Sleeveless will provide women with softer, smoother, and more attractive underarms in just five days.

According to a Dove press release dated April 4, Go Sleeveless was inspired by a 2010 study conducted by Unilever with market research firms Omnibus and TNS that supposedly found that 93 percent of women think their underarms are unattractive. Dove to the rescue, right?

Well, no. Many in the media quickly (and appropriately) called shenanigans on Dove’s new deodorant and its associated marketing campaign: Slate called it yet another instance in a long line of marketing efforts designed to foment fake insecurities in women and profit off selling them a cure. American Public Media’s Marketplace did a couple of person-on-the-street interviews with women who expressed no pre-existing concerns about their underarms.

And Stephen Colbert, who’s one of the media’s most astute advertising critics, had the definitive takedown of Dove during the April 13 episode of The Colbert Report. Colbert said “women have now learned that their armpits are hideous” with Unilever’s help and called the Dove’s new deodorant a “breakthrough shame-o-vation.”

Their points are all well-taken, and this should be pretty embarrassing for Dove. After all, even if we give Dove the benefit of the doubt and say that there are probably some women who do feel concerned about the appearance of their underarms, it's audacious for Dove to present its product as a solution to some sort of universal problem affecting virtually every American woman.

What hasn’t been mentioned quite as often is why this should be very embarrassing for Dove: with this advertising campaign—a campaign predicated on shamelessly fabricating a fictitious anxiety and foisting it upon women—Dove has completely blown its brand identity as the beauty brand that cares about the self-esteem and emotional well-being of young women. Oops.

 Cam-pain for real beauty

The last time a Dove campaign got any real attention was its much-lauded and award-winning Campaign for Real Beauty, launched in 2004. In the campaign, Dove criticized our culture’s perception of beauty as unrealistic and damaging to young girls with smartprovocative ads showing the artificiality of the beauty industry.

Can't see the ad? Click here to watch it on YouTube.

In other ads, Dove hired models more representative of average women, specifically eschewing thinner, perfectly airbrushed models. (Dove later got criticized when it turned out those models' photos were retouched, just like any other model's photos.)

But even without word of the entirely-unsurprising Photoshop job, it’s easy to see why Dove launching a Campaign for Real Beauty is at least a little hypocritical. Dove, after all, is still a beauty brand, and ultimately, its goal is to sell girls and women products that will make them prettier. And that just doesn’t jibe with a “feel good about the body you have” message—because, in essence, Dove is encouraging girls to feel good enough about themselves that they don’t need to buy all those products from Dove’s competitors, but not good enough that they can forego Dove products.

Don’t get me wrong—as an advertising cynic who takes an ardently feminist approach to advertising to and about women, I really like the general spirit of the campaign. But it’s entirely insincere for a beauty company to criticize the beauty industry while pretending it’s not a part of the problem. Would we give any credence to McDonald’s running ads on childhood obesity if they didn’t acknowledge the existence of Happy Meals?

And that’s the problem with the Campaign for Real Beauty—it’s co-opting a message of positive body image for the purpose of selling products that rely on girls and women feeling that their bodies are imperfect. To put it in starker terms, it’s commodifying girls’ self-esteem by turning it into just another marketing opportunity, and that’s particularly messed up. It’s one thing, after all, to be upfront and tell girls they’re ugly and that they should buy your product; it’s quite another to pretend to be empowering and supportive while passive-aggressively reminding girls that this great new Dove cream can really help them look great.

A couple of other criticisms of Dove’s campaign: There’s the implicit message that some women are more “real” than others, and that attractive or thin women don’t count as real. I’m pretty sure pitting groups of women against each other isn’t the most productive way to go about the whole empowerment thing.

Most damningly, there’s the stubborn little fact that Dove’s parent company, Unilever, also owns the Axe brand, which has no problem celebrating supposedly “fake” beauty in its ads. This was most ably demonstrated in Rye Clifton’s remarkable mash-up between a Dove Campaign for Real Beauty ad and various Axe ads.

Can't see the video? Click here to watch it on YouTube.

Lucky for Dove, those criticisms mostly didn’t stick, and the Campaign for Real Beauty is generally regarded as a major success for Dove. Its brand image as an empowering, pro-girl image is one Dove both values and actively cultivates—it has a “Dove Movement for Self-Esteem” website and brags about its Dove Self-Esteem Fund, which supports organizations like the Girl Scouts and Girls Inc.

In fact, as recently as April 4, Dove proudly tweeted that “[e]ach time you buy @Dove, you help us and our partners provide inspiring self-esteem programs to girls”—self-esteem that couldn’t possibly be belied by a tweet calling Dove Go Sleeveless Deodorant a “must have accessory” if girls are considering going sleeveless.

 Poll dancing

So this brings us back to Dove’s new deodorant. Here’s the marketing challenge facing Dove: How do you create a fake insecurity, push it onto women, and sell them a product that will fix that insecurity, all without looking like complete hypocrites after all that Campaign for Real Beauty stuff?

The answer lies in that press release, which repeatedly notes the results of Dove’s supposed survey: “Perceptions of unattractive underarms have women shunning sleeveless attire,” the headline screams. According to the study, “almost all women feel their underarms are unattractive”—the exact figure being “93 percent,” as mentioned later in the press release.

Remember, press releases are nothing more than marketing departments’ attempts to get the attention of news outlets to do a story on their products, framed the way the marketing department would like. It may be too much to hope that a press release is simply reprinted verbatim (although, sadly, that does happen a lot), but a slyly-written release can nudge a reporter into unwittingly writing a puff piece for a product.

So why does Dove want to emphasize the results of its market research so badly? (You don't see Mountain Dew putting out press releases stating that “90 percent of men think that Baja Blast is the most extreme flavor” when they roll out a new flavor.) Because that’s the way it can get out of looking like a bunch of self-esteem-crushing hypocrites. By pretending that Dove Go Sleeveless Deodorant is merely addressing a well-established problem rather than inventing one, Dove gets to shrug off criticism by saying, “Hey, we’re just responding to women’s needs.”

And it kind of worked. In a March 30 Wall Street Journal article, the “93 percent” statistic is repeated uncritically:

Dove Ultimate Go Sleeveless, which hits U.S. stores this week, claims its formula of specialized moisturizers will give women better-looking underarms in five days. It was inspired by Unilever PLC research that found 93% of women consider their armpits unattractive.

To its credit, the Wall Street Journal does cite the company as the source of the finding, and three paragraphs of the story are used to explain that being seen as fabricating a need is a risky strategy. But the reporting does, without question, take the company’s word that the research inspired the product, rather than raise the possibility that this purported “research” could just be another component of the marketing campaign.

On the other hand, WNBC, the NBC affiliate in New York City, just went ahead and handed Dove free ad time. During WNBC’s LX New York (which is admittedly more an entertainment and lifestyle show but is still hosted by what appear to be journalists), the host conducted an interview with Dove spokeswoman and Gossip Girl star Jessica Szohr about the deodorant. And by “interview,” I mean “hackish shill-session.”

Can't see the video? Click here to watch it on YouTube.

Consider this example of sparkling journalism from the clip above:

WNBC Host: A lot of women, apparently 93 percent of them who responded—

Szohr: Feel uncomfortable.

WNBC Host: Yeah!

Szohr: Yeah.

WNBC Host: Which I was kind of surprised by.

Szohr: Yeah. And this product basically, if you—when you use it, within five days, you get softer, smoother underarms. So that way, when you use halter tops, or tank tops, or dresses that show that area, you can feel comfortable and not self-conscious.

WNBC Host: So if people join in this campaign, they get to go shopping with you, maybe?

The press release offered by Dove doesn’t offer too much in terms of details about the survey itself; this is the entirety of the information about the survey:

Through a Unilever, Omnibus and TNS Research International the go sleeveless: Uncovering Underarms. An online, anonymous survey was taken by 534 women, ages 18-64 to gather insights on understanding the level of anxiety women feel about the appearance of their underarms.

And that’s not helpful. It’s journalistically unsound to cite poll results in news stories without seeing the complete questionnaire and results. Trusting a company to interpret results in an unbiased matter is foolish, and trusting a company’s poll as legitimate without at least giving it a once-over is just silly. This is doubly true when a poll itself is used as a cornerstone in an actual marketing campaign.

Plus, it’s unclear as to whether or not this poll is statistically valid. Dove wrote its press release suggesting some of its results can be generalized to the population (“93 percent of women think their underarms are unattractive,” Dove writes, and not 93 percent of women surveyed), but other statistics use the qualification (“Over half … of the women in the study”; “1 in 3 of the women studied”). The online nature of the survey raises more questions about the poll’s validity, too—the validity of online polls is often questioned due to issues with random sampling and self-selection bias.

So, I emailed Unilever and a media contact at Edelman (Unilever’s PR firm) and asked for the full questionnaire, with the exact wording and order of the questions, as well as the full results. I also asked very specific questions about the poll’s methodology, including if a random sample was used, how participants were recruited for the poll, and what the poll’s margin of error was.

After some friendly emails back and forth, Annette Evans of Edelman wound up emailing me this statement:

Thank you for your inquiry regarding the testing methodology of the go sleeveless: Uncovering Underarms study. This study was conducted by an independent third party market research supplier, to gather insights and provide statistically sound evidence.  This online, anonymous survey was taken by 534 women aged 18-64. The survey is statistically sound and adhered to industry-recognized guidelines and the highest quality standards.

Which is basically what the press release said. I emailed back, reiterating my desire for the full questionnaire and results and repeating my questions about methodology; alas, I didn’t hear back.

To be clear, I don’t blame Evans for sending an incredibly (and, more likely than not, deliberately) unhelpful statement. She’s just doing what her client likely instructed her to do. Frankly, I wouldn’t be surprised if I were the first person to ask for the full questionnaire and results—that would explain a lack of even a boilerplate, “sorry, we’re not releasing that” refusal.

But here’s the thing. If Dove is willing to release a part of its market research—the part that helps its marketing effort—without letting reporters see the whole picture, it makes it look like Dove is actively hiding something. And if Dove is unwilling to answer very basic questions about the poll’s methodology, it makes the whole marketing campaign—as well as Dove’s brand image—look completely bogus.

(For a detailed explanation on why I think it's so important to get the full questionnaire and full results from Dove, please click here.)

 Why this matters

With its campaign for Dove Go Sleeveless Deodorant, Dove appears to be trying to both manufacture and profit from the insecurities of its female customers. Dove is in the beauty industry, so that isn’t surprising.

It shouldn’t even be surprising that Dove is doing this after its whole Campaign for Real Beauty nonsense. A company that harshly criticizes the beauty industry while pretending that, somehow, it’s exempt from the criticism is hardly a paragon of consistency or self-awareness—and no number of donations to the Girl Scouts can change that.

What this should make abundantly clear is that “brand identity” or “brand personality” are meaningless concepts—or, at least, they should be. I don’t really get this tendency to anthropomorphize companies, but we should knock it off: by and large, companies exist to make money, and any personality they show is merely a means to that end. Nike isn’t an athletic company any more than Hallmark is a sentimental company. They’re companies. They make money. They assume whatever identity they feel best helps them make money.

And yes, this includes Dove. I don’t doubt that there are people who work for the Dove brand that care about the self-esteem and body images of young women; I would presume that many of its employees are women who have daughters. And I don’t doubt that many of its employees are earnest in their efforts to use the marketing resources of Dove to do a little good.

But we shouldn’t confuse those employees with the brand or the company. Dove makes its decisions based on what can help them sell enough body wash, deodorants, and creams to keep Unilever profitable. The Campaign for Real Beauty wouldn’t exist if Dove didn’t think it'd help Unilever make money, just as Dove Go Sleeveless Deodorant wouldn’t exist if Dove didn’t think it’d be profitable.

On one hand, I’m tempted to just shrug and chalk it up to the amorality of big business. But on the other hand, I’m pretty annoyed that Dove’s presumption that the cause of helping girls develop a healthy body image is a purchasable commodity, to be bought when it suits the brand’s marketing goals and dropped when it doesn’t.

The people at Dove who actually care about young women should be embarrassed by this campaign, and the people who actually believe that Dove cares about young women should be disabused of that notion.

And at the very least, nobody should be stressing out about their underarms that much. I mean, even before this product came out, Jessica Szohr hardly seems to be losing sleep over hers…

…so neither should you.

You can email me at, or tweet me @JoeDellosa.

Monday, May 9, 2011

Unilever’s Dove uses dubious “Go Sleeveless” poll to address criticism

This post originally appeared on my old “Joe Dellosa on Advertising” blog.

► Dove, criticized for inventing an insecurity for women, is justifying its new Go Sleeveless Deodorant by pointing to a study that says 93 percent of women think their underarms are unattractive but is refusing to release the actual poll or the full results.

This is an introduction of sorts to a fuller piece about Dove, its marketing campaign for the Dove Go Sleeveless Deodorant, and the inconsistency of its "pro-self esteem" brand image. The expanded piece will appear either later today or tomorrow.

Last month, Unilever's Dove began a marketing campaign to promote its new Dove Go Sleeveless Deodorant. The TV commercials for the deodorant promised women that they could have more beautiful underarms after using the product for just five days, which prompted many in the media to accuse Dove of attempting to invent a fabricated insecurity for women -- uh-oh, unattractive underarms! -- and profit from the resulting anxiety. Here's Stephen Colbert on the April 13 episode of The Colbert Report.

That a company in the beauty industry is trying to foment negative feelings in women about their bodies is certainly no surprise, but for Dove to do it seems hypocritical. Dove, after all, actively cultivates a reputation as a supportive, empowering brand for young women -- from its Campaign for Real Beauty to its Movement for Self-Esteem.

Possibly to fend off accusations of hypocrisy, Dove emphasized repeatedly in its press release about Dove Go Sleeveless Deodorant that the deodorant is the result of market research that found that "almost all women feel their underarms are unattractive" -- 93 percent, apparently. (This statistic was repeated unquestioningly in media outlets like the Wall Street Journal and WNBC.) One can imagine Dove arguing, poll in hand, that it's not inventing a new insecurity but merely addressing a pre-existing insecurity.

Unfortunately, the press release contained very little information about the poll itself -- no mention of the exact questions asked, how the poll was conducted, or any discussion of the methodology. It's even not clear if the poll is really statistically valid.

So, I asked Dove if I could take a look at the poll myself. I wanted to see the complete questionnaire, with the exact wording and exact order of the questions, and I wanted to see the complete results. I also had specific questions about the poll's methodology, including whether a random sample was used, how participants were recruited, and what the margin of error for the poll was. After a few emails back and forth, Dove, through its PR firm, declined to let me see the complete questionnaire and results and chose not to answer my questions about methodology. Apparently, I was to take Dove's word for it.

That, needless to say, is a bad idea. If a company is willing to mention some poll results -- presumably, the results that benefit the company -- in its press releases and other marketing materials, it should be willing to release the full results, with a full questionnaire, with a full explanation of the poll's methodology. To do any less isn't just shady and dishonest; it also insults the intelligence of that company's customers and undermine that company's credibility. More importantly, reporters -- especially those who cover advertising and media -- shouldn't let companies get away with it.

What follows below is kind of a Stats 101 mini-primer as to why seeing the questionnaire -- with the exact wording, exact sequence of questions, and full results -- is so important, and why, without it, even a statistically-valid poll can be completely dishonest and manipulative. Since Dove refuses to release any of that, I can only speculate as to how it got the results it did. For all I know, Dove asked fair questions that yielded legitimate results; however, until Dove decides to be more forthcoming, we'd do well to assume the worst.

 Exact wording

Consider the most prominent statistic in the marketing campaign, the “93 percent of women think their underarms are unattractive” stat. It could be Dove just very plainly asked:

Do you think your underarms are unattractive?

A: Yes
B: No

Would 496 women out of 534 women in a supposedly representative sample say they have an active belief that their underarms are unattractive? That seems unlikely. So what if Dove actually asked this:

Do you think your underarms are attractive?

A: Yes
B: No

Only 38 women answering "yes" still seems low, but it’s a little more plausible than 496 actively affirming a belief that their underarms are unattractive. And in this case, if 496 women answered “no,” they’re not necessarily saying they think their underarms are unattractive; they just never thought of describing their underarms as particularly attractive. (It's like if someone asked me if I thought my index finger is attractive -- I'd answer "no," not because I think it's ugly, but because, as a normal index finger, it's neither attractive or unattractive.) Of course, without the original question phrasing, we just don't know.

 Full results

Now, what if the Dove poll asked the question in this manner (with hypothetical results in brackets):

How do you feel about your underarms’ appearance?

A: My underarms are always attractive. [7%]
B: Every once in a while, my underarms don’t look the best. [50%]
C: Occasionally, my underarms don’t look the best. [30%]
D: Often, my underarms don’t look the best. [10%]
E: My underarms are always unattractive. [3%]

I could imagine that, presented with these choices, only 7 percent of respondents affirming that their underarms are always attractive, with a vast majority picking less absolute choices. But if Dove wanted the statistic that 93 percent of women think their underarms are unattractive, it can just lump all the non-“always attractive” responses together -- it's a tricky interpretation of data that could be seen as technically true but is patently dishonest. Hence, the importance of full results.

 Exact sequence of questions

Finally, consider if before asking about their underarms’ appearance, respondents were asked a series of questions like these:

Have you ever noticed any bumps or pimples on your underarms?

Have you ever noticed how rough your underarms can be?

Have you ever noticed any underarm decoloration?

Do you ever experience soreness or itchiness on your underarms?

Can you recall any times when you’ve been embarrassed by your underarms?

Followed up with an innocent, “Do you think your underarms are unattractive?” Clearly, the responses would be more than a little skewed.

It’s like asking, “Do you think Senator Jones is lying about having an affair?,” “Do you agree with Senator Jones’s decision to defund the troops, possibly leaving them without body armor?,” and “Do you think Senator Jones’s health care vote was the result of bribery, as some have suggested?”—and then asking, “Do you approve or disapprove of the job Senator Jones is doing?”

 In closing (for now)

I'll have much more on this in a fuller blog post about Dove, including an explanation of the inherent inconsistency of Dove's Campaign for Real Beauty and why I believe its campaign for its Go Sleeveless Deodorant is embarrassingly hypocritical. (I'm holding off posting that in part to give time for Dove to provide a statement and, hopefully, the full poll; a rep at Dove's PR firm said they'd have something by today.)

In the meantime, it's important to note that Dove is far from the only company to conduct dubious polls, cite the results in its marketing materials, and then balk that the notion of releasing the full questionnaire and results.

For instance, last year, FedEx attempted to smear UPS with its "Brown Bailout" campaign, in which FedEx falsely accused UPS of asking for a government bailout. FedEx cited polls on its website and in its press releases claiming public support for FedEx's position, and FedEx director of corporate communication Maury Lane mentioned polls when he was interviewed about the issue. (He did so during his interview with me, too.)

So I asked to see the full questionnaire and results. Lane told me that I would have to get it from FedEx's pollster, Public Opinion Strategies. Public Opinion Strategies said that they weren't authorized to release that information, and only FedEx could release it. When I emailed and left messages with Lane relaying what the pollster told me, Lane stopped returning my calls and emails.

This may be obvious, but it's a good idea to be suspicious of any poll a company conducts and then cites in marketing materials. Companies don't do polls because they're just so darn curious about people; they do polls because it helps them somehow -- maybe to have snappy soundbites in interviews, or to make press releases sound more authoritative, or, in what I believe is Dove's case, to attempt to preemptively fend off criticism.

Regardless, whenever companies offer poll results in a marketing campaign, customers and (especially) journalists should get into the habit of automatically asking for the full questionnaire and the full results. And if a company refuses to do so, we should get into the habit of telling them that they are being dishonest and very likely lying by omission.

There's only one reason a company would feel comfortable releasing only some but not all of a poll's results: they are hiding something. That's not breaking news, but it's good to remember.

You can email me at, or tweet me @JoeDellosa.