On Facebook and a myriad of other social media platforms, you can find out who your friends are dating, see pictures of their last vacation, and even know what they had for lunch yesterday. It is now becoming more unusual when someone chooses not to divulge their business than when they do.
Two research studies by Harvard Business School faculty explore this brave new world of "oversharing" — asking what it means to organizations and to reputation when we decide to buck the trend and keep personal information, well, personal.
The studies' surprising — and seemingly contradictory — conclusions about the costs of hiding information carry implications for individuals and companies alike. It turns out that who benefits from disclosing information has everything to do with how they reveal it.
Match Game
In What Hiding Reveals, Assistant Professor Leslie John, in the Negotiations, Organizations & Markets (NOM) unit, found that keeping unsavory information to ourselves may not always be in our best interest.
In fact, sometimes people think better of others who reveal ugly truths over those who keep mum.
To come to this conclusion, John and her co-researchers, HBS's Michael I. Norton and Kate Barasz, conducted an experiment asking participants to decide between two different dating partners based on their online profiles. Each profile contained answers to intimate and provocative questions, such as "Have you ever stolen anything worth more than $100?" and "Have you ever neglected to tell a partner about an STD you are currently suffering from?"
New research looks at the benefits and downsides of information sharing. ©iPhoto/rawpixel
Possible answers, given in multiple-choice format, included Never, Once, Sometimes, Frequently, and Choose Not to Answer.
When John and colleagues tested these various conditions, they found that participants were much more likely to choose a dating partner who answered all the questions, rather than someone who chose not to answer. Surprisingly, that was the case even when potential partners answered "frequently" to bad behavior.
"They would rather have someone who disclosed the worst possible thing they could than choose someone who doesn't disclose," says John.
On average, 80 percent of participants chose the "revealer" over the "hider." Even in cases where the respondent admitted to frequently hiding a sexually transmitted disease from a partner, 64 percent of participants chose that person over the person who decided not to answer the STD question.
One explanation for this result may be that subjects assumed that those who chose not to answer were engaging in bad behavior even more often than "frequently"— that is, they inferred an extra answer of "very frequently." When the researchers tested this possibility by asking participants to guess how often they thought the hiders did those things, however, they chose, on average, somewhere between "sometimes" and "frequently," meaning they assumed that they engaged in bad behavior less than the partner who did it "frequently"-yet they still chose the other partner.
"I thought this was a false positive at first," admits John. "But we replicated it many, many times. I was shocked."
The question is, why? In a series of follow-up studies, the researchers determined that the explanation may come down to one word: trust.
Honesty, The Best Policy?
In one experiment, for example, the researchers had participants play a game in which a person is given an amount of money, and then must decide how much of the money to give to a partner. Every dollar participants give is tripled. However, it is the partner who decides how much to give back to them-none, some, or all. Thus the amount of money participants give is heavily determined by how much they trust their partners.
When shown profile questionnaires filled out by their partners (who had been induced to either answer the questions or leave them blank), participants routinely gave less money to those who had chosen not to answer the questions, even compared to those who said they "frequently" tried to gain access to another person's email account, for instance, or faked a sick day at work.
"We like people who are honest," concludes John. "It signals trustworthiness, and that seems to have a positive "halo" effect, such that we are willing to overlook an honest person's bad behavior."
There may be completely innocuous reasons someone may wish to keep personal information private
The implication may be that people overcompensate in hiding bad information about themselves.
In another experiment participants were asked whether they would admit that they used drugs on a job application; a different set of participants acting as prospective employers were asked whom they'd rather hire, someone who admitted using drugs, or someone who chose not to answer.
Even though only 23 percent of respondents said they'd admit using drugs, prospective employers found drug users hireable 62 percent of the time, versus only 45 percent of the time for those who chose not to answer that question.
Of course, such honesty has its limits, John hastens to add. "You might not want to say you are a heroin addict," she says. "But if you are trying to decide whether to hide or reveal information, people often have a knee-jerk reaction that they shouldn't say something bad about themselves, when they might be better off being honest."
On the other side, there may be perfectly benign reasons why people might withhold information-from a job application, a dating profile, or a Facebook page-starting with the fact that they don't think it's anyone else's business. In this case, it helps for observers to be aware that hiding information isn't necessarily an admission of guilt.
"As observers, we may be prone to missing opportunities to form friendships or hire people by unfairly inferring [that] they are untrustworthy," she says. "There may be completely innocuous reasons someone may wish to keep personal information private."
Full Disclosure
While John's study shows that people think badly of people who withhold information, another recent HBS study found differently.
In Is No News (Perceived As) Bad News? An Experimental Investigation of Information Exposure, Assistant Professor Michael Luca, also from the NOM unit, found that people are likely to give others the benefit of the doubt when they fail to fully disclose bad news about themselves. While on the face of it, Luca's findings would seem to contradict John's paper, in reality, the two studies complement each other, showing just how subtle can be the way we process information.
Luca, who works a few doors down the hall from John, has studied the ways in which organizations hide information from consumers-sometimes duplicitously. In a previous paper about U.S. News & World Report college rankings of MBA programs, for example, he found a strong link between where a school fell on the rankings and how likely it was to list that ranking on its website.
"Outside of the top 25 programs, business schools with worse rankings become less and less likely to mention them on their websites, and more and more likely to include other information instead," says Luca.
The problem is that in some cases keeping information private can directly harm consumers. After Los Angeles required mandatory hygiene information at restaurants, for example, hygiene rates rose and foodborne illnesses dropped.
"Just by disclosing the information, and letting markets take action, it led to a positive social effect," Luca says. In this case, however, it took the direct intervention of government to persuade restaurants to reveal this information which hadn't been done voluntarily.
According to game theory, however, that shouldn't be necessary. The logic goes like this: The best restaurants or schools should loudly trumpet their A rankings as a matter of course. Then B-ranked restaurants or schools would reveal their rankings, to separate themselves from the Cs. The pattern would continue to the C establishments and so on.
"The theory is that the information would unravel, and everyone but the very lowest grade would have the incentive to disclose," says Luca.
Despite that theory of "information unraveling," however, in reality that is generally not what happens. In the case of restaurants, very few voluntarily disclosed their hygiene ratings, even when they were above average. In order to test why, Luca, along with Ginger Jin of the University of Maryland and Daniel Martin of the Paris School of Economics, set up a simple experiment they called the "disclosure game."
Guessing Game
For the experiment, the researchers separated participants into pairs. One person, the "sender," was randomly assigned a number from 1 to 5, and could choose to reveal or hide that number to the other person, the "receiver." If they chose to hide it, then the receiver tried to guess the number as close as possible, using increments of .5.
Here's the catch: Receivers were paid more money depending on how close they were to the number, while senders were paid more depending on how high the receiver guessed. That meant the senders had more incentive to hide the number when the number was smaller. That's exactly what happened. For number 1, senders only revealed the number 5.7 percent of the time; for 4 or 5, they revealed the number 97.7 percent of the time. For numbers 2 and 3, meanwhile, senders varied, reporting the number 40.8 percent and 88.6 percent of the time, respectively.
Where the experiment gets more interesting, however, is looking at the numbers guessed by the "receivers." While the average non-reported number was 1.584, the average guess was 2.022-meaning that the receivers routinely guessed too high, underestimating the extent to which the "senders" were hiding bad information.
While this experiment was performed in the lab, Luca extrapolates the findings to apply to consumers (receivers), who want to know the true quality of a product, while sellers (senders) hide it from them.
"Customers were not inferring the worst, and sellers take advantage of this," concludes Luca. "They were guessing a higher-quality rating than the actual quality rating." That would explain why information doesn't "unravel" according to game theory predictions, and why companies don't voluntarily release information even when it is not the worst it could be.
Customers give too much credit to companies for not disclosing information
"Customers give too much credit to companies for not disclosing information. That was the big takeaway for us," says Luca. "Policymakers need to be more heavy-handed in making sure organizations are disclosing information. And customers should be leery of the sound of silence."
As another example of this phenomenon, Luca points to movie studios that, when they know they have a flop on their hands, withhold movie previews from critics to avoid bad reviews during the critical opening days. Studios trumpet good reviews in their marketing, but of course exclude bad reviews. Customers should understand that no news (or reviews) is bad news in this situation — but oftentimes they ignore the lack of reviews, and flock to opening weekend anyway.
"Consumers should think about what it means when a company is not giving you information — and think about what information they could have given you."
Dogs That Don't Bark
On the face of it, John's and Luca's studies seem to be showing different things. In John's study, people think worse of those who hide information, while in Luca's, they seem to give more benefit of the doubt to them than they should. The discrepancy may come from just how apparent it is that information is being hidden.
"People don't notice the dog that doesn't bark," says John. By including the "choose not to answer" choice in their study, she and her colleagues intentionally made it clear that the person who answered the profile was hiding information-leading observers to conclude that the individual was less trustworthy.
In the case of restaurants, movies, or college rankings, on the other hand, consumers may not realize that information is being withheld from them. "If restaurants were required to say that they were choosing not to reveal their hygiene [rating], I think it would be a day before everyone would stop going," says Luca. Of course, requiring restaurants to reveal that they are choosing not to reveal is probably just as difficult as requiring them to reveal in the first place.
The bigger takeaway from both studies may be that it pays for consumers to be aware of the information they should be looking for — whether that means going directly to U.S. News & World Report for the complete list of college rankings or looking up film reviews on Rotten Tomatoes — rather than just considering information a company provides. As for that organization and person choosing not to answer a question, they may very well be hiding something. Or they may just be choosing not to answer.
