BOB GARFIELD:
This is On the Media. I'm Bob Garfield.
BROOKE GLADSTONE:
And I'm Brooke Gladstone. Throughout the short life of the new media, we at On the Media have always held one truth to be self evident, that, of course, being that porn drives the Internet. That fundamental belief was shaken a few weeks ago when we spotted a blasphemous headline exclaiming, “Social networking surpasses porn.”

That assertion comes from Bill Tancer, author of Click: What Millions of People are Doing Online and Why it Matters. Tancer, who studies all of our online search habits, has the data to back that assertion up.
BILL TANCER:
Our theory is, is that it’s a zero sum game. We only have so many hours in the day. Social networking now taking up so much time for the younger demographic they're spending less time on porn sites.
BROOKE GLADSTONE:
And you had one singular anecdote to sort of underpin all of this.
BILL TANCER:
Yes. That was one of the most fascinating times is when we looked at the two days in summer — it was, I think, two years ago — that MySpace was unavailable. Where did people go when they didn't have access to what at the time was the number one site visited across all Internet sites that we track?

The thing that we found that happened when MySpace went down was that visits to dating sites and adult entertainment sites were the two categories of sites that increased.
BROOKE GLADSTONE:
By how much?
BILL TANCER:
About a 10 percent increase to both which is pretty substantial in terms of the data that we track.
BROOKE GLADSTONE:
So let's seamlessly segue way into Viagra. You have a lot of data on this from two streams. One is the billions of emails that we get — spam about these products — and the other involves searches that seem to spike during the Super Bowl.
BILL TANCER:
If you look at the trending of spam on Viagra, there also is a corresponding increase in searches on Viagra, which might indicate that although it’s extremely annoying, it may actually be raising awareness of erectile dysfunction drugs.

You also mentioned that Super Bowl is when we see the spike in Viagra searches. We think that’s caused by Super Bowl commercials. And, of course, we found with all of the erectile dysfunction drugs that have been advertised during Super Bowl, they all experience increases during that week.
BROOKE GLADSTONE:
So you don't relate it to the, you know, testosteronic impact of watching the game.
BILL TANCER:
Yeah, that’s, of course, a theory that you could pursue, but, looking at the data, I think it’s more directly driven from the actual advertisements.
BROOKE GLADSTONE:
I was really interested by what you found with online gambling. You found an inverse relationship between online sports betting and online poker.
BILL TANCER:
That's right. I noticed this negative correlation between visits to online poker sites versus visits to sportsbook sites — when I say negative correlation, as one goes up, the other goes down — almost a perfect negative correlation.

What it told us about the online gambler in aggregate is that it’s probably also a zero sum game. As sports season starts to die down, they'll take the same amount of money they would have bet in a sportsbook and then visit poker sites, for example, to place bets, so a certain amount of money to spend on betting at any given time and a tradeoff between the two activities.
BROOKE GLADSTONE:
Bill, have you wondered why there’s so much vice [LAUGHS] that shows up in your data? And, in fact, PPC generally stands for pay per click, but you've adopted the acronym to stand for porn, pills and casinos. If porn isn't driving the Net, is vice in general?
BILL TANCER:
I don't know that it actually is driving the Net. You know, as we mentioned from the other statistics, there’s other categories of sites that are increasing. But the thing I found most interesting about porn, pills and casinos is that it highlights a deficiency in the traditional ways we conduct market research.

If you did a telephone poll, for example, and asked people how much time you spend on adult entertainment sites, probably not going to get an accurate answer something that in the research field we call cognitive dissonance. People want to appear in the best light possible, and sometimes they will change their answers from reality to an answer that shows them in a more positive light.

Now, if we compare that type of research, traditional surveys, to observed behavior, which is what we're looking at, what people are doing in aggregate on the Internet, we see a big difference. What we think about the economy and what we think about politicians, all of that data, what we might not admit to on a survey may show up in our aggregate search behavior.
BROOKE GLADSTONE:
Now, I get that. On the other hand, not everybody is online yet. In fact, many, many more people have telephones. So why do you think that you get a truer picture of the population in general from just evaluating what the online population does?
BILL TANCER:
And Brooke, that’s a very good point that you bring up. There is still a segment of the population who’s not online, and this data doesn't apply to them. The flip side of that is looking at how traditional polls are conducted today. I mention this in the book.

One of the challenges with conducting traditional polls is getting to those households that no longer have a landline phone. Most polls conducted by random digit dial, or a machine that’s randomly dialing up phone numbers, you can't use those machines when dialing cell phones.
BROOKE GLADSTONE:
And so we're going to assume that all those guys have computers.

BILL TANCER:
I don't think you can assume that, either. But know that there’s problems with getting a representative sample on both sides.
BROOKE GLADSTONE:
Since people use the Internet alone, usually, and since people vote alone in the voting booth, do you think that you can predict better how people may behave on that fateful day?
BILL TANCER:
That’s one area I can't predict. And those that know me know that I've predicted American Idol results, I've predicted Dancing with the Stars. [BROOKE LAUGHS] I'm often asked, can I predict who’s going to win in November? And the answer is, no, for a very interesting reason.

The last presidential election we saw this phenomenon in our politics category. We were tracking political blogs, and we actually coded them for those that were more liberal, those that were more conservative and those that were more centric. And the closer we got to election, the more crossover traffic was happening between those blogs.

And what we think was happening is that the closer you got to election, people were visiting the opposing viewpoint, not because they supported it but because they wanted to arm their arguments with what was ever featured on the opponent’s site.

And so this crossover going back and forth, it confounds the variable makes it almost impossible to separate out — is someone popular because they're popular or because somebody wants to find dirt on them or find something to fuel their own arguments?
BROOKE GLADSTONE:
Now, are you just being cynical, or do you know that when somebody goes from the American Spectator site to a relatively neutral site, like RealClearPolitics, to The Huffington Post, that they're really just looking to arm themselves against the arguments of their opponents? Isn't there just the remote possibility that they're looking for information that may run counter to their convictions?
BILL TANCER:
It may be a little bit. I think you’re right. There are also is some swing vote in there that might be people that are truly trying to make a decision and are using the Internet as a tool to make that decision.

My point is, is that the variables become confounded and you really can't separate the two, which makes it almost impossible to say who’s going to win.
BROOKE GLADSTONE:
All right, thanks very much for being on the show.
BILL TANCER:
Thanks, Brooke.
BROOKE GLADSTONE:
Bill Tancer’s new book is Click: What Millions of People are Doing Online and Why it Matters.