How To Measure A Crowd, Without The Political Numbers

Transcript

ARI SHAPIRO, HOST:

This is WEEKEND EDITION from NPR News. I'm Ari Shapiro.

UNIDENTIFIED WOMAN #1: 52.

UNIDENTIFIED WOMAN #2: 11:52.

UNIDENTIFIED MAN: 25.

UNIDENTIFIED WOMAN #2: 6.12

SHAPIRO: Numbers are everywhere. They rule our lives. They shape our world. And we've been checking in with numbers expert Mona Chalabi from the data journalism website FiveThirtyEight.com, who gives us this number of the week.

UNIDENTIFIED WOMAN #2: 80,000.

SHAPIRO: That's the number of people who've participated in Hong Kong's protests according to organizers. Mona Chalabi joins us from our New York bureau. Welcome.

MONA CHALABI: Hi.

SHAPIRO: So organizers say 80,000 people marched one night, and yet, you read the news coverage of this and it says tens of thousands of people or masses of people. Why are news organizations avoiding that number 80,000?

CHALABI: I think they're just skeptical about how accurate it is. And as far as I'm concerned, they're right to be. The number of people in a crowd can be a really, really politicized figure. So people who want to undermine a protest movement might be inclined to underestimate the figure by just saying not that many people are taking part.

And equally, people who organize something or the supporters of an issue might want to overestimate crowd sizes because it can lend credibility to their cause. A big number in a crowd is basically adds more legitimacy to a movement, if you like.

SHAPIRO: So if you don't trust the organizers and we don't trust the opponents, is there a way to get a trustworthy number for crowd count?

CHALABI: Well, there are independent sources that sometimes attempt to count crowds by using methods like telephone surveys or aerial photography. But all of those methods are incredibly time-consuming.

So if you want real-time numbers, which is basically what the media want or security forces need if they're trying to take care of a protest, then you have to turn to, kind of, other techniques.

SHAPIRO: When you talk about aerial photography, are you talking about actually counting tens of thousands of dots, each dot being a head?

CHALABI: Not quite. So one way to do it is to take an image of the area where the protest is taking place and basically divide it up into a grid. And if you can work out the average number of people in any given square in the grid and just multiply it by the number of squares, then you're somewhere close to a kind of rough estimate.

SHAPIRO: And more recently, we also saw a lot of dispute over how many people participated in climate marches around the world.

CHALABI: Yeah, last month's People's Climate March in New York was a good example of that. So the organizer said that 310,000 people went. And then just a couple of hours later, they updated that figure to 400,000.

And the problem is that a figure like 310,000 kind of sounds a bit too suspiciously precise because unless you've got turnstiles and you can physically count the number of people attending, it's very hard to put that fine a number on it.

SHAPIRO: I used to report on American politics. And when we would attend a campaign for a candidate or for the president, there was a headcount down to the exact number. Is it possible to ever get that specific?

CHALABI: Absolutely. One of the reasons why rallies are easier to provide a precise figure on is because they often have a security presence there who can count the number of people that are coming in, whether it's even the number of people that are kind of having their bags checked or going through a metal detector. But the problem is that a lot of protests in the world simply aren't that organized.

SHAPIRO: So we began this conversation with the number 80,000 people marching in Hong Kong at one point a week ago. Is it at all reliable? Is there a more reliable figure we can use?

CHALABI: Well, you could look at other sources, so things like social media and the number of people who say that they're going to go and attend a protest like this. But that's also imprecise, too. I would say that it's difficult to say exactly how reliable that figure of 80,000 is.

But there's one indication that it's not too far off.

So Joshua Wong, who was the protest leader, asked supporters to download this app called FireChat that basically allows them to communicate even if they lose Internet connection. And since Sunday, the app's been downloaded about 100,000 times by people who presumably want to communicate whilst they're in that protest area.

It's not too far off 80,000 but again, it's just another number that is very difficult to verify.

SHAPIRO: Interesting. Mona Chalabi of the website FiveThirtyEight.com. Thanks a lot.

CHALABI: Thanks a lot. Transcript provided by NPR, Copyright NPR.