Counting Crowds

Glenn Beck's Restore America rally at the Lincoln Memorial last Saturday has once more raised the issue of how to estimate the size of large crowds. Attendance at events, particularly political ones, is contentious.

I don't know about you, but I heard numbers for Restore America that ranged from a noncommittal "10s of thousands" to Michelle Bachmann's clearly hyperbolic "We're not going to let anyone get away with saying there were less than a million."

CBS News had the brilliant thought to pay some experienced crowd counters to estimate the crowd, based on aerial photos. One of the counters, Steve Doig, has explained how he came to his estimate (which was 80,000). A second counter, using the same photos independently, arrived at an estimate of 87,000.


Doig is a Pulitzer-Prize-winning journalist (at the Miami Herald) who is now an endowed chair in the journalism school at Arizona State University. He started estimating crowds while in Miami, including a visit by Pope John Paul II. He also counted the crowd at Obama's inauguration, and came to a number of 800,000...which conservative commentators applauded at the time, since mainstream media outlets were using numbers like 2 million. Doig writes, "I am amused to see that those who embraced my Obama inauguration estimate as soberly realistic are now attacking the Beck rally estimate, produced using exactly the same methods, as deliberately biased."

Doig does not reveal his density methods, but he does note in the comments that 10 square feet per person is a fairly tight crowd and "5 sf would be mosh-pit scary."

I did some square foot appraising at my office today, using a ruler and our convenient carpeting, which is marked off in squares 18" to a side. A density of 5 square feet (a square about 27" on a side) would be fine for a few minutes, but I agree that it's not likely most people at an outdoor event would be comfortable remaining at that distance for long. A moderate-sized adult sitting in a chair requires just about exactly 3 feet x 2 feet from head to toe, with not much leg room, and elbow to elbow (6 square feet). That doesn't allow any space between the chairs or rows of chairs. So clearly it's likely to be more like 8 - 10 feet per person if the crowd is seated, even on the ground.

Part of the problem with the Mall space is that some crowd estimates from the past are still considered accurate, even though there's no proof they were based on anything but a wild guess. 200,000 are said to have been at the I Have a Dream speech; supposedly, Lyndon Johnson's 1965 inauguration drew 1.2 - 1.5 million (what!?). I've seen estimates that the area between the Lincoln Memorial and Capitol can hold from 1.5 to 3 million people, depending on how densely they gathered, but no detailed mention of how this was arrived at.

So every time there's a large gathering on the Mall, organizers feel compelled to fluff up their event's metaphorical feathers to fill this mythically large space.

As a former professor of mine used to intone, "It's an empirical question." How hard would it be, really, to come up with some standards methods -- which are pretty likely to resemble Steve Doig's -- that we can all agree to?

Note: The photos above were supplemented with others to determine the size of the area and varying crowd densities. Unfortunately, the 16MB full-size versions of these photos have not yet been (and may never be) released for public viewing.