It was big! I was expecting only 10,000 on the outside, but at least 40,000 people turned up. And those numbers are from police estimates, which have historically been (surprise, surprise) on the low side.

My favorite comment was from Ma Lik who claimed that the family activities day in Ma On Shan was not a boondoggle trying to divert people away from the march!

Before leaving the march at Cheung Kong at around 7pm to catch a ferry, I had independently said it was 200 to 250 thousand people.

If that was 40 or 63 thousand, I'm a monkey's uncle. Though I'm sure there will be at least one blogger who will contend that I and the organisers are in fact just that.

There will be a standard of truth that will emerge in a day or two: Robert Chung's HKU-POP count/survey.

First, his methodology was adopted by the Civil Human Rights Front at the 7/1 march. So there should not be any question whether this is the right or wrong way of doing it.

Second, his personal interests are simple. On one hand, there is this particular march whose outcome that everybody has declared beforehand is immaterial. On the other hand, he has to lay his entire career and legacy on the line. Will he fudge the numbers? Fat chance.

So just wait for it. Somebody is going to be found to be lying big time because 63,000 vs. 250,000+ is not even remotely close. I don't have a clue who's lying but I look forward to finding out.

Dr Chung may be independent, but not necessarily accurate.

Quite correct about Dr. Chung.
I'm expect that there will be others from other HKU departments, CUHK, HK Polytech, etc, like on July 1, 2004.

They can't all be dupes, can they? I mean, they must all have their careers on the line if they did. And this is a good and easy way of consolidating one's credentials.

more on accuracy.

if dr. chung is doing a telephone survey, there may be concerns about coverage, response rates, respondent selection, language issues and all that.

the basic questin is this: does the staff know how to count the number of people who pass through a certain point in a designated minute? could they be wrong by a factor of 4? No. HKU POP will be producing minute-by-minute counts at the observation points. I don't think anything can go wrong by that order of magnitude.

Yes it certainly does seem as though it was much higher than 40,000 or even 69,000, which was the revised police estimate. The problem with the police methodology is that what they do is pick one specific spot, and try to count how many people go through it over the course of the day. Unfortunately, that discounts many people that may sometimes find an alternative route, and if crowds movement over a set period of time changes through that particular point then their count is thrown off. It was a big turnout, no doubt.

The Donald has of course denied that this will change anything, saying that there was no room for maneuver.

What it must show, both to him and Beijing, however, is how fragile his mandate or claim to legitimacy is, when so few people had a say in his election. A wider franchise means greater legitimacy - full stop. It may also mean that Hong Kong will sometimes come up with decisions that China will not like - but perhaps that is a better outcome than China having to reinterpret the Basic Law every few years and completely undermining Hong Kong's rule of law in the process.

The Don has indicated he has 'got the message' but does that mean Beijing has? HK is no more closer to universal suffrage than it was on Saturday...because it is all up to Beijing. The have told The Don what the package will look like and what the end result will be...the only thing he can do is tweak the logistics of it...the finished product remains the same.

Same comment I posted to Glutter, but seems relevant here:

Clearly 63,000 is too low, since Victoria Park was packed to capacity, and well known to hold 80,000. My estimate, as posted here:
http://www.chattergarden.com/node/889

puts it at between 100,000 and 150,000. This was taken on Causeway Road near the library.

I think the HKU estimate last year July was fairly accurate and rigorous, while that of the organizer obviously flawed.

I tend to agree with the estimates from the academics.

however, if the organizer is founded to be significantly wrong. then i have questions, what is the big deal between 63 and 250? it doesn't really change a lot about the message. why is there such a need to exaggerate?

I think from July 1 March to December 4 March, 89.64% of marchers don't care who the organiser is. They are surely inexperienced and not always credible.

But the message from the marchers, I think, is quite clear.

Capacity of Hong Kong Stadium is roughly 40,000. That would put the 63,000 estimate of the police in the same ballpark as the Rugby 7s crowd.

You're gonna tell me that yesterday's crowd stretched from CGO to Victoria Park with full lanes covered is just your run of the mill Rugby 7s crowd? yeah, right.

this is a pretty useless exercise. but well, for the heck of it, let's talk about it.

andres lih seems to provide a pretty good reasoning. although the number may be slightly higher if one counts those who joined in the middle of the path.

tom, how did you arrive at 200-250k? care to explain your reasoning?

Who cares about the numbers although I like the point made above about the Rugby Sevens. It took a lot longer to move the crowd than it takes to clear Hong Kong Stadium.

As I said in my latest post the important thing is that this turn out was big and that with the difference in the economic situation compared to the original big march there is no confusing this public statement with general dissatisfaction with the government.

Tihs march was a simple call for universal sufferage. Donald Tsang says he will create a roadmap for democracy in his second term (if he gets one). That is rubbish. Any politician could come up with roadmap in two hours.

No offense to Andres, he might have picked the worst spot to count numbers. I remember that area as being a horrible bottleneck and his comments about periods of "breakouts" confirms this.

But look at the photos of the head of the march or any other portion of the march and you don't see the 10-15/second that Andres is reporting.

Count the number of people across the typical row on this photo which should be very close to the front of the march.

Check the TVB footage they took from the tram rolling through Wan Chai and you'll see that there aren't gaps in the march flow that would be necessary to get that density at the front of the march and only having 10-15/sec at Andres' checkpoint.

Could be that because of the bottleneck people skipped a block and went around Andres' checkpoint. Could be that people didn't make it down to Victoria Park and joined at Commercial Books or Sogo and were never counted at Andres' checkpoint.

But as a flow density problem, Andres' numbers would lead to much lighter density of marchers further up the road than what is displayed in any of the photos.

tom,

that is how you arrived at 200-250k? that is quite qualitative for a number range.

about andrew's number. let me clarify, it does not matter which spot you pick. the answer should be the same (assuming those who entered or left in the middle is small, i think about 10% according to some estimate in ESWN's links). everybody who completes the march pass through all spots.

as for aerial photoshot, ESWN has one source that counted the a snapshot and concluded the number was 92k. i think that is pretty accurate and consistent with Andrew's.

i believe the organizer has some calculation but i am curious how it differ from others. i suppose it is alright to exaggerate within the margin of error by taking the top of the range, but not further.

anyway, i agree with pretty much all fragrant-harbour said.

---
btw, i do not understand why the police will have any incentive to under-estimate the number. because the higher the turn out is, the more resources they can claim from the government. and a better job they can report in the performance evaluation.
it is possible that there is top-down order to tweak to the lower range, but then if they would announce their methodology and let more junior staff do the data collection it would be credible.

A simple and humble suggestion - can't they get the people that count crowds for Disneyland to help out?

No because then they would not tell us the count.

eswn has now a list from the academics.
http://www.zonaeuropa.com/200512brief.htm#020

but i thought the statistics and mathematics needed have all been taught in high school. but they really brought in acturialist and professors to do this.

the chinese saying call it
"killing a chick with an ox-knife"
杀鸡不用牛刀
presumably ox knives are too big for chicken.

yes, i think the disney watchers are good enough for this job. :)

---
talking about professors in HK, a couple years ago apple daily asked a HKU physics professor to answer the question given to applicants in an interview with Beijing and Tsinghua University. He totally flunked it but the appledaily reported believed him. it was a huge embarassment.
example: why is the hole of the sewage lid a circle? the professor went on to talk a lot of irrelevant BS i couldn't remember.

Sun Bin, I thought that was the nature of academia. ;-)

There's also a key philosophical question here:

If one hundred thousand (or so) people march for democracy, does anyone in Beijing see or hear it?

A "side issue":

I noted that some Chinese marchers (speaking other Chinese dialects) are not Hong Kongers. They got quite excited. Mainlanders could also watch the Cable TV news on KCR train.

EJ EXCLUSIVE: Universal Suffrage Not Later Than 2017: Beijing