# How to estimate the number of people who attended the Bersih 2.0 street rally. Answer: between 25,000 to 30,000

How many people turned up for the Bersih 2.0’s Kuala Lumpur street rally on July 9, 2011? Police estimated between 5,000 to 6,000 people, whereas the media reported closer to 20,000 people. The event organizer of Bersih 2.0, however, estimated that around 50,000 people showed up for the street demonstrations. Some bloggers who were also involved in the street rally estimated an even higher number of 200,000 people. So which figure is correct?

But first: why is it important to know the number of people who showed up for the street rally? It is important because the number of attendees roughly measures the people’s support of Bersih 2.0 and, in turn, how many people oppose the government that strongly enough to vent their frustrations on the streets.

So, a figure of a couple of hundreds would be deemed a failure by the Bersih 2.0 organizers, whereas a turnout of one million people (as the opposition parties were hoping for) would be considered an effective blow by the opposition on the present government.

In getting the number of attendees, I wasn’t interested in rough guesses or skewed or emotional estimations. Fortunately, a more rigorous estimation approach was done by PoliTweet.org which is a non-partisan research firm that analyzes the online social interactions, mainly via Tweeter, between Malaysians and their politicians. Analyzing Tweeter traffic and what people were saying at that time, PoliTweet determined that there were five hot spots (or hubs) of large crowds during the rally in Kuala Lumpur.

PoliTweet then determined the total area of the streets in each of those hot spots and calculated the number of people who could “fit” into or occupy those areas. For example, the Puduraya hotspot has a total street area of 127,536 square feet, and taking one person per four square feet, the estimated number of people at Puduraya area was (127,536 / 4) or 31,884 people.

So, adding all the numbers for the five hotspots revealed that the total number of people at the Bersih 2.0 rally was nearly 50,000.

However, PoliTweet’s method assumes that, for each hotspot, people occupied the full or the whole area of the streets. For instance, PoliTweet assumed that the streets in the Puduraya hotspot was completely filled with people (taking into account of “free space” for each person – recall, one person for every four square feet).

So, this figure of 50,000 should be taken as the * maximum number* of people who

*attended the street rally, rather than the actual number of attendees. In other words, PoliTweet’s work puts a cap (or maximum) on the number of people who could potentially attend the street rally.*

**could have**So, if 50,000 is the maximum number of people who could have attended, what is the actual attendance then? For that, I am going to make some very rough calculations based on a single photo! I like to stress this is merely an educational exercise rather than a journal publication-worthy work.

Look at the photo below which shows a crowd of people facing the security forces. This is somewhat a convenient photo to work with because it shows the full crowd which we can estimate their numbers. It is as if all the Bersih 2.0 people gathered at the same place at this particular hotspot to have their group photo taken.

To obtain the number of people in this crowd, I have to make some necessary assumptions. The assumptions are as follows.

**One,** the width of the entire road is 14.4 m. I got this figure by looking at the number of people in the front row of the crowd. I estimate that 24 persons could comfortably stand (or sit) across the entire road width. Assuming further that each person occupies a space of 0.6 m, this means that the entire road width is (24 persons x 0.6 m) or 14.4 m. This road has four lanes, so each lane is (14.4 m / 4 lanes) or about 3.6 m wide. This single lane width of 3.6 m is about right for major, heavy traffic urban roads in Malaysia.

**Second,** in a packed crowd, I assume every person would occupy an area of two square feet (or 0.19 square meter). One square feet would be too small. People would literally be standing shoulder-to-shoulder and breathing down the front person’s neck. Four square feet is the area of people walking in a packed crowd. So, two square feet per person is about right for a packed but static crowd.

**Third,** the person who took the above photo is assumed to be standing 20 m away from the first row of the crowd. This distance looks about right from the photo. This assumption is necessary because no camera focal length was given in the photo. Searching for the same photo from other websites proved fruitless. Information about the camera focal length is absent.

Why do we need to know the camera focal length? We need this information to determine the distance between the first and last row in the crowd. I would use simple geometry to determine this distance.

We are trying to determine X1, the distance between an object (the front row of the crowd, in this case) and the camera. To determine X1, we use: X1 = (X2/Y2) x Y1.

The average height of a Malaysian man is 1.65 m. This height is Y1. And from the photo, I measured that the height of several men in the front row of the crowd measured at an average of 1.6 cm or 0.16 m. This height is Y2. Previously, I mentioned that I had to assume that the photographer was standing 20 m away from the crowd. This assumed distance is X1. Hence, the focal length, X2, is (X1/Y1 x Y2) or (20/1.65 x 0.16) to give 194 cm.

Now, let’s look at the last row in the crowd. The last row looks quite blurry from the photo, but I managed to measure that the average height of the men in the last row as 0.4 cm or 0.004 m. This is Y2. Again, I am going to take the actual average height of the men in the last row as 1.65 m. This is Y1. Taking the previously calculated X2 value as 0.194 m, X1 is (X2/Y2 x Y1) or (0.194/0.004 x 1.65) to give 80 m.

Consequently, the distance between the first and last row of the crowd is (80 – 20) or 60 m. Taking the road as a rectangular, the total area occupied by the crowd on the road is (60 m long x 14.4 m wide) or 864 square meter. Since this is a four-lane street, each lane has an area of (864/4) or 216 square meter.

The number of people occupying a single road lane is thus (216 square meter / 0.19 square meter per person) to give 1,137 persons.

Looking at the photo, it appeared the crowd was occupying about five lanes (note the “spillover” of people from the four lanes into the next lane on the right of the photo).

Thus, the estimated total number of people in this crowd is (1,137 persons per lane x 5 lanes) or 5,685 persons. So, perhaps the police are correct that between 5,000 to 6,000 people attended the street rally. However, this figure is only valid for one hotspot.

Recall that PoliTweet determined that there were five hotspots at the capital that day. So, assuming this crowd number is representative of the crowds in the other hot spots, I estimate that the **total** number of people who attended the Bersih 2.0 rally was (5,685 persons per hotspot x 5 hotspots) to give 28,425 persons.

This figure is nearly 30,000, a figure closer to that estimated by the media.

Interestingly, Syed Akbar Ali, a well-known book author and newspaper columnist, received information that no less than 3,000 police undercover officers were placed among the crowd. If this is true, the actual figure of the total number *genuine protesters* who turned up for the rally is reduced to 25,425, which is about one-half of the figure estimated by Bersih 2.0 organizers and about five times more than that given by the police.

So, my final answer is this: Between 25,000 to 30,000 people attended the Bersih 2.0 street rally on July 9, 2011.

And is this figure a significant number of people? I believe Bersih 2.0 is a stalemate between the opposition and the government. This figure of 25,000 to 30,000 is too few a number for a critical mass for a government overhaul and too large a number for the government to say to the opposition parties, “Crowd? What crowd?”

### Further readings:

- Malaysia Military Power – A blog of another way to estimate the crowd numbers.
- From a Malaysian in UK on Bersih – An interesting and more rationale view on Bersih 2.0.