How to estimate the number of people in a crowd at a street action? Today in the world there are a large number of possibilities for counting the number of participants in various street events.
The exact or at least an approximate number of people at public events is a question that has long tormented experts and political scientists.
There are several ways to estimate the crowd: estimate the approximate number of participants; count literally "over the heads"; count the number of people in a certain area and then extrapolate the figure to the entire territory; finally, apply formulas. Let's take a look at all these methods. Immediately, we note that the more people in the crowd, the easier it is to make a mistake and the more difficult it is for us to compare large values. For example, in a small crowd, we can relatively accurately say by eye: there are 40-50 people here, and not 100-200. But if the crowd is large, then for an ordinary person who has not been specially trained in the methods of estimating the number of people, it is difficult to say how many people are in the square: a thousand or ten thousand.
To begin with, here is a seemingly simple question from RIA Novosti: how many people are in the selected fragment? Make a guess and we'll show you the correct answer later in the article.
We estimate "by eye"
This method is good when you do not need high accuracy of the result and when you are limited in time to apply more accurate methods. If you need to clarify - there are several dozen people in the crowd or several hundred, then it is quite possible to estimate "by eye". But at the same time, it is desirable that people stand freely, and not huddled close to each other, otherwise the quality of the assessment may seriously suffer.
Promotional video:
We count "over the heads"
Recently, my fellow journalists and I were at the cranberry festival in the Polesye village of Olmany, and we needed to estimate the approximate number of spectators. Since the territory was compact, and most of the audience watched the concert (plus a part of them “hung out” at several outlets), the easiest way was to count people literally “over their heads”: almost all of them stood still and looked at the stage. We quickly counted the "static" participants, then counted the "dynamic" ones, making allowance for their movements. It turned out that the celebration attracted about 300 spectators.
We count and multiply
The essence of this method is to count people in a small area, and then estimate how many times the total area of the territory is larger than the area of our calculation. Let's say there are flags at regular intervals on a city square. We calculated that 20 people fit between the flags, and there are ten such segments in total. Therefore, we are observing two hundred people. Knowing in advance where the mass action will take place, you can measure the size of the territory using a satellite image and outline for yourself the size of individual sections of this territory. For example, Oktyabrskaya Square in Minsk is 200 meters from Engels Street to the Palace of Culture of Trade Unions and 95 meters from the steps of the Palace of the Republic to the roadway of Independence Avenue, and between the lampposts - 45-50 meters.
Applying the Jacobs formula
Herbert Jacobs watched the Vietnam War protests in the 1960s from his office at the University of California at Berkeley. He became interested in the methodology for assessing the number of people and developed his own formula. The method is now called the “Jacobs method”. In the original, it boils down to the following figures: with a relatively free crowd, there is about 10 square feet of land per person, with a closer one - 4.5 square feet, and in a very dense crowd - only 2.5 square feet. If translated into values that are more understandable to us, then we get 1 person per square meter (people stand at arm's length), 2.4 people per "square" (dense crowd, but you can still walk between people) and 4.3 people per square meter (people stand shoulder to shoulder), respectively.
This method is most often combined with the extrapolation method described above. Let's say we observe a crowd in a certain area. By visual reference points (buildings, benches, trees, poles, trash cans), we can divide this area into equal areas, and we can also estimate the size of each area (or calculate in advance using maps and satellite images). We peer into the crowd: it turns out that it is not so dense - there is a lot of free space between people. We apply the Jacobs formula (1 person per "square"), estimate the number of people in a small area, and then multiply the result by the number of such sites.
We count the flow per unit of time
When people move in a column (for example, from one square to another, occupying the entire sidewalk), then it is convenient to estimate the total number of participants in the procession by how many people pass through the observer per unit of time. Suppose an observer stands at the edge of the sidewalk and counts that 20 people passed by him in a minute, and the whole procession past him lasted 20 minutes. It turns out that 400 people passed through the observer. It is better to take a larger time interval - say, 5 minutes, and for the accuracy of observations it is useful to place the second observer a little further away and take the average of their results, because during the procession someone joins the crowd, while someone, on the contrary, diverges.
At the same time, it is better to observe not from the side, but from some elevated point: if you just look “at the side” of the column, it seems that there are a lot of people, and there can be large gaps between the demonstrators, which are visible only from above. The same advice can be applied when assessing the number of people from a photograph: from the side photograph, one can say that people were "out of the way", and from the photograph from above, there were not so many demonstrators.
When estimating the number of people in a crowd, it is helpful to listen to what colleagues and other sources have to say about this in order to try to get an average. However, this does not always lead to increased accuracy of calculations: some sources are mistaken due to ignorance of the calculation method, some deliberately overestimate the number of people, and some deliberately underestimate.
In the West, where the history of street politics goes back more than one decade, they have also been trying to solve the problem of counting the participants in rallies for a long time. In recent years, modern technical means have come to the rescue. Here are some of them:
1) Photoelectric light barriers
Light barriers are usually used to protect areas, but they can be used with success when counting passing through "light frames". Usually such devices are small and can be easily built into the same frames of metal detectors, which are used to equip "gates" for access to meetings and demonstrations. Sensors pick up motion.
2) Acoustic touch floor panels
Small, floor-mounted touchpads are placed in the aisles of event attendees The visitor steps on the touch surface, thus counting. Unfortunately, this technology is not designed for use in low temperatures and on uneven surfaces.
3) Seismic wave counting techniques
The measurement is carried out with the help of special devices installed in the "ground", which transmit a signal to the receiving device at the moment when the event visitor steps on it. This technology is also not designed for use at low temperatures and on some types of surfaces.
4) Radar installations
Radar installations measure the number of visitors to events by measuring reflected waves. The types and forms of these systems can have various variations. It is difficult to count visitors passing in groups in such conditions.
5) Aerial photography
With the help of special devices, you can shoot large areas where public events are taking place on a camera, and count the participants using specially developed computer programs.
6) Counting passing through the turnstiles
This method is used by the Ministry of Internal Affairs - special reading devices are installed on the frames of metal detectors.
The sensors record only incoming ones, so if a person has left and entered several times, then they will be counted as not one, but two, three or more participants. Often the participants in the rally deliberately use this opportunity so that the organizers can account for a large number."
Therefore, along with the data from the counters, the police calculate how many people came to the action, based on the average density of participants in the area of the event. “For the calculation, the average density is taken, since it is clear that there are many more people at the stage at the rally and they stand closer to each other. At a considerable distance from it, everything is exactly the opposite. Moreover, each season has its own standards: for winter it is two square meters per person, for the summer period - one meter per person”, - said the representative of the press service.
In addition, an expert assessment of the filming of actions taken from above - from a helicopter or high-altitude points - is also used. “All these methods of counting, as a rule, are used in combination and complement each other,” the chapter emphasized.
7) Using counters
Two groups of "counters" (observers) are located at the following points: one hundred meters to the head of the column and at mark 2/3 of the path of the column. When the columns of protesters pass along an imaginary straight line of eyes, the "counters" use the pressing of a manual counter, recording every 10 people.
The head of the Analytical Division of the Paris Police Prefecture, Bayy, believes that this method is “rather artisanal, but effective. The margin of error is about 10-15% in favor of the protesters."
8) Measurements using photo and video cameras
In some cases, audience measurements can be carried out by filming an event using a photo and video camera. The data is then processed using various computer programs.
What do we know about the White Meter?
White counter is an informal, non-profit association of volunteers who count people at rallies and processions, with the aim of publicly distributing the intermediate and final results obtained, working on the principles of maximum openness of the methods used, the data obtained and the assessment of errors.
The White Meter project emerged from the Sonar election observer movement in early 2013.
At first, even at the rallies in 2012, not all frames of metal detectors were counted, but only selective ones, since there were not enough volunteers to count all the frames, and one person is able to reliably calculate two frames. Then the results were interpolated for the total number of people who came. As of May 2017, there were more than 70 people on the White Counter volunteer list. The core of the organization - the people who regularly take part in the calculations - are members of various communities of election observers. Since mid-2013, volunteers have been counting the flow of meeting participants at all levels, without using interpolation.
Counting method
Counting the number of participants in the rallies is carried out by monitoring all approaches to the venue of the event. Volunteers count everyone who comes by pressing a mechanical clicker counter. The counting is conducted from the moment the participants open the passage to the rally until the moment when free access is closed or the event ends. At regular intervals, the readings of all meters are summed up, and at the end the final results are summed up. To estimate the error, in parallel with manual counting, complete or selective video filming of the counting places is made.
All interim and final results of counting the number of rally participants are immediately published on public pages of social networks on the Internet. In addition, anyone, including representatives of the media, can get the current counting results right on the spot from the volunteer coordinator. After some time, a report on the calculation is published on the Internet.