Million Man March

Michael Godin Arlotto (154aamga@pic.ucla.edu)
Wed, 22 Nov 95 13:25:30 -0800

I have been having trouble thinking of an alternative way to count the people
at the march, but I think Thanhnga's method may be more inefficient and less
accurate then just taking pictures. By standing in the middle of a group and
counting those in a specified area (10sqft I believe), you are assuming several
things about the person doing the counting. First you are assuming he or she
can see over everyone else. Second, you are assuming he or she will be able to
see those really short people. Third and most importantly, you are assuming he
or she can tell when a person is within the area and should be counted or
whether that person is in another area. I think these three problems will
result in misleading figures for the attendence at the march.

I think video tape could help a little. First, you should place the camera on
top of the stage facing the audience. The camera will be high enough as to see
the people in the back, while everyone will be facing the camera. Start the
camera on the far east side of the march and count those that can be seen. Now
slowly pan to the west and count the numbers there, and so on. You could make
a pattern somewhat like this:
>>>>>>>>>>>>>>>end
^<<<<<<<<<<<<<<
>>>>>>>>>>>>>>^
^<<<<<<<<<<<<<<start
stage and camera
This way you could keep an accurate count and now precisely where you have
already counted. It would be very helpful if the camera was on some sort mobil
divice so that your sample spaces were rectangular.

The article mentioned a 870,000 were in attendance with a resolution error of
25%. So the numbers range from 1.1 million to 650K, while the park service
came up with a total attendance of 440K. It seems the resolution error may
even be higher than originally thought.

Talk to ya soon,
Mike

Back to the Chatroom Homepage...

Back to the Stats 154A Homepage...

Back to the listings...

Send a message...

This archive was generated by hypermail 1.02.