Ivan, you've asked several times for an explanation or interpretation of the quincunx. Your three-paragraph explanation seems to sum it all up fairly well. What "more" did you think you want? All it is, is a device to illustrate a certain kind of random behavior. There really isn't any more to be said about it. (Unless you want your students to construct one, but then the "more" is nothing by way of "explaining" and a lot about taking care in manufacture.)
Or, to put it another way, what do you not understand in the explanation you've quoted? On Tue, 2 Sep 2003, Ivan Balduci wrote: > Hi Dr J Williams, Dr David Heiser, Dr Rich Ulrich, Dr Dawson, Dr > Auslenders and all members > Thanks for your answers > ....................................................................... > Now, I would like to know what I can to know more to explain for my > students about the Galton Board (quincunx) more than this > explanation of this site > <http://www.stattucino.com/berrie/dsl/Galton.html>http://www.stattucino.com/berrie/dsl/Galton.html > > This applet simulates Galton's Board, in which balls are dropped > through a triangular array of nails. This device is also called a > quincunx. Every time a ball hits a nail it has a probability of 50 > percent to fall to the left of the nail and a probability of 50 > percent to fall to the right of the nail. Well, approximately; though I suppose if one is *simulating* a quincunx one can arrange P = 0.5 fairly precisely. With a real, physical board with nails in it, P = 0.5 only (a) to the precision with which the device was manufactured and (b) if the board is not leaning at all to left or right while one is doing the demonstration. > The piles of balls which accumulate in the slots beneath the > triangle will resemble a binomial distribution. To reach the bin at > the far left the ball must fall to the left every time it hits a > nail. > > Because Galton's board consists of a series of experiments the piles in the > slots are the sum of random variables. Therefore, this simulation provides > also an illustration of the central-limit theorem, which states that the > distribution of the sum of n random variables approaches the normal > distribution when n is large. When we add more rows of nails to the board > the approximation would be better. ----------------------------------------------------------------------- Donald F. Burrill [EMAIL PROTECTED] 56 Sebbins Pond Drive, Bedford, NH 03110 (603) 626-0816 . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
