This is a plot on a graph that looks like the cross section of a bell, a line arcing (relatively) smoothly upwards to a peak, then dropping in an inverse fashion back down. It is often described by a series of random events within a determined set of values where most of the events occur in a similar fashion, with a few results diverging from the mean. Not all random events fall into a bell curve. However, things happening within a defined set of circumstances tend to group themselves into a pattern. (see Gaussian Distribution)

For example, if you graphed the death rate in a given place vs. year of birth, the data would not follow a bell curve. But, if you graph deaths by age, or munber of births by family, Or birth rate by income, the data would follow such a curve. Another example would be to listen to a bag of microwave popcorn as it pops. First a few early kernels go, then the mass of popcorn pops, then you get some stragglers. If you charted corn popped over time, you'd also get a bell curve.

A famous experiment demonstrating this can be conducted with a board, one hundred nails, one hundred marbles, and some cardboard. Drive the nails partially into the board to create a 10 x 10 grid of posts, staggered by a half-row to create something looking like a pachinko machine, and tilt the board to make a ramp. Then, make ten little boxes with the cardboard at the bottom of the ramp to line up with the ten lanes created by the rows of nails, and let the 100 marbles fall down the middle of the ramp. Most of the balls will fall down the center, some will fall to one side or the other of the center, and a few will come to rest at the extreme edges. If you count the marbles, and perform the test several times, when you graph the results you will find an almost perfect bell-shaped curve.

Thanks to Professor Pi and Tdent for their input, forcing me to turn this from a quickie nodeshell rescue into a halfway decent w/u!

Log in or register to write something here or to contact authors.