Figure 3.1 shows a distribution based on a total of 57 children; the frequency
distribution consists of intervals with a width of 10 lb. Now imagine that we
increase the number of children to 50,000 and decrease the width of the inter-
vals to 0.01 lb. The histogram would now look more like the one in Figure 3.2,
where the step to go from one rectangular bar to the next is very small. Finally,
suppose that we increase the number of children to 10 million and decrease the
width of the interval to 0.00001 lb. You can now imagine a histogram with bars
having practically no widths and thus the steps have all but disappeared. If we
continue to increase the size of the data set and decrease the interval width, we
eventually arrive at a smooth curve superimposed on the histogram of Figure
3.2 called adensity curve. You may already have heard about thenormal dis-
tribution; it is described as being a bell-shaped distribution, sort of like a han-
dlebar moustache, similar to Figure 3.2. The name may suggest that most dis-
Figure 3.1 Distribution of weights of 57 children.Figure 3.2 Histogram based on a large data set of weights.NORMAL DISTRIBUTION 121