When we describe a histogram, there are many things that we can talk about.   The most important though are:

1. Shape:   Is the distribution symmetric, skewed (tail on one side and that is the direction that the distribution is skewed to), or neither.

2. Center:   Where is the middle?   You can talk about the mean or median since both are measures of center, but the median is much easier to find in a histogram by looking for the middle observation.

3.   Spread:   Does the data cover a large or small range of values?   Many times this is not as important unless you are comparing histograms (the spread of the male scores is larger than the female scores)

Other things to note when describing a histogram you could use are:                        

gaps, outliers (extreme values), peaks (mode - observations that happen the most often), and min/max.

Modifié le: mardi, 21 juin 2011, 12:00