Understanding and using Histograms
How to read Histograms
Histograms split a single continuous measure into bins, or groups, that represent a specific range of values. Data points are then grouped into these equally sized bins. The bins are then displayed visually as bars stacked next to each other. To measure the scores of Olympic divers, one bin could contain scores between 2 and 4, the next between 4 and 6 and another the scores between 6 and 8.
Bins are measured by the number of occurrences within each range of values. This count will alter the appearance of the view depending on where the values from the data are concentrated. When the values are concentrated on one side or the other of the middle it is called skew. Other ways the data could display are called:
- Bimodal distribution - which has two peaks
- Plateau Distribution - which rises to a certain level and stays there for most of the bins
- Edge Peak Distribution - which looks like a normal frequency distribution, except one bin at the end is higher than the rest, serving as a sort of tail
The x-axis used on a histogram functions as the width of the bars, while the y-axis functions as the height. Unless a particular category has no frequency at all, there should be no spaces between bars.
There are two primary ways of displaying data in a histogram: Count of values within bins, and Density of values (% of Total).