A histogram is a visual representation of data, rather than a numerical representation, making histogram frequency distributions more easily intuited. Histogram frequency is plotted along the y-axis, since a histogram is a series of columns representing binned data and the frequency distribution representing how often a circumstance occurs is shown in higher bars. To find the actual number of occurrences of an event, track the peak of the column representing events of interest to the y-axis and the number found at the height of that peak.
On a histogram, data is often binned into classes or columns, each of which is represented on a histogram by a vertical bar. The height of this bar indicates histogram frequency. The number of bars on a histogram can vary but is usually more than two because a histogram with only two columns can also be a bar graph. Often, data is normalized before it is graphed, so histograms showing "bell curves" or standard curves are often symmetrical and show a range of values, the highest of which are in the center and the smallest of which are on either end of the curve.
A population data curve can be interpolated from histogram data by drawing a curve across the top of the concatenated histogram class columns. This and other basic statistical analyses are easily ascertained by looking at a histogram. A viewer can get an idea for the total number within a given population or occurrence data set, and outliers are easily spotted. If there are skewed results and the data distribution is not a standard or normal curve, this is also shown on a histogram by seeing elevated columns on one end of the data set. No math is necessary for these simple, qualitative analyses.
Modes, medians and means are easily detected on histogram frequency graphs. The median, representing the point between the two farthest outliers, sits in the exact center of the histogram along the x-axis. The mode, representing the most common or most often occurring value, is the column that rises to the highest value along the y-axis. The mean is more difficult to spot without doing some calculations, but generally, given a normal bell-shaped histogram and not many outliers, the mean, or average, will be found around the halfway point of the x-axis. This average value will shift based on datapoints that sit far from the center of the histogram frequency distribution.
