A histogram is a plot that lets you discover, and show, the underlying frequency distribution (shape) of a set of continuous data.
A histogram is a chart showing the distribution of variables data. The variable is shown in order along the X axis, and the number of occurrences of each value (or group of values) is on the Y axis.
If your variables data is discrete with not too many values (for example a rating scale as in the picture above) you would have one column for each value. If the measure is continuous or has a lot of values, the scale would be divided into between 5 and 20 equal sized ‘bins’.
A histogram looks similar to a bar chart but it is not quite the same thing. A bar chart shows numbers with particular qualities or attributes, which have no particular order. The underlying data for a histogram has an order to it which is reflected in the chart.
Use a histogram if you want to understand the distribution of data associated with:
• scales, for example satisfaction ratings or other measurements including age
This type of data is known as continuous (variables) data. The histogram will give you an idea of the location, shape and spread of the data.
It is useful to look at the distribution of data if you have collected a lot of data (at least 30 data items) to see whether there are any patterns occurring. For example, you might spot errors or anomalies in the data. Sometimes it is also useful to understand whether the data follows a typical ‘bell shape’ or if it is skewed or asymmetrical.
When investigating your data it can also be really useful to split it into different groups and compare histograms. For example, you might be interested to see whether different teams involved in the above project have similar distributions of happiness scores. This might help you to decide whether to focus in on specific teams or take some other approach.
Histograms are a bit fiddly to create in excel. There are two options:
1. Work out the numbers in each bin yourself, enter the numbers and create a column chart.
2. Use the histogram function in the Analysis ToolPak add-in. Before trying to create the histogram, you need to define and enter the bins i.e. the ranges of data that will correspond to each bar on your chart. You may need to experiment with your choice of bins.
This wiki-how page has illustrated instructions on how to add the Analysis ToolPak and use the histogram function: https://www.wikihow.com/Create-a-Histogram-in-Excel.
IHI also have a guide to creating histograms: http://www.ihi.org/resources/Pages/Tools/Histogram.aspx.
Alternatively, please refer to the useful excel tool in section 6 created by Richard Scoville of IHI allows you to enter your data and the number of bins you want, and creates a histogram for you. (Do we need to ask permission to use this on the website?
Once you have created your histogram you will be able to see the shape of the distribution – whether it is symmetric, skewed one way or another, or has some other pattern, and you will have a better idea of the amount of variation in your measure.
The histogram gives you a view of variation that does not capture any patterns over time.