A bar chart plots the number of times a particular value or category occurs in a data set, with the length of the bar representing the number of observations with that score or in that category.
Bar charts are often primarily used for displaying the quantities of qualitative or categorical data (e.g. the number of people in a sample falling into a given income, age range, ethnic group or religious affiliation), although they can also be used for quantitative data if the number of unique scores in the data set is not large.
While one axis represents the different (nominal) categories, the other axis can represent any measurement unit: relative frequency, raw count, percent, or whatever else is appropriate for the situation. Graphing by percent is most common.
Bar charts can be displayed horizontally or vertically (which are sometimes referred to as column charts) and they are drawn with a gap between the bars, whereas the bars of a histogram are drawn immediately next to each other.
Investment by area of impact
This example takes a close look at a vertical bar graph and breaks down several changes which transform it into an easier to interpret, more communicative horizontal bar graph (shown above).
This isn't to say that horizontal bar graphs will always be preferable to their vertical counterparts, but rather to highlight some things to think about as you are choosing between the two. When in doubt, plot your data both ways and compare side by side to judge which will be the easiest for your audience to consume.
Runs scored by national league division
These examples are simple bar and column charts prepared in Excel. Default formatting in Excel was adjusted to soften the gridlines and data labels to make the data and message stand out.
Advice for choosing this method
People are most accurate at judging length, thus making bar charts one of the best choices for communicating data. They are fairly well understood and easily interpreted.
Advice for using this method
Order the data in some meaningful way, usually greatest to least. Only place data on the chart that is for the comparison. Too many columns or bars can oversaturate the chart and become confusing. To solve this, prepare multiple comparative charts for different variables.
Nussbaumer, C. (2012). storytelling with data: my penchant for horizontal bar graphs. Retrieved September 2014, from https://www.storytellingwithdata.com/blog/2012/10/my-penchant-for-horizontal-bar-graphs
School of Psychology, University of New England (2000). Histograms and Bar charts. In: Chapter 4 Analysing the data (Research Options and Statistics course): http://www.une.edu.au/WebStat/unit_materials/c4_descriptive_statistics/histograms_barcharts.html (archived link)
Valery J Easton and John H. McColl (1997). Statistics Glossary: http://www.stats.gla.ac.uk/steps/glossary/presenting_data.html#bar (archived link)
'Bar chart' is referenced in:
- Rainbow Framework :