A bar chart presents grouped data with rectangular bars sized proportionally to the values they represent. The bars can be plotted vertically or horizontally.
A histogram is a graphical representation of the distribution of numerical data. It is basically a bar chart showing strictly numerical values.
What's the difference between a bar chart and a histogram? Bar chart data is grouped by category, histogram data is grouped by a quantitative variable. Because a histogram represents a continuous set of numerical values, the bars should touch.
An area chart is similar to a line chart, except the area below the line is filled to indicate volume. Different volumes can be overlapped. When multiple variables are compared, the fill color should be transparent so data is not hidden.
Stacked Area Chart
Alternatively, an area chart can compare values by stacking them. This also allows you to relate each value to the total.
A scatter plot displays a series of points. The position of each point is determined by its relationship to the horizontal and vertical axis.
A bubble chart uses the area of a circle to add a third axis of information to a representation of a series of values.
A pie chart is a circle divided into sections proportional to percentages. A pie chart works best when comparing only 2-3 values, as it becomes difficult to distinguish proportions when the number of values increases.If more than 3 values need to be displayed, a bar chart is probably a better option.
Skewing a pie chart with 3D effects should typically be avoided. This kind of effect skews our perception of the data, making it less accurate and less effective. Portions of a pie chart can be called out by applying non-skewing depth. When in doubt, it is best to keep pie charts flat and circular.
A radar chart is a graphical way of displaying data with three or more variables. Each axis starts from the same point. Like the polar area diagram, this chart works well for comparing cyclical data.
Keep your charts simple
Although it can be tempting to embellish charts to make the data look more important or a presentation more flashy, a good chart should be as simple as possible and easy to read. Here is an example of how not to do it, followed by suggestions about how to make it better.
Placement of dates and fruit type are switched.
Visual Structure is simplified and highlights data.
Bars are distinct colors with ample contrast.
The charts discussed here are just the tip of the iceberg. Each of these chart types was originally created for a specific research need. If you can’t find a chart that represents your research data well, don’t hesitate to think outside of the box and make something new.
Give us a call us at (541) 687 5778 or email us at [email protected]. We would love to discuss your next research project.