How to choose a chart? Which chart would best fit the data set that I have? How do I choose the better chart type from Excel? Finding the right type of chart for the data set that you have can be a bit of a challenge. Given a set of seemingly equivalent alternatives, picking one chart over the other can turn out to be quite an interesting exercise.
The chart chooser diagram can help you make a choice in situations like these. The chart highlights some of the most frequently used charts for plotting data and helps the reader pick the right chart type. (Please bear in mind that this list is neither an exhaustive nor a conclusive list of all chart types and may be subject to further modifications.)
You can download the larger version of the chart chooser (PNG) here.
For those would might want to look at some of the chart types used in the above chart chooser a little bit closer, here’s a quick summary:
A Bar Chart or bar graph is a chart with rectangular bars with the height (or length) of each bar being proportional to the actual values they represent. Bar charts along with the line chart are the two most commonly used charts. Bar charts are very versatile and can be used to depict the frequency distribution (histogram), the change in the variable over time or for comparing multiple variables at any given point in time. Other variations and alternatives include the stacked bar chart, bar chart scaled to 100% (same size) and the 3-D bar chart. Bar charts can also be horizontally inclined though they typically are shown with a vertical inclination.
Line Chart is a chart in which the various data points are represented by points which are connected with lines. A line chart can be used to show the variation in data over a period of time and its frequency distribution for one or more variables. Line charts are often used in combination with bar and area charts.
A Pie Chart (or a circle graph) is a circular chart which is divided into sectors. Each sector of the pie chart shows the portion of the total corresponding that particular component. The area under each component, the length of the arc and the angle all are in the same ratio as the actual value of components. Variations of the basic circular pie chart include donut charts, blow-up pie charts and 3-D pie charts although the simple circular version is almost always preferable.
An Area Chart or area graph is used to represent data on a chart with the portion under each point (and the line connecting each point) being shaded out. The area chart is similar to a line chart and given an option between the two, in most cases the choice would tend to tilt towards the line chart.
A Heat Map is a graphical representation of data where the change in values of one or more variables are represented as change in colors. A heat map is often times linear although other options such as the circular and 3 dimensional heat maps are also used.
Treemaps are similar to heat maps. In a treemap, data is shown as nested rectangles. A large rectangle will represent a major logical division of data while smaller nested rectangles show the various sub-divisions within the larger portion. A treemap can be thought of as a tree structure in reverse where the smaller units are packed within larger units rather than projecting outwards. Oftentimes the various divisions are separately colored to add yet another dimension (in which case it becomes a combination of a heatmap and a treemap)
Box Plots (also known as box-and-whisker diagram) are typically used in statistics and depict a finite set of number (oftentimes limited to four, five or six data points per box plot). They consist of a line and a rectangle with the line being vertically inclined and piercing the rectangle in the middle. The top part of the line visible above the rectangle, the top and bottom edges of the rectangle and the portion of line visible below the rectangle are all used to indicate one separate data point. (In stock market, a box plot will commonly indicate the highest price of the stock, the opening, the closing and the lowest price of a stock over a period of time. The color of the rectangle can be used to indicate a fifth data point. The rectangle itself can be divided into two with the length of each part being used to indicate yet another (6th) data point.
A Radar Chart is a two dimensional chart use to plot data around a circular scale. Although they can be used to chart two or more variables over a given cyclical time frame, their use is limited and a line chart oftentimes can show the same variation with lesser complexity.
Funnel Charts are often used to represent a diminishing quantity of a single variable over a number of stages. In sales and marketing, they depict a reduction in the number of “leads” as they filer down from the prospect stage to the “buying” stage. This type of a chart can also used to visually identify the stages which account for the biggest reductions in the available resources. However, just like the radar chart, it’s usefulness is limited and can easily be replaced by a bar chart which can help reduce the complexity.
A Bubble Chart depicts a series of numbers as bubbles with the size of the bubble indicating the value of a variable along a particular dimension (and the X-Y position representative of two more dimensions). When plotted in three dimensions, the bubble chart can indicate a potentially four dimensions of the same data point.
A Surface Charts are used to indicate a continuous interplay of two or more variables over a three dimensional space. It is oftentimes used to show complex formations resulting from the interaction of variables in a graphical manner.
Bump charts got their name from “bumps race’”, a term used to refer to a boat race where each boat tries to “bump” the one in front and move up the chart. Bump charts have become quite common of late and are typically used to represent changes in the position of a given number of competing entities over a fixed time duration.
A Flowchart is a diagram or a chart that shows an algorithm or process flow in sequential steps. Flowcharts are used in documenting the flow of steps in a given process or a procedure and are commonly used for designing and managing a process. They are oftentimes used in software, telecommunications and other industries where there is high process orientation.
Bullet Charts are useful in indicating the level of a single variable as measured against pre-set values such as the average, higher and lower quartiles etc. Normally the bullet chart will be shown as single prominent horizontal bar (repressing the given value) alongwith with a red mark indicating the desired or expected value. Other values such as the mean/average are represented as lighter colored bands around the bar representing the main data point. A bullet chart can be used in lieu of a bar chart and can be a better alternative due to its smaller size.
A Scatter Plot is a graphical representation of the value generated by a interplay of two or more variables. A scatter plot is also known as the scatter diagram or the scatter graph. Scatter plots can be two or three dimensional with the former being more widely used than the later. Each point in a scatter plot represents the value resulting from a given value of each input variable. The level of each input variable can be read from their respective axis of the chart.