How to Create a Box Plot in Excel Prerequisite knowledge: Box Plot also called a box-and-whisker plot provides a horizontal or vertical graphical representation of a distribution of data where the end points are the minimum and maximum values (the whiskers) connected to a rectangular box whose ends represent the first and third quartiles with an embedded line parallel to the ends representing the median of the data. Excel does not have a tool that would automatically create a box plot. Therefore, we need to prepare the data and perform some calculations before we can create the plot. 1. Launch Excel. 2. Input some data and create two additional tables to prepare and calculate the values used by the box plot. For this example, we use fictitious sales data. Algebra 1 Summer Institute, 2014 Page 1
3. Begin by setting up the calculations for the quartiles. Entering the formulas on the left should result in the values on the right. Enter the formulas shown into the corresponding cells and hit enter. The formulas should automatically cause the values on the right to be calculated. Note that the percentile function uses a decimal value to indicate the correct percentile value to be calculated. 4. Highlight the cells that have the data values. Copy and paste these formulas to the other three columns. Completing this step correctly should result in the following values in the table. Algebra 1 Summer Institute, 2014 Page 2
5. Next we need to prepare the data for our chart in a series of rows. The values in the Data for Chart table represent the cell locations from the Summary Range table. The =J6 in cell J15 takes the minimum value from the J6 cell. The remaining cells in the North column of the Data for Chart table calculate the difference between the current value and the previous value. For example, =J7-J6 calculates the difference between the values in J7 and J6 (965-630=335). The blue arrow below indicates the resulting values after the cell calculations. 6. Copy and paste the cell data from the North column in the Data for Chart table to the other three columns as done in a previous step. Completing this step correctly should result in the following values in the table. It s this data that will be used to draw the box plot. Algebra 1 Summer Institute, 2014 Page 3
7. To start the box plot, highlight the labels and the first four rows of data. We are purposely not including the data for Series 5. We will use it later. 8. On the insert ribbon, go to the charts section, select column type chart. Then, select the 2D stacked column. Algebra 1 Summer Institute, 2014 Page 4
9. Once the chart is inserted, drag it over to the left-hand side and resize if desired. 10. The chart needs to be modified to resemble a box plot. Looking at the chart, we can see the spread of our four quarters of the data represented by blue (lowest 25%), red (second 25%), green (third 25%), and purple (highest 25%). To start, we will hide the Series 1 data. Algebra 1 Summer Institute, 2014 Page 5
11. Once the Format Data Series box pops up, select Fill and check No Fill. Then select Border Color and check No line. After selecting close, you will see that the data is still there, but now it is invisible. 12. Now repeat these same steps to make the Series 2 data (red) invisible as well. This time, leave the invisible Series 2 data selected as shown here. These will be reformatted. Algebra 1 Summer Institute, 2014 Page 6
13. Under Chart Tools, select the Layout tab. Then select the Error Bars pull down menu and select More Error Bars Options. 14. The Format Error Bars window should display. Here select Minus and choose Percentage and enter 100 in the percentage field. Close the window. Algebra 1 Summer Institute, 2014 Page 7
15. You can now see the whiskers that represent the minimum values for each of the four regions. 16. Next select Series 4 (purple). Repeat the previous steps to display the Format Error Bars window again. This time, select Plus. Then select Custom and click on the Specify Value button. Algebra 1 Summer Institute, 2014 Page 8
17. The Custom Error Bars window displays. Now we ll use the Series 5 data. Place your cursor in the North cell of the Series 5 row (cell J19). Click and drag to select the other three cells (to the right) in the Series 5 row. Then click OK on the Custom Error Bars window. Close the Format Error Bars window. 18. The box plot is now finished. Delete the legend since it is no longer meaningful by selecting the legend box and pressing the Delete key on your keyboard. Algebra 1 Summer Institute, 2014 Page 9
19. The box plot is now complete displaying our data, the min, max, mean, and first and third quartiles! 20. To make the box plot horizontal, follow the same steps except when selecting the chart type, choose Bar and then select the 2-D Stacked Bar button. This document is based entirely on the work of Eugene O Loughlin s YouTube video cited below. Source: https://www.youtube.com/watch?v=l_roxgxiwpu Eugene O Loughlin How To Draw a Simple Box Plot in Excel 2010 YouTube 29 August 2012 For additional information on using the analysis tools, refer to the Algebra 1 Summer Institute resource repository. Algebra 1 Summer Institute, 2014 Page 10