TopBoxTornado for multiple ratings

38d87d9852a66eea29da00ae16511664

Updated at October 16th, 2019

A TopBoxTornado plot is a concise way to present top-box scores, and rate multiple items on Likert-type scales.


Market research surveys often ask respondents to rate multiple items e.g., "How much do you agree with the following statements?" The response scale could be 5-, 7-, 10- or 11-point scales.

An effective way to present the results is to look at top-box scores, showing percentages for the top one, two or three scale items, e.g., "Somewhat agree or Strongly agree."

However, there are some instances where the client also wants to know the bottom-box score. The tornado plot is a visually interesting way to show both top- and bottom-box. The middle scale points (and any missing/NA values) are hidden.

Export to PowerPoint

These also export to PowerPoint as native Office chart objects:


Create a TopBoxTornado

To create a TopBoxTornado chart in Protobi, press the blue square edit icon for a group of rating questions and select "Edit JSON...." Then specify a chartType attribute:

    {
        "chartType": "TopBoxTornado"
    }

The default is to show the bottom-two box scores on the left and the top-two box scores on the right, using the format attribute to identify the boxes and sorting them to identify top vs bottom.

You can override the defaults by specifying a chartOptions attribute as well:

    {
        "chartType": "TopBoxTornado",
        "chartOptions": {
           "left": [1,2],
           "right": [4,5],
           "leftTitle": "Very strong or strong conflicts",
           "rightTitle": "Not very strong or no conflicts",
          "leftColor": "purple",
          "rightColor": "gold"
    }

The left and right options specify one or more values to include. You can list one, two or more. If you don't list any, Protobi assumes the first value from the format attribute.

If no title is specified, the titles are inferred from the formatted values.

The colors can be any CSS color, e.g., "#39C" or "blue".

Was this article helpful?