Skip to content
Grad Coach

Diagram · CC BY 4.0

Likert scale chart example (diverging stacked bars)

Five survey items as diverging stacked bars, centered on neutral and sorted by agreement.

Diverging stacked bar chart of five Likert items on a five-point agreement scale. The data are illustrative, made up for this example, with 200 respondents per item. Each bar is centered on the neutral midpoint, with agreement to the right of the zero line and disagreement to the left. The "Neither agree nor disagree" share is split in half across zero, following Robbins and Heiberger (2011). Items are sorted by the percentage who agree. The columns on either side of the plot give each item's total percentage disagreeing (left) and agreeing (right).

Swipe sideways to see the whole diagram, or download the PNG.

Likert scale diverging stacked bar chart example. Derek Jansen, Grad Coach, CC BY 4.0.

What it shows

A diverging stacked bar chart, the layout Robbins and Heiberger recommend for Likert data. Each of five items from an illustrative course-evaluation survey is one bar, split at the neutral midpoint so that agreement runs to the right of zero and disagreement to the left.

The neutral share is divided in half across the zero line. Items are sorted by the percentage who agree, and each item’s totals are given in columns on either side. The data are invented for the example, with 200 respondents per item.

How to read it

  1. The right-hand column gives total agreement. “The course was well organized” has the most, at 78%, so it sits at the top.

  2. Read across a bar to compare the two sides. For “Feedback was timely”, 44% disagree and 39% agree, so its bar sits furthest to the left.

  3. The gray in the middle is the neutral response, half of it on each side of zero.

  4. Gray-blue marks disagreement and orange agreement, a pair that remains distinguishable under the common forms of color-vision deficiency.

The numbers behind it

  • 200 respondents per item. Counts are strongly disagree, disagree, neither, agree, strongly agree.
  • The course was well organized: 6, 14, 24, 92, 64.
  • I would recommend the course: 8, 16, 30, 78, 68.
  • The readings were useful: 10, 24, 36, 84, 46.
  • The workload was manageable: 22, 46, 40, 66, 26.
  • Feedback was timely: 30, 58, 34, 52, 26.

Ways to use it in teaching

  • Give students the counts and have them rebuild the chart in a spreadsheet or in R, including the split neutral category.

  • A plain stacked bar chart of the same counts makes a useful comparison: which shows the weakest item more clearly?

  • In a survey-design module, it can open a discussion of whether a neutral midpoint belongs on the scale.

For designing the questionnaire behind data like these, see our survey design guide.

Sources

Robbins, N. B. and Heiberger, R. M. (2011), "Plotting Likert and other rating scales", JSM Proceedings, Section on Survey Research Methods, American Statistical Association, 1058–1066; Heiberger, R. M. and Robbins, N. B. (2014), "Design of diverging stacked bar charts for Likert scales and other applications", Journal of Statistical Software 57(5): 1–32, doi:10.18637/jss.v057.i05.

Statistics

Related diagrams

All open educational resources