Lollipop Plot Explained: How to Make, Read and Report It
A lollipop plot displays one value per category as a dot on a stem. It conveys the same information as a bar chart with less ink, and it is useful for ranked values, deviations from a reference and many categories. This post covers the general lollipop chart, not the mutation lollipop plot used in genomics.
When it works well
One numeric value per category, with a meaningful order (ranked or grouped). Viewers judge position along a common scale accurately (Cleveland & McGill, 1984), so the dot position carries the message; keep the axis starting at zero if stems encode length, as with bars.
Original example (illustrative)
Mean satisfaction (1–5) for six university services: Library 4.6, IT 4.1, Registry 3.8, Advising 3.5, Cafeteria 3.2, Parking 2.4. The overall mean is 21.6/6 = 3.60. Deviations from the mean are +1.0, +0.5, +0.2, −0.1, −0.4 and −1.2, which sum to zero as they must. Sorting the lollipops from highest to lowest shows immediately that Library and IT sit above average and Parking is the clear outlier.
Plain means conceal uncertainty. If you show these as statistics, add confidence intervals or n per service; otherwise small differences such as Advising versus Cafeteria may be noise.
How to make it in StatFAS
Open the Lollipop Plot Generator. Provide a category column and a numeric value column. The existing StatFAS post covers only the mutation lollipop plot, so link it as a distinct topic.
Reading and reporting
State what the dots represent (mean, percentage, count), the sample size and any sorting rule. Caption: Figure 1. Mean satisfaction by service (1–5 scale), sorted from highest to lowest.
Common mistakes and alternatives
Truncating the axis, leaving categories unsorted, using it for data with several values per category (use box or violin plots; see the violin plot guide), and crowding too many labels. Alternatives: bar charts (more familiar) and dot plots (Cleveland, 1985).
Conclusion
A lollipop plot is a clean bar-chart alternative for ranked values, best paired with uncertainty and clear labelling.
1.
statfas-lollipop-plot-interface.png, alt: "StatFAS Lollipop Plot Generator interface"2.
statfas-lollipop-plot-example-data.png, alt: "Example data entered in the StatFAS Lollipop Plot Generator"3.
statfas-lollipop-plot-settings.png, alt: "Lollipop Plot Generator settings in StatFAS"4.
statfas-lollipop-plot-result.png, alt: "Lollipop Plot Generator result generated in StatFAS"References
- Cleveland, W. S., & McGill, R. (1984). Graphical perception. Journal of the American Statistical Association, 79(387), 531–554.
- Cleveland, W. S. (1985). The elements of graphing data. Wadsworth.