Use this tool when each sample is a mixture of three parts that add up to a constant, for example soil sand, silt and clay, microbiome phyla fractions or macronutrient shares. Upload your data below (CSV, Excel or SPSS, or paste it), optionally add a grouping column, and run the analysis.
About the ternary plot generator
A ternary plot is a triangle in which each point is a three-part composition. The closer a point is to a corner, the larger the share of that component. It is the standard figure for soil texture, mineral and alloy composition, microbiome fractions and mixture designs.
Shares that add to a constant cannot vary independently, so ordinary statistics on raw proportions can mislead. This generator converts each composition to isometric log-ratio (ilr) coordinates, so group centres, confidence ellipses and tests respect the constant sum, and it tests group differences with PERMANOVA on Aitchison distance.
When to use it
Use it to display and compare three-part compositions between groups: gut microbiome phyla in healthy and diseased groups, soil texture across sites, tissue or diet composition, and formulation experiments. If you have more than three parts, combine them into three meaningful groups first.
What this generator reports
A ternary plot with optional group colours, group centres (closed geometric means) and log-ratio confidence ellipses.
Group centres and descriptive statistics for each component.
PERMANOVA on Aitchison distance with pseudo-F, R² and a permutation p-value.
Options for gridlines, labels and axis titles, with the three components re-scaled to sum to 100%.
An APA7 results paragraph and an interpretation of every table and figure, available as a Word download after logging in.
Assumptions to check
Each row has three non-negative parts. Zeros need care because log-ratios are undefined at zero.
Observations are independent.
For PERMANOVA, groups should have similar dispersion; unequal spread can produce a significant result without a difference in centres.
Group sizes are reasonably balanced, and each group has enough samples to estimate a centre.
How to report it in APA7
Report the group centres (for example as percentages), then the PERMANOVA pseudo-F with degrees of freedom, R² and the permutation p-value, stating the distance (Aitchison) and the number of permutations. The generator writes this paragraph and lays out the table in APA7 style.
Customize and download
Change colors, fonts, font sizes and line widths in the figure style panel, then download each figure (PNG or SVG) and table (CSV), or the full set of figures, tables and interpretations as one Word file. Downloads need a free StatFas account; running the analysis does not.
Yes. You can analyze data and read all results without an account. Create a free account or log in only when you want to download figures, tables or the Word report.
Do my three columns have to add to 100?
No. The tool re-scales each row so that its three parts add to 100% before plotting.
Why use log-ratios instead of the raw shares?
Because the parts add to a constant, an increase in one forces a decrease in the others. Log-ratio coordinates remove this constraint, so means, ellipses and tests are valid.
Is my data uploaded to a server?
The analysis runs in your browser. Your data stays on your device while you work with the generator.