Repository navigation
Releases: Edgar-Mendonca/Datum-Plot
Release list
Datum Plot 0.3.11
Dates and times.
Dates in tables
- The X column of an XY table can hold dates or dates and times: right-click its heading and choose how they are shown (5 Nov 2026, 05-Nov-26, 2026-11-05, 05/11/2026 or Nov 2026). Nothing changes for tables you leave as numbers, and older files open as before.
- Type or paste dates in many forms: 2026-11-05, 2026-11-05 14:30, 5-Nov-26, 5 Nov 2026, Nov 5, 2026, 05/11/2026; impossible dates (31 Feb) are refused. File › Options › Dates sets whether 05/11/2026 is day first or month first (the first time, from your system's language).
- Dates are kept as numbers, days since 1 January 1970, so everything else works on them unchanged.
Dates on graphs
- A date X axis with automatic ticks (the year once, then months or days), or day and month, month and year, ISO or week numbers (Format Graph › X Axis), with the range typed as dates. Hovering shows the date, and every export keeps the date ticks.
Analyses and exports
- Straight-line fits on dates give the slope per day. Workbook exports (.xlsx, .ods) write real dates.
From Datum Plan
- Datum Plot 0.3.11 is the first version that opens Datum Plan's time charts with dates (
xFormat: "date", days since 1970): they open on a date axis with no questions asked.
New sample
- Readings over time: weekly dissolved oxygen at two river sites, on a date axis, with the fall per day.
Datum Plot 0.3.10
Control charts and design of experiments.
Control charts (Analyze › Quality › Control Chart)
- Individuals and moving range for one value per row, X̄ and R or X̄ and S for subgroups (each row a subgroup in its sub-columns), and the p chart for defectives out of the number inspected.
- 3σ limits, points beyond them and runs of 8 on one side of the centre line are flagged; give the specification limits for Cp and Cpk.
Design of experiments
- New Design makes a two-level factorial design as a table: 2 to 7 factors, full factorial or half fraction, replicates and centre points, the factors coded −1/+1 and the runs in a random order (row titles). Type each result in the Response column.
- Factorial Analysis gives each factor's and interaction's effect with its test (Lenth's method when there are no replicates), a Pareto chart of the effects and the main-effect plot.
Checked: the control limits equal the textbook constants (A2, D4, A3, B4, 2.66, 3.267) and the binomial limits of the p chart; known effects are recovered from a simulated design; designs are balanced and orthogonal.
Datum Plot 0.3.9
More statistics: three-way ANOVA, ANCOVA, agreement, bootstrap confidence intervals and spline fits.
Compare groups
- Three-way ANOVA: a grouped table whose rows are the first factor and whose data sets are named by two levels with a slash ("Young / Drug", "Old / Placebo"…). Main effects and every interaction, with effect sizes.
- ANCOVA: an XY table with the covariate in X and one data set per group. Checks that the slopes are parallel, then compares the groups' adjusted means (with Šidák comparisons) and draws the parallel lines on the graph.
Relationships
- Agreement: a contingency table of paired ratings (the first rating in the rows, the second in the data sets). Cohen's kappa with its 95% CI and strength, weighted for ordered categories, and McNemar's test (2 × 2) or Bowker's test (larger tables) for a systematic difference.
Statistics and fitting
- Bootstrap confidence intervals of each data set's mean, median or SD, and of the difference between the first two, without assuming a distribution (BCa, repeatable).
- Spline fit: Smooth has a smoothing spline whose smoothness is chosen automatically; it is drawn on the graph.
Fixed
- Double-clicking an item in the Analyze or Graph gallery now closes the gallery behind the window it opens.
Checked against independent results: kappa by its formula, McNemar's exact test by the binomial distribution, three-way sums of squares by hand, ANCOVA by nested models, the spline by SciPy.
Datum Plot 0.3.8
Vector fields and 3D scatter graphs.
New graph styles for tables of many variables (Graph › Graph Style, or Format Graph › Graph)
- 3D scatter: the first three variables as X, Y and Z, coloured by a fourth when there is one. Tilt and turn the view in Format Graph.
- Vector field: an arrow at each X, Y with its two components from the next two variables, coloured by its length. Arrows are scaled so the longest is about the distance between neighbouring arrows.
Datum Plot 0.3.7
Ternary, waterfall, radar and bubble graphs.
New graph styles (Graph › Graph Style, or Format Graph › Graph)
- Waterfall (XY tables): each data set's curve is lifted above the last and labelled at its end, for spectra, runs or temperatures. The lift is automatic, or set it in Format Graph.
- Ternary (tables of many variables): the first three variables of each row as shares of 100 % on a triangle, for compositions.
- Bubble (tables of many variables): X, Y, bubble size from the third variable, and colour from a fourth if there is one, with a size key and a colour bar.
- Radar (grouped tables): the rows are the spokes and each data set is an outline through its means, for comparing profiles.
- A graph style a table cannot make says why (a radar needs at least three rows, a ternary three variables).
New sample
- Composition triangle: copper, nickel and zinc in 18 alloys on a ternary graph; switch to Bubble to see hardness as bubble size.
Datum Plot 0.3.6
Multiple t-tests and volcano plots.
Multiple t-tests
- Analyze › Multiple t-tests (in Compare Groups): a grouped table with one row per gene, protein or compound and two data sets (for example Control and Treated) with their replicates. Each row gets its own test: Welch's or ordinary unpaired t-test, paired t-test, or the Mann-Whitney test.
- The P values are corrected for the number of rows by the false discovery rate (Benjamini-Hochberg, or the two-stage method of Benjamini, Krieger and Yekutieli) or by Holm-Šidák or Bonferroni. Rows below the threshold (q < 0.05 by default) are discoveries, and the answer says how many went up and down.
- The results list every row by P value with the means, the difference, the log₂ fold change, t, P and q.
Volcano plots
- The result shows a volcano plot: log₂ fold change (or the difference) against −log₁₀ P, discoveries coloured and the strongest labelled by their row names.
- New table from result turns it into an editable graph, with every row's name kept.
Checked
- Each row's test equals SciPy's, and the false discovery rates equal the Benjamini-Hochberg step-up worked by hand.
Datum Plot 0.3.5
Repeated measures and mixed-effects two-way ANOVA; Copy Linked is removed again.
Repeated measures
- Two-way ANOVA has a new Repeated measures option for grouped tables where the same subjects are measured again (one sub-column per subject): in every row (for example at each time point), in every data set, or in both.
- With complete data it is a repeated measures ANOVA with the right error term for each effect and the Geisser-Greenhouse correction for sphericity. When values are missing it fits a mixed-effects model instead, so no subject is thrown away.
- Multiple comparisons within each row or each data set, with stars on the graph, come from the fitted model. The results show the variance between subjects and the sphericity estimates.
- Checked against independent calculations: the fully repeated design matches a reference repeated measures ANOVA exactly, and the classic and mixed-model tests agree for complete data.
Removed
- Copy Linked (0.3.4) is gone: it depended on other programs' linking steps and was too awkward to use. Copy Picture, Export, Slides, Document and Workbook remain. Projects saved with 0.3.4 open as before.
Datum Plot 0.3.4
Copy Linked: slides and documents follow your graph.
Copy Linked
- Graph › Copy Linked (also on the graph's right-click menu) keeps the graph as a picture in a folder beside the project, and redraws it every time you save. Insert that picture into your slides or document as a link once: from then on they show the new graph whenever their links are updated, in any program. The Linked Picture window gives the steps; its path is already on the clipboard.
- Tools › Update Linked Pictures redraws every linked picture at once; Stop Linking ends it for one graph.
- A project file can only name pictures inside its own pictures folder.
Roadmap
- The next releases are planned in Help › Roadmap: repeated measures and mixed-effects ANOVA, multiple t-tests and volcano plots, ternary, waterfall, radar, bubble, vector and 3D scatter graphs, more statistics, control charts and design of experiments, wavelets, a graph digitizer, Bayesian tests, and then Mac and Linux versions.
- The roadmap now says plainly that two-way ANOVA is ordinary only for now; repeated measures and mixed models for two and three factors come next.
Datum Plot 0.3.3
Testers' requests: colour schemes beside the graph, OpenDocument export, and plain names for file types.
Colours
- Graph › Colour Schemes now sits beside the graph, so each scheme you click shows on the graph at once. Its table is compact again (the colour columns were far too wide).
- Schemes whose colours stay easy to tell apart for colour-blind readers carry a small colour-blind safe label: Okabe-Ito (the default), Bright, Muted and Cividis. We checked them by simulating the three common kinds of colour-blindness.
- Format Graph › Data Sets shows each data set's colour as it is drawn in the chosen scheme, and its quick colours are that scheme's colours.
- Every window can be moved by dragging its title bar.
OpenDocument
- Tools › Workbook saves .xlsx or .ods, Tools › Slides .pptx or .odp, and Tools › Document .docx or .odt: choose the type in the Save dialog. Datum Plot writes the OpenDocument files itself.
Plain names
- Buttons and texts name the file type, not another company's product: Workbook, Slides and Document; texts say "a spreadsheet", "a document" or "slides". The "Classic" colour scheme and the "Open axes" and "Framed box" graph styles have plain names too.
Datum Plot 0.3.2
The Datum Plot Guide in English, Chinese and French, and five new samples.
Guide
- Help › Guide (and the Guide tile on the File page): a step-by-step guide with numbered pictures, from your first table to analyses, graphs, layouts, export and the Python console, with a "which test when" table, a glossary and shortcuts. Choose English, 中文 or Français: the PDF downloads from this release, is checked against its checksum, is saved in your Downloads folder and opens. The installer stays the same size.
New samples
- Survival study: months to relapse on two treatments, with Kaplan-Meier curves, the log-rank test and the hazard ratio.
- Method comparison: a bedside glucose meter against the lab, with Bland-Altman bias and limits of agreement.
- Diagnostic test: a biomarker in healthy people and patients, with the ROC curve, its AUC and the best cut-off.
- Response surface: process yield over temperature and time, with a contour map and a quadratic surface fit that finds the best settings.
- Polar pattern: the radiation pattern of two antennas on a polar graph.
Installer
- An installation in a folder of your own choosing is updated in place after last version's publisher change.