Summary
Linkr ships a library of ready-made plugins covering most common descriptive and statistical analyses — without writing a line of code. Each widget is configured in a dedicated panel, with a live preview.
The catalogue
These widgets are plugins shipped with Linkr: you add them from Add a widget, the Plugin tab (they carry a Built-in badge). You then pick the columns to analyse, an optional grouping variable, and the appearance; the result updates live (see Configuring a widget).
Scatter, line, bar, histogram, boxplot, violin.
Aggregate value, icon and optional mini-chart.
Descriptive statistics: counts, %, means, medians, IQR.
Kaplan-Meier: curves, log-rank, median, at-risk table.
t-test, Mann-Whitney, χ², Fisher, ANOVA, Kruskal-Wallis.
Linear and logistic: coefficients, odds ratios, forest plot.
Pearson or Spearman, p-values, significance.
Flows and transitions: patient pathways, funnels.
Geographic points (longitude / latitude), colour, size.
Widget by widget
Plot builder
A versatile chart builder: scatter, line, bar, histogram, boxplot, violin. You pick the X and Y variables, a colour/grouping variable, and a per-entity aggregation (first, last, mean, median, min, max, sum) when an individual has several values. Legends, axes, colour palettes, opacity and point size are adjustable.
Key indicator
A single highlighted aggregate value (mean, median, sum, count, proportion, standard deviation, quantiles…), with an icon, a colour, a statistical subtitle and, optionally, a mini-chart (histogram, bar, pie).
n = 3,558 · SD 15.2
Table 1
The classic descriptive table of a clinical paper: counts, percentages, means (± SD), medians (with interquartile range), min/max, missing values. An optional grouping variable compares two or more populations, column by column. You choose which metrics are shown.
| Characteristic | Survivors | Deceased | p |
|---|---|---|---|
| n | 2,847 | 711 | |
| Age, mean (SD) | 62.4 (15.1) | 71.8 (12.3) | <0.001 |
| Female, n (%) | 1,310 (46) | 291 (41) | 0.02 |
| SAPS II, med. [IQR] | 34 [26–44] | 52 [41–63] | <0.001 |
| Ventilation, n (%) | 996 (35) | 498 (70) | <0.001 |
Survival analysis
Survival analysis: Kaplan-Meier curves, log-rank test, median survival and an at-risk table. Confidence bands and censor marks can be shown; a grouping variable compares several strata.
Correlation matrix
A heatmap of correlations between numeric variables, as Pearson or Spearman coefficients. Optional display of p-values and highlighting of significant correlations at an adjustable α level.
| Age | SAPS II | Lactate | LOS | |
|---|---|---|---|---|
| Age | 1.00 | 0.42 | 0.18 | 0.31 |
| SAPS II | 0.42 | 1.00 | 0.55 | 0.48 |
| Lactate | 0.18 | 0.55 | 1.00 | 0.27 |
| LOS | 0.31 | 0.48 | 0.27 | 1.00 |
Statistical tests
A widget that picks the right test automatically: t-test, Mann-Whitney, χ², Fisher, ANOVA, Kruskal-Wallis. You switch between parametric and non-parametric approaches, set the significance threshold, and choose the columns shown (statistic, degrees of freedom, p-value, confidence interval, effect size).
| Variable | Test | Stat. | p |
|---|---|---|---|
| Age | Student's t | t = 9.2 | <0.001 |
| SAPS II | Mann-Whitney | U = 6.1e5 | <0.001 |
| Sex | χ² | χ² = 5.3 | 0.02 |
| Lactate | Mann-Whitney | U = 4.8e5 | 0.003 |
Regression
Linear and logistic regression, with multiple variables and interaction terms. The widget produces the coefficient table, model statistics and, if needed, a train/test split or cross-validation for predictions.
Sankey diagram
The flow visualisation: patient pathways, selection funnels, state transitions. Three data shapes are accepted (rows = steps, columns = levels, or a path string). You can hide flows that are too rare or too frequent, and show the diagram, a table, or both.
Map
A geographic visualisation based on Leaflet: marker positioning, clustering, with colour, size and label driven by a variable.
Warehouse widgets
Beyond these analyses, Linkr offers patient-data widgets (demographic summary, timeline of measurements, clinical notes). They belong to exploring individual data rather than composing dashboards; they are covered in the Exploring and analysing section.
When the catalogue is not enough
If no plugin in the catalogue matches your need, two options:
- Custom code in R, Python or SQL right inside a widget — ideal for a one-off, custom analysis.
- Build a plugin — to package a reusable, configurable, shareable analysis that then appears in the catalogue.