PCA biplot
Show observation scores and variable directions in the same reduced-dimensional view.
Chart data · 22 rows
| Variable | PC1 loading | PC2 loading |
|---|---|---|
| Diabetes | -2.71 | 1.06 |
| Obesity | -2.22 | -0.06 |
| COPD | -2.71 | -0.99 |
| Heart dis. | -2.83 | -0.36 |
| Depression | -1.02 | -2.33 |
| Arthritis | -2.27 | -1.61 |
| Asthma | -1.52 | -1.85 |
| Stroke | -2.8 | 0.44 |
| High chol. | -1.84 | 0.48 |
| Cancer | 1.17 | -2.51 |
| High BP | -2.67 | 0.37 |
| Fair/poor | -2.82 | 0.71 |
| Mental dist. | -2.48 | -1.21 |
| Phys. dist. | -2.83 | -0.15 |
| Inactivity | -2.65 | 0.77 |
| Binge drink | 1.9 | -0.65 |
| Short sleep | -2.32 | 0.37 |
| Smoking | -2.61 | -0.89 |
| ADI | -2.75 | 0.28 |
| ADI:Finance | 2.13 | 0.64 |
Page 1 of 2
The chart labels the 12 longest loadings; the table includes all variables. Correlation-based PCA (variables standardized). Scores and loadings co-scaled for a proper biplot.
Captured ZIP health and deprivation PCA reference fixture
Preview
How to read it
The cloud contains observation scores. Nearby arrow directions suggest variables that align in this projection. Axis labels report the variance captured by each component.
When to use it
Exploring correlated multivariate data after the analysis and its standardization choices have been established.
Watch out for
Do not treat two dimensions as the whole dataset, infer causation, or read absolute arrow lengths as original measurement units.
Data shape & units
scores: { x, y }[]; loadings: { x, y, label }[]; variance: { x, y } in 0–1 proportions, not 0–100 percentages.
Missing values
Filter invalid scores and loadings without silently replacing them with zeros. Document upstream handling of missing survey answers.
Controls & accessibility
The reusable component spreads loading labels and limits their number for legibility. A separate table lists every loading, including any not labelled on the plot.
Use the component
Install from the CTZN registry, then pass your prepared rows. Examples below illustrate the component contract.
npx shadcn add @ctzn-viz/pca-biplot<PcaBiplot scores={scores} loadings={loadings}
variance={{ x: 0.42, y: 0.18 }}
ariaLabel="Principal components of survey attitudes" />
