Home▸Statistics▸Factor Analysis Simulator

Factor Analysis Simulator

Interactive factor analysis simulator: watch correlated test scores decompose into hidden latent factors — rotate a 3D factor-score plot, inspect the correlation heatmap, and see each variable's loading vectors on the extracted factors.

Statistics3DAdvanced60 FPS

Details

CategoryStatistics

RenderingThree.js · WebGL

DifficultyAdvanced

Target FPS60

ControlsOrbit-drag to rotate the 3D factor-score plot, scroll to zoom; factors-to-extract slider (1-3) selects how many eigenvector axes are treated as real factors; correlation-strength slider (0.3-0.95) regenerates the dataset with weaker or stronger true loadings; Loadings toggle shows/hides the per-variable loading arrows; click a variable in the list to inspect its exact loadings and communality; New dataset button resamples the students.

Three.jsStatisticsFactor AnalysisPCAEigenvectorsData Visualization

Related simulations

A/B Test Lab: Peeking, SRM & BanditsWatch two live variants collect visitors while a p-value trace, a sample-ratio-mismatch…

PCA — Principal Component AnalysisRotate a cloud of correlated points onto its principal components, watching eigenvectors…

A/B Testing Pitfalls: Peeking, Sample Ratio Mismatch, and When to Use a Bandit InsteadSample-size calculation, the peeking problem, multiple-testing correction, and sample ratio… ⊞ All 23 in Statistics →

Six simulated test scores are generated from two hidden abilities plus noise, then the simulator computes their correlation matrix and extracts hidden factors by diagonalizing it — exactly the eigenvector step at the core of real factor analysis. Rotate the 3D plot to see students positioned by their estimated factor scores, watch the loading arrows show how strongly each test depends on each extracted factor, and click any variable to read its exact loadings. Adjust correlation strength to control how cleanly the two ability clusters separate, and choose how many factors to extract to see variance explained shift between real signal and noise.

⚙ Under the hood

Interactive factor analysis: six correlated test scores generated from two hidden latent abilities are decomposed by Jacobi eigen-decomposition of their correlation matrix, plotted as a 3D factor-score cloud with loading-vector arrows.

Three.jsStatisticsFactor AnalysisPCAEigenvectorsData Visualization

3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install