Screeplotit

Screeplotit is a software tool designed to assist researchers in creating scree plots and related diagnostics for multivariate analysis. The primary purpose is to visualize the distribution of eigenvalues obtained from dimensionality reduction methods such as principal component analysis (PCA) to aid in selecting the number of components to retain.

Its features typically include automatic computation of eigenvalues from data matrices or from user-supplied values, generation of scree plots and cumulative explained variance plots, and options to overlay parallel analysis thresholds. The tool emphasizes clarity and reproducibility, offering customizable axes, labels, colors, and export formats such as SVG, PNG, and PDF. It may also provide bootstrap-based confidence bands for eigenvalues and optional annotations of elbow points.

A typical workflow begins with loading a dataset, performing PCA, and passing the resulting eigenvalues to screeplotit to render the plots. Analysts interpret the elbow or the point where the explained variance gains diminish to decide how many components to retain. Some implementations also compare different methods (e.g., eigenvalue > 1 rule) or include parallel analysis guidance.

Screeplotit is described as cross-platform and available as a library for programming languages commonly used in statistics, such as Python and R, with various packaging options and documentation. It is released under an open-source license and maintains a repository with issue tracking and examples.

See also: Scree plot, Principal component analysis, Dimensionality reduction, Eigenvalues.