PCA Correlation Tool Calculator

Calculator Input

Delimiter

Components to score

First row contains variable names

Paste dataset

Use rows as observations and columns as variables. PCA is computed from the correlation matrix, so variables are standardized automatically.

StudyHours	Attendance	Assignments	ExamScore
5	88	78	81
6	90	82	85
4	76	70	74
7	94	88	91
3	72	68	69
8	96	90	94
5	84	80	79
6	89	84	86

Standardization: z = (x − mean) / s, where each variable is centered and scaled by its sample standard deviation.

Correlation entry: r_ij = Σ(z_iz_j) / (n − 1), which forms the correlation matrix used in the decomposition.

Eigenvalue equation: Rv = λv, where R is the correlation matrix, λ is an eigenvalue, and v is the associated eigenvector.

Explained variance: Explained % = (λ / Σλ) × 100, which shows how much standardized variance each component captures.

Loadings: Loading = eigenvector × √eigenvalue, which links each original variable to each principal component.

Component scores: Scores = ZV, where Z is the standardized data matrix and V contains the selected component directions.

Paste your dataset into the text area. Keep observations in rows and variables in columns.
Select the correct delimiter and tick the header option when the first row contains variable names.
Choose how many component scores you want returned for each observation.
Click Calculate PCA to generate the correlation matrix, eigenvalues, loadings, explained variance, and scores.
Review the scree plot and the cumulative variance table to decide how many components to retain.
Use the CSV and PDF export buttons to save the generated results for reporting or documentation.