Pro Statistics Toolkit
Advanced calculators for research analysis
Multiple regression, ANOVA, chi-square goodness-of-fit and Bayesian updating — with visual output for every result.
Multiple Linear Regression Calculator
Fit a model with one or more predictors, with full coefficient statistics.
About this calculator
Enter one row per observation as comma-separated values: the dependent variable (y) first, followed by one or more independent variables (x₁, x₂, …). For example:
10, 2, 515, 3, 720, 4, 9
- 1 independent variable → simple regression with a 2D scatter and line
- 2 independent variables → multiple regression with a 3D scatter and plane
- 3 or more independent variables → coefficients and statistics are shown; visualisation is not possible beyond 2 predictors
Result
One-Way ANOVA Calculator
Compare the means of three or more independent groups.
About this test
One-way ANOVA tests whether at least one group mean differs significantly from the others. It reports:
- F-statistic and degrees of freedom (between, within)
- p-value, computed from the exact F-distribution
- η² (eta squared), the proportion of variance explained by group membership
Result
Boxplot — shows the spread and outliers within each group.
Group means ± 1.96 SE (approx. 95% CI) — non-overlapping bars suggest a real difference.
Chi-Square Goodness-of-Fit Calculator
Test whether observed frequencies match an expected distribution.
About this test
Goodness-of-fit compares your observed category counts against what you’d expect under a hypothesis (e.g. an equal split across categories). Leave the expected field blank to assume all categories are equally likely.
Result
Observed vs. expected frequency, by category.
Standardized residuals — bars past ±1.96 flag categories driving the result.
Bayesian Updating Calculator
Update a probability estimate in light of new evidence.
About Bayes’ theorem
Bayes’ theorem updates a prior probability into a posterior probability once evidence is observed:
P(H|E) = P(E|H) × P(H) / P(E)
- Prior — your belief in the hypothesis before seeing evidence
- Sensitivity — P(evidence | hypothesis is true)
- False positive rate — P(evidence | hypothesis is false), i.e. 1 − specificity
Result