Statistical rigor.
Reproducible notebooks.
The data-science tool you don't have to apologize for. Every result shows the Python, every notebook re-runs, and every test reports the effect size — not just the p-value.
Hypothesis tests built in
A full stats toolkit. No R, no SPSS, no jargon.
Welch's t-test
Independent samples, unequal variance + Cohen's d effect size
One-way ANOVA
F-statistic + Tukey HSD post-hoc + eta-squared
Chi-square + Cramér's V
Independence test + categorical effect size
Shapiro–Wilk + KS
Normality dual-check with QQ-plot data
Pearson + Spearman
Linear + monotonic correlation with significance
OLS regression
Coefficients, p-values, VIF, residuals — full statsmodels output
Workflow
From CSV to a publishable result in four cells.
Upload your experiment CSV
Treatment group, control group, every measurement. WorkLiq auto-detects which columns are numeric and which are categorical. No prep.
Pick the test — or let WorkLiq pick
t-test for two groups, ANOVA for three or more, chi-square for categorical, a normality check when you're unsure. It suggests the right one from the data's shape.
See the result, effect size & interpretation
A p-value alone is thin. WorkLiq reports Cohen's d, eta-squared, Cramér's V — and explains in one sentence whether the difference actually matters.
Export the notebook for peer review
Every cell carries its Python. A reviewer downloads the .ipynb, re-runs it locally, and gets the same number. No “trust me” science.
Real research scenarios
What labs actually use WorkLiq for.
Ready to test your hypothesis?
Upload a CSV, pick two variables, and get the test, the effect size, and a plain-English interpretation — with the Python shown.
Start free →