Why this exists
Most statistics calculators return a number and stop. Paste your data, receive 4.7213, close the tab. If you already understood the method that is a convenience. If you were trying to learn it, or check a result by hand, or explain to a colleague how a figure was reached, it is worthless — you finish exactly as capable as you started.
DatrixCalc is built the other way round. The final number is the smallest part of every page. What matters is the working: the deviation of each observation from the mean, the sum of squares, the degrees of freedom, which critical value was looked up and why, the residual for every point. That is what a statistics tool should show, and almost none of them do.
Who it is for
- Students checking coursework, who need to reproduce the intermediate steps rather than just the answer.
- Researchers sanity-checking a figure from software before it goes into a paper.
- Analysts who need to explain to a non-technical stakeholder where a number came from.
- Anyone who has been handed a spreadsheet and a question and needs a defensible answer quickly.
How we build
- Client-side only. Every calculation is JavaScript running in your browser. Nothing you paste is transmitted, logged or stored. There is no server-side processing on this site at all.
- The working is always visible. Not behind a toggle, not summarised — the actual intermediate table, on the page, updating live.
- Assumptions stated. Where a method needs approximate normality, a minimum sample size, or independent observations, it is written down. Where an approximation breaks (a proportion at the boundary, a slope from six points) the tool says so rather than returning a confident wrong number.
- Methods documented. Quantiles use linear interpolation between order statistics, matching R type 7 and Excel PERCENTILE.INC. The normal CDF uses the Abramowitz & Stegun 7.1.26 approximation, accurate to about 1.5 × 10−7. Critical t values are exact tabulated figures to df = 30, interpolated between the standard anchors above that.
- No dependencies. Each page is a single self-contained file. No frameworks, no external fonts, no third-party libraries loading before the content.
- Slowly. Five thorough instruments beat thirty thin ones. Tools get added when they are finished, not when they run.
Why privacy matters here specifically
The data people paste into a statistics calculator is frequently confidential — unpublished research, clinical measurements, salary figures, exam marks, commercial performance. A tool that transmits that to a server is asking for a level of trust it has not earned and does not need. Doing the arithmetic in the browser removes the question entirely. There is nothing to leak because there is nothing to send.
What we are not
DatrixCalc is not validated statistical software and does not claim to be. It has not been through the verification process that R, SAS, SPSS or Stata have. It is an educational and checking instrument. For anything submitted, published or used in a regulated context, reproduce the result in software your institution or field recognises, and cite that. See the disclaimer.
We are also not a statistics consultancy. We cannot tell you which test to run on your data, whether your study design is sound, or how to interpret a result in your field. Those questions need someone who knows your context.
Funding
Display advertising, placed so it never sits between you and a result. No sponsored content, no affiliate links dressed as recommendations, no paid placements in the article text. Nothing you enter into a calculator is used for ad targeting, because none of it is ever transmitted. The privacy policy sets out exactly what advertising partners can and cannot see.
What is queued
Twenty-seven further instruments, in rough order: p-value from any test statistic, A/B test significance, one- and two-sample t-tests, combinations and permutations, the normal, binomial and Poisson distributions, percentiles and quartiles, outlier detection, chi-square, odds ratio and relative risk, effect size, and a frequency distribution builder. If there is one you keep needing, send it through — requests reorder the queue.
Corrections
If a figure here is wrong, we want to know immediately — a statistics site with an arithmetic error is worse than no site. Corrections go to the front of the queue ahead of every other kind of work. Use the contact form or email corrections@datrixcalc.com, and include the exact inputs you used so we can reproduce it.