Yes. Every calculation runs inside your own browser. Your dataset is never uploaded, stored or sent anywhere, including Excel files, which are opened locally. The only thing Rigora receives is the email address you confirm to unlock the result.

Should I report p-values in Table 1?

For randomised trials, no. CONSORT advises against it because any baseline difference in a randomised trial is by definition due to chance. For observational studies a comparison can be informative. The tool leaves p-values switched off by default and names the test used for each variable when you switch them on.

Mean and SD or median and IQR?

Report mean with standard deviation when a continuous variable is reasonably symmetric, and median with interquartile range when it is skewed, which is common for biomarkers, length of stay and many laboratory values. The tool judges this per variable from the shape of your data and you can override it.

Free tool, built by physician-scientists

Build a publication-ready Table 1 from your dataset.

Table 1 is the baseline characteristics table at the start of almost every clinical study. Paste or upload your data, choose the grouping variable, and get it formatted with the right summary for every variable. Ready for your methods section in seconds.

Your data never leaves your browser Distribution-aware summaries Word, PDF and R code export

1

Add your data

Excel, CSV or tab-separated · first row = column names

Upload Excel or CSV

2

Configure your table

Sensible defaults are pre-selected

Group by (columns) Variables to include · drag to reorder, or use the arrow keys

Continuous variables shown as

Show a “Missing” columnCounts missing values per variable Add p-values comparing groupsOff by default. Test chosen per variable.

For randomised trials, reporting p-values in Table 1 is discouraged (CONSORT). A baseline difference in a randomised trial is by definition due to chance.

Your Table 1

Email confirmed. Your table is unlocked.

Your Table 1 is ready

We only receive your email. Your data stays in your browser.

Why the details matter

A Table 1 is easy to make and easy to get wrong

This tool gets the mechanics right. The judgement calls, the ones reviewers notice, are where a statistician earns their fee.

Distribution

Mean or median?

Reporting mean with SD for skewed data such as biomarkers, length of stay or lab values misleads the reader. We choose per variable from its shape, and you can override it.

Randomised trials

p-values that should not be there

In a randomised trial a significant baseline difference is chance by definition, and CONSORT advises against those p-values. Knowing when to drop them shows you understand your own design.

Completeness

Missing data, stated plainly

Silent denominators are a common desk-reject trigger. We surface how much is missing per variable so nothing gets buried and you can decide how to handle it.

The real work

Table 1 is not your analysis

Baseline description is the start. Adjusted models, survival analysis, sample size and the reporting reviewers scrutinise are where conclusions are won or lost.

When it has to be right

Let a named physician-scientist run the analysis behind your paper

Rigora's statistics service is led by a physician-scientist with 82+ peer-reviewed publications and active peer-review work for high-impact journals. You get the analysis, publication-ready figures, the methods text and a fully reproducible R script. Flat fee in EUR, with a money-back guarantee.

See statistics packages Book a free scoping call Free sample size calculator

Questions

What is a Table 1 in a research paper?

Table 1 is the baseline characteristics table at the start of almost every clinical study. It describes who was in the study, usually split by treatment or comparison group. Continuous variables are reported as mean with standard deviation or median with interquartile range, and categorical variables as counts and percentages.

Is my data safe?

Yes. Every calculation runs inside your own browser. Your dataset is never uploaded, stored or sent anywhere, and that includes Excel files, which are opened locally. The only thing we receive is the email address you confirm to unlock the result. This matters for unpublished work and patient data, and it is how a GDPR-conscious workflow should behave.

Why do I have to enter a code from my email?

To confirm the address is real and yours. We send a 6-digit code to the address you enter, and you type it back to reveal your table. It takes a few seconds and means we can actually reach you with your table and, later, with help on the full analysis.

Which file formats can I upload?

Excel .xlsx files, CSV, tab-separated and plain text. Older .xls files and OpenDocument .ods files need to be saved as .xlsx or CSV first. Excel files are unpacked and read inside your browser, so they are never uploaded.

What summaries and tests does it use?

Continuous variables are summarised as mean with SD when reasonably symmetric and as median with IQR when skewed, judged from the shape of your sample. Categorical variables are shown as counts and percentages within each group. If you switch on p-values, the tool picks the test per variable, using a t-test or ANOVA for symmetric continuous data, Mann-Whitney or Kruskal-Wallis when skewed, and chi-square or Fisher's exact for categorical variables. Every test used is named in a footnote.

Can I use this table in my manuscript?

Yes. It is meant to drop straight into your paper. Export it to Word, save it as a PDF, or take the R script and reproduce it yourself. You do not need to cite the tool. If your study is high-stakes or a reviewer has already raised statistical concerns, have the underlying analysis reviewed by a statistician before you submit.