Simple Data Sets: 3 Small Tables to Practice and Publish

If you want to practice turning a spreadsheet into a public page, start with a table whose rows and columns you can explain in one sentence. Use the fictional CSV below to check the publishing workflow, or choose Iris or Palmer Penguins for practice with real observations.

A useful starter dataset has one clear row type, a few well-defined columns and a question readers can answer by filtering or comparing values. Small size helps, but clear definitions matter more.

1. Fictional library loans: a copyable starter CSV

Use this for: checking headers, dates, numeric sorting and a simple summary.

Copy this into a plain-text file named sample-library-loans.csv, or enter the same values in a spreadsheet and export it as CSV:

date,branch,loans_count
2026-01-05,East,18
2026-01-05,West,25
2026-01-06,East,21
2026-01-06,West,30
2026-01-07,East,15
2026-01-07,West,27

These six rows are invented for practice. They do not describe a real library. Each row represents one branch on one day; loans_count is the number of items loaned that day.

Try answering: Which branch had more loans across these three days? East totals 54 and West totals 82, for 136 overall. Those totals give you a checkable reference when preparing a summary.

For publication, keep dates in the consistent YYYY-MM-DD form shown above and counts as numbers without words such as “items” inside the cells. Label the public page “Fictional library loans — practice dataset” so nobody mistakes it for reporting.

2. Iris: a small table with no missing values

Use this for: grouping observations and comparing measurements.

The UCI Iris dataset contains 150 observations: 50 in each of three iris classes. Each row represents a plant. It has four measurements in centimetres—sepal length, sepal width, petal length and petal width—plus a class column. UCI reports no missing values.

Download the comma-delimited iris.data file. Despite its extension, it contains comma-separated values and starts with an observation, not a header. Import it as comma-delimited text, then add this header above the first observation:

sepal_length_cm,sepal_width_cm,petal_length_cm,petal_width_cm,species

Save the prepared table as CSV. It should still contain 150 data rows; adding column labels must not replace the first plant’s measurements. For more on this check, see CSV header row vs data row.

Try answering: How does petal length vary across species? Filter by species and compare the measurements, or make a species-level summary with observation counts.

Before publishing, record which file you used. UCI’s dataset notes identify differences between its data and Fisher’s article, so “Iris” alone is not a sufficient version description. UCI licenses the dataset under CC BY 4.0; when sharing an adapted table, provide appropriate credit, link the license and indicate changes, such as adding headers.

3. Palmer Penguins: practice documenting missing values

Use this for: working with categories, measurement units and incomplete observations.

The simplified Palmer Penguins dataset has 344 rows and eight columns covering three penguin species from three islands in Antarctica. Each row represents a penguin; its fields include species, island, bill dimensions, flipper length, body mass, sex and year.

The ready-to-save penguins CSV already has a header row. Measurement columns include their units: millimetres for bill and flipper dimensions, and grams for body mass. The file uses NA for missing values.

Try answering: How does body mass vary by species? When calculating averages, treat missing measurements as missing—not as zero—and report how many measured observations contributed to each average. Keep the original file unchanged; document any replacement of NA with blank cells in your publication copy.

The project’s license and additional-use guidance lists the data as CC0 and provides recommended citations. Separately, it asks anyone interested in publishing the data to contact the data collector about analysis and collaboration on final products. Read that guidance before publishing; distinguish the project’s request from its stated license.

Prepare the file, then check the public page

Before uploading any of these simple data sets:

• Keep one header row. Remove title rows, merged cells and totals embedded among observations.
• Preserve the row meaning. Do not automatically remove identical measurements; identical values do not prove duplicate observations.
• Write a source note. Include the dataset name, source, access date, units, missing-value convention and changes made. Clearly label invented data.
• Review for public release. Never upload sensitive information, private records or credentials.

TablePage accepts CSV, TSV, XLSX and XLS files and generates a public dataset page with a filterable table. Upload only the reviewed publication copy.

Before sharing the link, compare the page with your file: check the data-row count, first observation, column labels and numeric values. For the fictional sample, confirm that filtering to East leaves three rows. Share the page alongside its source and interpretation notes so readers know what the table represents—not just how to explore it.