Exporting results
Pooled can export a single table as CSV or Excel, or a selected piece of the pooled file in several formats. Values that are missing are written as empty cells.
Quick downloads of one table
- On a dataset page, use Download and choose CSV file or Excel file. The export is the table for the tab you are viewing, with the Countries and Years filters applied.
- On All datasets, the CSV and Excel buttons export the table currently selected in the list, also filtered.
| Format | File name | Contents |
|---|---|---|
| CSV | the table name and date, for example Coverage_-_National_2026-01-31.csv | One table, header row, one row per country, year and area |
| Excel | pooled_rmncah_2026-01-31.xlsx | One sheet named after the table |
Extract a piece
Use Extract a piece when you need several tables, or only part of them. The chips at the top choose Datasets, Countries and Years; choosing nothing means all. Under Columns, keep All columns or choose the columns per dataset: country, year and area columns are always kept. The Preview shows up to six rows per dataset.
The Get the piece card counts datasets, countries, years and rows, then offers three formats:
| Format | Result |
|---|---|
| CSV | One CSV for one dataset, or a .zip with one CSV per dataset for several |
| Excel | One workbook with one sheet per dataset |
| Pooled file | A smaller .rds you can reopen in Pooled |
Files are named with the domain and date, for example pooled_rmncah_2026-01-31_piece.zip or _piece.xlsx. Sheet names are limited to 31 characters and the characters not allowed in Excel sheet names are replaced by spaces. Excel sheets have a bold header row, a frozen first row and automatic column widths. Explore this piece opens the result in the app. Start again resets all filters and columns.
Excel and CSV files do not contain the build log. To keep the list of countries, the files that were left out and the log, keep the pooled .rds from Build pooled file.
What is inside the pooled file
The .rds is a plain R list, not a country cache. It holds domain, the creation time, the app version, a table of countries (country, ISO3, source file, warning), a table of left_out files with reasons, datasets (one table per dataset) and a time-stamped log. Tables are ordered as in the app’s dataset list: Parameters and score, Indicator coverage, Coverage, Mortality, Service utilization, Data quality, Denominators, Estimates and surveys, Equity, Health systems.
Next steps for analysis
- Read the CSV or Excel file in R, Stata, Python or Excel.
countryandiso3identify the country, and coverage columns already share names across countries. - For a full analysis in R, read the pooled file with
readRDS()and work onpooled$datasets. - Compare your results with the methods in the analytical framework, including national coverage.
- For issues with a single country, return to its cache in the RMNCAH App, fix it, save, and rebuild the pooled file.
Using the DataSuite chat assistant
The DataSuite chat assistant works on a single country’s dataset open in the RMNCAH or Vaxx app; see the AI assistant page. The pooled file has a different structure (a list of tables): ask the assistant to read the file and analyse its datasets directly, and check the results against the tables you downloaded.