The redacted metadata is split into multiple files to make it more likely that they could be opened in a spreadsheet program. We initially did this by congressional year, but most files are still too large, so do this by calendar year instead. Researchers with enough knowledge to care about congressional year should be able to determine from the dates which congressional year they are from.