Data from 1998-2015
This page collects the data collected between 1998 and 2015 for download and use on your computer. It is not available in our web-based search or API, and is provided here as part of the historical record.
Full data from 1998-2015
This is a single zip file containing all the Census data from 1998–2015. When decompressed, there is a separate folder for each year.
- census-1998-2015.zip (413MB, SHA1)
By audit year
This is the same data as above, broken out by audit year.
- census-1998.zip (15MB, SHA1)
- census-1999.zip (16MB, SHA1)
- census-2000.zip (16MB, SHA1)
- census-2001.zip (20MB, SHA1)
- census-2002.zip (22MB, SHA1)
- census-2003.zip (23MB, SHA1)
- census-2004.zip (22MB, SHA1)
- census-2005.zip (23MB, SHA1)
- census-2006.zip (24MB, SHA1)
- census-2007.zip (24MB, SHA1)
- census-2008.zip (24MB, SHA1)
- census-2009.zip (27MB, SHA1)
- census-2010.zip (32MB, SHA1)
- census-2011.zip (32MB, SHA1)
- census-2012.zip (30MB, SHA1)
- census-2013.zip (30MB, SHA1)
- census-2014.zip (28MB, SHA1)
- census-2015.zip (29MB, SHA1)
How to work with this data
We describe both the data itself, and how to get you started exploring and doing work with this data.
Data dictionary
This data is provided as CSV (comma-separated value) files. CSV files are, in many ways, like spreadsheets; you can use Excel, Google Docs, or an open source alternative (like LibreOffice) to open these files.
Each file has a header row, and then the rest of the file is data.
Using the data files in a spreadsheet
The 7 files beginning with the characters ELEC
contain the data.
ELECAUDITHEADER
contains information similar to thegeneral
table in the GSA FACELECAUDITS
contains information similar to thefederal_awards
tableELECAUDITFINDINGS
is similar to thefindings
tableELECPASSTHROUGH
is similar to thepassthrough
tableELECCPAS
,ELECEINS
, andELECUEIS
are similar to the tablessecondary_auditors
,additional_eins
, andadditional_ueis
For example, you could:
- unzip the file
census-1998.zip
, - open the file
ELECAUDITS.csv
, and - sum the column
AMOUNT
.
This will tell you what the total (audited) federal expenditures were in 1998.
Using the data files in code
The CSV files should import cleanly using tooling like SAS, SPSS, or any number of other programming languages (PowerBI, Python, etc.). If you work with code, you might write something like the following to get the same total expenditures as above.
import pandas as pd
file_path = "ELECAUDITS.csv"
chunk_size = 10_000
total_expenditures = 0
for chunk in pd.read_csv(file_path, chunksize=chunk_size):
total_expend_series = chunk.loc[:, "AMOUNT"]
total_expend_numeric = pd.to_numeric(total_expend_series, errors="coerce")
total_expenditures += total_expend_numeric.sum()
print(f"There were ${total_expenditures} audited")
to achieve the same result. We calculated $485,687,753,099 audited in 1998.
Cleaning up the data
Note that these datafiles may need cleaning and manipulation before you can use them. For example, the TOTFEDEXPEND
column in ELECAUDITHEADER
contains both numbers and text. This is how the data was collected, and we present it here as-is.