Subject Guides: Data Science: Finding Data

Data Librarian

she/her/hers

Email Me

Open Data

Finding data for your own research

Finding Data and Statistics
Guide to statistical information resources by subject, for U.S. and international.
Online databases that AU Library subscribes to containing numeric data
The database descriptions provide an overview of their contents and any important access information. Selected ones from this large group are listed further below.
AU Library FAQ "I need a dataset for the quantitative section of my paper"
Google Advanced Search
Can be used to focus searches on specific file types, such as Excel, and Internet domains, such as .ca.gov for California government agencies or .edu for accredited education institutions in the USA. (For more on domains, see: https://www.pcmag.com/encyclopedia/term/internet-domain-name.) Can also be entered directly into basic Google search box, for example (for latest Excel file format):

pollution data site:.ca.gov filetype:xlsx

Caveat: just because a file is in Excel format does not mean it contains analyzable data.
Eleven quick tips for finding research data
A 2018 article on finding research data, natural/physical science-centric but with some generally applicable advice.

Selected (out of over 100) online resources that AU (Library) subscribes to that contain numeric data - these are suitable for statistical analyses:

ICPSR: Inter-University Consortium for Political and Social Science Research This link opens in a new window
This resource includes both publicly available and non-publicly available datasets on public policy and social science research.
**Access to datasets requires a free user-specific login.
To create or log into your ICPSR account, click on the 'Log In' button found in the upper right corner of most ICPSR pages.**
Additional information, particularly about off-campus access, is available in the library's Accessing ICPSR datasets at AU Subject Guide.
Gallup Poll Social Series (GPSS) microdata This link opens in a new window
Datasets (in SPSS format) and their codebooks (in Excel format) from the Gallup Poll Social Series (GPSS). For each month, the dataset contains a common set of basic indicator variables, and different sets of topical variables, for years going back to 2001, as indicated in its codebook.
For more information on the GPSS, please see How Does the Gallup Poll Social Series Work?
** Items in this collection are labeled 'under embargo' and can be accessed when connected to the secure campus network (includes AU VPN). **
Cross-National Time-Series Data Archive This link opens in a new window
The Cross-National Time-Series Data Archive, launched in 1968, contains domestic conflict, political, legislative and economic data (around 200 variables) on over 200 countries from 1815 onward.
iPOLL and RoperExpress (Roper Center) This link opens in a new window
U.S. and international public opinion polls (iPOLL) and datasets (RoperExpress) from 1935 to present. Includes the major academic, commercial and media survey organizations.
DataCite Commons This link opens in a new window
Search for works that have been issued a Digital Object Identifier (DOI) by DataCite. Narrow results to "Work Type: Dataset" to select only those from search results. Results can be hit-and-miss, but datasets on "niche" topics may be found here.