It can be hard to describe, much less find, data you want. The VT library's Discovery Search does not distinguish data from other kinds of information. It won't help you identify aggregate numerical data in tables nor datasets of microdata analyzed with sophisticated quantitative or qualitative software.
We provide Data Planet both as a
- tool to discover with scope of data from around the world and, often more important, to point you to data providers who might have more for you
- and as a rich and varied source of numerical information in its own right, with some interesting visualization options
It's tempting to start search engines to find numerical and geospatial data. This is a good start for finding what data exist about something. They may not be as effective if you're interested in the variety of data available about a place.
- Their reliability depends on how dataset providers comply with technical standards for describing data (ie, metadata). As with any search, your search terms have to match the works the people who compiled the data put in the titles or headings of tables.
- Only a few of our data providers (notably ICPSR, Roper Center, Harvard's Dataverse network) permit searching for data by variable name (which varies with the researcher).
- Often the best way to start finding data is by asking yourself what kind of agency or business or research institution might have an interest in counting the people, things, behaviors (or whatever) that you want data on. Go to the source and dig around. Some words/labels that can signal where data lurk: data, dataset. indicators, repository, statistics, archive, visualization.
Compilations of data, such as RAND State Statistics, PolicyMap, and traditional statistical abstracts usually group aggregate data according to predefined themes that may reflect the goals and practices of the compilers rather than your own.
- News articles will often identify researchers who have not yet published their data in academic sources.
- When you find a relevant table of data use it as a discovery tool: scrutinize source notes and other annotations and trace back to the providers for further -- perhaps more recent -- data.
- Sometimes, the words in notes, titles, and column headings that can be effective search terms.