Data Literacy for the Social Sciences: Define the data need

Like all good research, when when planning the type of data you will need, first start with a reasonably well defined research question. Then, to determine where to search for datasets, consider these questions:

• How could your research question be measured? (e.g. by recorded observations, a survey instrument, demographic counts)
• What agency or organization likely gathers data on your topic?
• What population(s) will you study?
• What are the geographic constraints or units?
• What are the time constraints (a range of years; monthly, quarterly or annually)?
• Do you need cross-sectional data (observations of many different subjects at a given time) or longitudinal data (tracks the same type of information on the same subjects at multiple points in time)?
• Do you need quantitative data (measurements, counts, rankings) or qualitative data (texts, surveys, opinions, words), or both?

Questions adapted from Partlo, Kristin. 2009. "The Pedagogical Data Reference Interview." IASSIST Quarterly 33, (4): 6-10. Available at: https://iassistquarterly.com/public/pdfs/iqvol334_341partlo.pdf.

Next:

• Conduct a literature review to identify existing data sets that are used by studies relevant to your research question, as well as gaps in these datasets.
• Following a literature review, consider how your research question addresses existing gaps and whether generating new data sets will help you answer that question.

A useful tool for keeping track of potential dataset sources and research methodologies is the Data Reference Worksheet.