Love Data Week

Love Data Week Day One: Find Data

Let Penn Libraries Help You Find a Dataset!

If you know a little bit about what you want to do, we can help you find your data match:

Know what scale of data you are looking for?

If you could imagine the smallest unit you'd like to analyze, would it be individual people, households, firms, or something else? What is it that you hope to draw conclusions about?

• Macro data?
• Micro data from Worldbank or GDELT?

Know the type of data you are looking for?

What should the data tell you about these people or other units? How would you measure that, and what kind of categories would you create? These are the variables you need.

• Quantitative data
• Qualitative data (try the Qualitative Data Repository!)
• Images
• Text
• Open metadata

Know the analytical method you'd like to use to explore the data?

• Social network analysis
• Text mining and analysis
• Geospatial analysis
• Machine learning
• Natural language processing

Some of our Data Source Faves!

Dataset search engine - Google Dataset Search 

• Kaggle - 
• UPenn Library Frequently Used Databases 
• Data repositories catalog
• U.S. Government’s open data
• ICPSR
• Roper Center
• IPUMS
• WRDS
• NBER 
• IQSS Dataverse Network
• Vanderbilt TV news archive

No idea where to begin?

Questions to Ask Yourself

• Do you know the topic / geographic location / subject area of the data you are looking for?
• What timeframe of data do you need? (Do you need data from a particular historical period? Do you need a snapshot (i.e. cross-sectional data) or changes over time (i.e. a time series)? Is the series yearly, weekly, once a decade?) Then try: 
  • Data World 
  • Google Dataset Search 
  • re3data.org (universal data repositories catalog)
• What location are you focusing on? (Do you need to know about a particular place--a city, county, state, or country? Within that place, are there smaller areas you would like to compare, e.g. neighborhoods within a city?) 
  • See our guide to finding spatial data
• What is the source for the data? (Why would someone record data on this subject? If you know who would be interested, then you can infer where you might find it. For example, the Centers for Disease Control is interested in the spread of diseases, so they might be a source for health data.)