{"id":573,"date":"2026-04-01T19:27:47","date_gmt":"2026-04-01T19:27:47","guid":{"rendered":"https:\/\/miriamposner.com\/classes\/dh150s26\/?page_id=573"},"modified":"2026-04-01T19:27:49","modified_gmt":"2026-04-01T19:27:49","slug":"milestone-a-identify-a-dataset","status":"publish","type":"page","link":"https:\/\/miriamposner.com\/classes\/dh150s26\/assignments\/data-storytelling-project\/milestone-a-identify-a-dataset\/","title":{"rendered":"Milestone A: Identify a Dataset"},"content":{"rendered":"\n

Due: In class, Tuesday, April 28<\/strong><\/p>\n\n\n\n

If you’re going to tell a story with data, you need…data! For this milestone, you’ll identify and obtain the dataset you plan to use in your group’s data storytelling project.<\/p>\n\n\n\n

What to choose<\/h2>\n\n\n\n

Topic<\/h3>\n\n\n\n

You should find a dataset related to an issue of social justice you care about and want to explore in depth. Of course, seldom do we find a freely available Golden Dataset that embodies everything we want to know. So it’s likely that your process of selection will be a dialogue between what you want<\/em> to do and what’s actually practical<\/em>.<\/p>\n\n\n\n

Size and format<\/h3>\n\n\n\n

I do not have hard-and-fast guidelines about the size or format of the dataset you use, but I do have suggestions. You’ll want the dataset to be large enough that it makes sense to work with it programmatically (as opposed to by hand). The number of records can vary, but it’s especially good if your dataset has a lot of attributes<\/em>, because that way you can see how different properties interact with each other.<\/p>\n\n\n\n

If you’re comfortable working with code, your dataset can be in any format. However, if you’re still figuring things out, it’s often helpful if your dataset is formatted as (or converted to) a CSV (AKA a spreadsheet). That way, you can open it in Excel and perform simple analyses. Plus, almost all visualization software can accept a CSV.<\/p>\n\n\n\n

APIs and other retrieval techniques<\/h3>\n\n\n\n

By April 28, you should have the data in-hand\u2014not just identified a source for it.<\/p>\n\n\n\n

If you’ve identified an API or some other kind of endpoint where you want to retrieve data, I’d like you to actually<\/em> retrieve the data for this milestone. That is, you should have it saved somewhere as a file. That’s because it’s often harder than people think to work with a new API, and you don’t want to be surprised down the road. It’s OK if the dataset changes later; I just want to make sure you know how to obtain it.<\/p>\n\n\n\n

Similarly, if you’ve found a resource (e.g., a table in a book or a library database) from which the data you want can be scraped, transcribed, or obtained in some other way, do that work before<\/em> the milestone is due. Again, that’s just so you aren’t surprised by unforeseen complications.<\/p>\n\n\n\n

Where to look<\/h2>\n\n\n\n

This is where you’ll put on your Sherlock Holmes hat! As you might be aware, there’s no one repository where you can reliably find the data you want, so you’ll probably have to do a fair amount of Googling.<\/p>\n\n\n\n

Here are some suggestions.<\/p>\n\n\n\n