# Putting it all together: Messy data exploration

For a real-life example of dealing with messy data, we're going to use a very important contemporary data set: COVID-19 reporting in Massachusetts.

# Finding the data in context

This website is where the State of Massachusetts makes its public Covid-19 datasets available.
- Spend some time looking at this site to see what kind of contextual data is available. Who is generating this data? How do we know that it is accurate?
If we wanted to make a map of COVID cases by town, we'd want to access the "Raw Data." Here's what that link looks like:

But if you click that link, you're not on a map page or a metadata page. You've just downloaded an Excel file to your computer, maybe without even realizing it!
- To simplify this exercise, we've copied this file to our cloud storage so that you can look at it in your browser. See it in Excel Online by clicking here.

This Excel spreadsheet is a pretty good example of the very messy data sets that you're likely to find "in the wild."
Spend some time clicking around the file together to see if you can answer the following:
- Which tab helps you understand what is going on in this document?
- Which tab or tabs might be most useful in making a map of Covid-19 cases?

Once you've made it to the tab Weekly_City_Town, you'll find yourself with some information that seems like it relates to geography
- Massachusetts only has 351 cities and towns, but this workbook has over 20,000 rows! What do you think is going on here?
- What attributes from this data do you think might be useful for making a map?
- Can you see any problems with the data that might confuse mapmaking software?
- What would you need to do in order to relate the City/Town column to a shape on a map?