Week 10 “viz”: “Interview” a dataset
This week is another “viz,” meaning that you’re not actually making a data visualization. Instead, I would like you to interview a dataset and write up your findings. When interviewing, think about the stuff you’ve learned from the How to Talk Back to a Statistic from How to Lie with Statistics, as well as the things we discussed from The art of the interview and How to ‘interview’ a big pile of data. Some questions you may want to consider:
- Where did you come from?
- Who says so?
- What’s missing?
- What’s the min? The max? The mean?
- Anything interesting from pivot tables?
You can interview any dataset you’d like, but a couple good ideas would be:
- Our community partner data
- The General Social Survey (either your own extract or mine)
- Some data from the Minnesota Department of Education, particularly the North Star File. (This is what we’ll be using for the One Number Story.) The MDE has a guide to uses of different data sources, which could be a good resource.
I would like just a single document this week, either an HTML file or a PDF. If you write your interview results in Quarto, render to HTML or PDF, and if you work in Word, print or save to PDF.