Setting your data up for success

sschuff

I wanted to share a resource that has helped me think about how to set up data for painless analysis. I have seen many variations of messy data including multiple variables stored in one column, variables stored in both columns and rows, and different data types in the same table. Tableau works very well when data is set up following the rules in the article, easily aggregating and grouping data how you want. The one exception I have found is when you want to perform calculations across a variable like a time series. I hope this helps for your next project! Feel free to share your best practices for data cleaning and data structure.

Tidy Data by Hadley Wickham: https://www.jstatsoft.org/article/view/v059i10

mroberts

Nice post Sam! This reminded me of a site I had used back when in Chicago. We would use this to clean datasets prior to posting. I think OpenRefine has maybe changed a bit since then (was Google Refine) but this might be useful to others.
https://digital.cityofchicago.org/index.php/how-a-table-becomes-a-dataset-openrefine/