I want to practice my data wrangling skills but can't seem to find any dirty data.

Trifacta is my main focus right now

At the very least can you help me with ideas on how to practice this?

2 Answers 2


You can start with clean data and perturb them artificially. This procedure has the advantage that you have a ground truth to compare your data cleaning results with.


Since this is ancient history, I'll give a short update on my preferences for obtaining data.

Webscraping. Webscraping is the best way to obtain interesting data.

Here is a reddit post outlining how to start webscraping: https://www.reddit.com/r/webscraping/comments/fsmsqe/learn_how_to_web_scrape_or_use_online_tool/

In addition to everything in that post, I also love GraphQL, because with GraphQL introspection you can see all the objects available to you! Then you just act like a frontend developer working for that site, requesting to your heart's content.

I don't want to self-promote too much, but I do a lot of Youtube videos showing how easy it is to get and clean data using just a few lines of JavaScript.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.