We are now in 2026. Happy new year, all.

The Courage to be Curious

When learners ask me what I look for when hiring a bioinformatician or data scientist, there is a specific quality that I look for I call data curiosity: the willingness and tenacity to ask and wonder questions about the data. If they have the data curiosity, and can demonstrate how they followed that curiosity.

If you can demonstrate where that curiosity takes you, all the better. A well written vignette is worth a thousand resumes.

This curiosity is a quality that is not encouraged in many educational systems, because it needs to be cultivated and encouraged individually. (I have seen a lot of original thinkers get squashed by educational systems.) This is where participating in communities such as the Data Science Learning Community can help.

What are some steps you can do to cultivate your data curiosity?

• Look at lots of different kinds of data, especially data types you're not familiar with. (Psst - this is what Tidy Tuesday is about!)
• Generate questions about the data you're curious about. This can be the hardest step - it's best to not be overly critical of yourself at this level. You can always edit these down - but don't prevent yourself from asking crazy questions at this point.
• Evaluate whether you can answer the questions and whether they make sense. Talk it over with someone else. Do a little research about the data - how was the data collected? How was a variable measured? How does that impact your question?
• Make visualizations, crosstabs, and filters that help you answer your question. Go with the simpler visualizations, and try to assess the relationships between your outcome variable and the other covariates.

What are some other ways you cultivate your own data curiosity? Leave your tips as a response to this post.

The Courage to be Wrong

I make a lot of mistakes. I'm in a fortunate position in that I can make mistakes without someone cutting me down. Making mistakes is an incredible privilege, and I try to use that privilege to help others.

I try to make use of that when I teach - it can be humbling when you are cut down, but oftentimes it can be damaging when you are new to the field.

I'm not an expert and I make mistakes. It was pointed out to me that the NHANES databot videos that I did some things that were questionable. I'm not a expert statistician when analyzing survey data - I know there are many intricacies when analysing survey data with weights.

I'm teachable - what were the things I did in the analysis that you objected to? Leave an answer in the comments.

Working with Databot on Endangered Languages

I have been using Posit's Databot as a tool for exploring data for Tidy Tuesday data. Here's an example of me using Databot for exploring the endangered languages dataset.

I know a little about doing spatial analysis with {sf} and {leaflet}, so I use that knowledge to start visualizing data about the languages of the world spatially.

Remember, Databot is not a flotation device. You need to use it responsibly and verify the code it generates is doing what you expect it to be doing. Don't turn off the critical parts of your brain when using it.

Some Wise Words from Travis Gerke

Annual Reminder: Spotify Wrapped is proof that sometimes count(x, sort=TRUE) is all the data science you need.

Thanks for Reading!

Take care of yourself, and let's get better and better at working with data.