Saturday, January 26, 2019

Dog days of code

morning.fun.with.dogs.before.coding.all.day <- c("Copper", "Spike", "Gabby", "Kelly")


Unfortunately, that's probably the extent of R I'll code today. For better or for worse, Stata is just so much faster for the data cleaning/formatting I need to reexecute in order to prep my project's data for new analyses next week. You win some, you lose some.

Today I learned...
*Technically, I learned this on Thursday, not today, but it's too good not to share*
While humans see in three primary colors (i.e., our eyes have three types of cones, which then allow us to see red, green, and blue light), chickens have 12 cones in their eyes, and thus can see in 12 different primary colors! Can you imagine how chickens would describe colors if they could do more than squawk - or if human beings could chat with chickens in...Chicken-ese? about how they experience the world?!

I learned in Chris Adolph's CSSS569 course, Visualizing Data and Methods, which has been phenomenal thus far. I highly recommend it to anyone interested in the theory (cognitive psychology, hurrah!) and generation of effective visualizations of social science methods and data.

Today I'm grateful for...
A good night's sleep. I slept long and hard last night, which I desparately needed after a few weeks of various ups, downs, and stress involving the royal triad of school, work, and life matters. Physical and psychological lethargy had been dragging me down recently, so it was quite lovely to finally feel more spritely this morning.

Today's best part was...
Thus far, it's been solving some tricky automation code I needed to debug and troubleshoot today. I had spent various chunks of the last week trying to figure out how to code up a systematic way to select patients' baseline and endline biometric measures while accounting gaps in baseline measurement timing (i.e., at enrollment, sometimes patients didn't have A1c or blood pressure readings so we needed to use the next most recent measures), but hadn't been able to fully implement it across various datasets. 

No comments:

Post a Comment