Wow, thanks so much for all the replies, everyone! I tried using z-scores like most of you suggested, and it worked pretty well. I also checked for outliers and did a quick log transform, which helped a lot.
One quick follow-up though—how do you guys handle data that’s not normally distributed? I’ve got a few columns that are pretty skewed, and I’m not sure if z-scores are still the best option. Any tips? Thanks again, you all rock!
One quick follow-up though—how do you guys handle data that’s not normally distributed? I’ve got a few columns that are pretty skewed, and I’m not sure if z-scores are still the best option. Any tips? Thanks again, you all rock!