Good correlations and bad flashcards
This week's article is a long one, on correlations! Even if you think you know correlations, there's probably a thing or two you don't.
Hello! I hope your week is going well.
New articles
The Surprising Richness of Correlations
When Fisher wrote Statistical Methods for Research Workers, it was one of the first books ever on frequentist statistics, so nobody knew how to present the topic. Fisher opted to do it through the perspective of correlations. This is a very unique point of view, which more people should be acquainted with!
Full article (10–25 minute read): The Surprising Richness of Correlations
Flashcard of the week
This week I will do something unusual and present an anti-example. This is a terrible, awful, really bad flashcard I made.
If one has an X = √(2T) where T is chi-squared, what is the mean of X?
The answer is
√(2n - 1) where n is the degrees of freedom of T.
I'm sure this is a useful fact – else I wouldn't have written it down. But I cannot for the life of me tie it into anything else I know. Since there's no framework in my brain on which to hang it up, I will not ever remember it. Every time I get this flashcard I will go "Oh no, not this again. I have no idea."
That means it's a bad flashcard and I summarily deleted it. If I run into this fact again in a context where it makes more sense to me, I will create the flashcard again. For now, it has no value to me.
With spaced repetition, one cannot be scared of creating new flashcards – but one also must not be scared of deleting ones that don't serve their purpose!
Premium newsletter
One question I have received a lot is if I can put together a short list of the best books I've read, along with brief summaries. I did that, and sent it out in the fourth premium newsletter that went out last week. (It also has some information on how the list was constructed.)
You can get access to this list (as well as other past and future issues of the premium newsletter) for only $2/month. As interest increases and I learn to hit my stride with publishing these, the price will be set higher for future subscribers. You can cancel at any time, no questions asked.
To upgrade, click the subscription link at the top of this newsletter and fill in your email again.
Your opinions
I cannot improve without feedback. Reply to this email to share your thoughts on any of the topics above, or anything else!