Teaching
More than seven years of tutoring quantitative methods and applied statistics, and courses built on one rule: start from a question the learner already cares about, then teach only the mathematics needed to answer it.
I have tutored quantitative methods and applied statistics for more than seven years. The same problem comes up again and again: learners can follow a procedure but cannot say what it is for, because they met the formula before they had a reason to want it.
The courses below are built the other way round. Each session opens with a concrete question, meets the idea as a picture or a simulation before its formula appears, and ends with code the learner runs on real or clearly labelled data. Each item is published with the materials that exist so far, such as notebooks, lab scripts and interactive tools you can open and use, and says plainly what is still being built.
The same approach can carry into professional training: refresher courses for analysts who are about to work with regressions, survey weights or indicators, and short sessions for programme teams on reading evidence and designing a test.