A Basic Data Science Workflow

Developing a clean and easy analysis workflow takes a really, really long time. In this post, I outline the workflow that I have developed over the last few years.

Rebecca Barter

Developing a seamless, clean workflow for data analysis is harder than it sounds, especially because this is something that is almost never explicitly taught. Apparently we are all just supposed to “figure it out for ourselves”. For most of us, when we start our first few analysis projects, we basically have no idea how we are going to structure all of our files, or even what files we will need to make.

The intuition behind inverse probability weighting in causal inference

Removing confounding can be done via a variety methods including IP-weighting. This post provides a summary of the intuition behind IP-weighting.

Rebecca Barter

In my previous post, I introduced causal inference as a field interested in estimating the unobservable causal effects of a treatment: i.e. the difference between some measured outcome when the individual is assigned a treatment and the same outcome when the individual is not assigned the treatment. If you’d like to quickly brush up on your causal inference, the fundamental issue associated with making causal inferences, and in particular, the troubles that arise in the presence of confounding, I suggest you read my previous post on this topic.

Confounding in causal inference: what is it, and what to do about it?

An introduction to the field of causal inference and the issues surrounding confounding.

Rebecca Barter

Often in science we want to be able to quantify the effect of an action on some outcome. For example, perhaps we are interested in estimating the effect of a drug on blood pressure. While it is easy to show whether or not taking the drug is associated with an increase in blood pressure, it is surprisingly difficult to show that taking the drug actually caused an increase (or decrease) in blood pressure.