💐Causal Mapping: a Garden of Ideas#
🌸 Starting to use the Causal Map app and want to know more about the theory and practice of causal mapping? What is it, what is it good for, how can you do it?
🌻 You've skimmed through a couple of our publications but want to see how it all fits together?
🪻 Here you'll find dozens of one-pagers setting out the key ideas in causal mapping as we see it, curated and assembled from existing publications and blog posts.
The titles of most of these "ideas" pages are expressed in the form of a single thought or claim or idea, like "Causal coding is easy to automate".
This Garden will also serve as a background companion to the Causal Map app. The app has its own Guide which is also integrated into the app itself, but that Guide is all about how to use the app whereas this Garden is a more discursive look at the theory and everything that surrounds causal mapping.
We have also included some of our most popular LinkedIn posts, especially on AI.
⚒️ This site is very much a work in progress! A lot more ideas are still to come!
Some essential terminology for causal mapping.
What is causal mapping? What are its strengths and weaknesses? How does a causal map differ from a systems diagram? This chapter has some answers.
At Causal Map we are relatively agnostic about data collection.
In this chapter we present some of key general principles about how to do causal mapping which we at Causal Map Ltd (and, most of the time, at BathSDR) have adopted.
In this chapter we will cover some (optional) special conventions which we use to make causal mapping more powerful and easier to apply.
The fundamental output of causal mapping is a database of causal links. If there are not too many links, this database can be visualised "as-is" in the form of a causal map or network. But usually there are too many links for this to be very useful, so we apply filters.
Because the output is a structured network, we can apply a range of queries to explore the data. This gives us a library of pre-existing approaches to ask practical questions about the causal landscape described by the participants...
Causal mapping has been used in many different fields. In this chapter we look at how it can be applied in evaluation; its strengths and weaknesses.
Causal mapping is also a kind of Qualitative Data Analysis (QDQ). How does that even work? This chapter explains.
This guide shows how to create a simple causal map in Kumu.io using a spreadsheet. It includes:
Causal coding is fascinating but can take a lot of time. Using AI to help you is pretty easy, especially if you provide a codebook of factor labels which the AI has to use, as we will explain in this chapter, when we've finished it!
Causal coding is fascinating but can take a lot of time. Using AI to help you is pretty easy, especially if you provide a codebook of factor labels which the AI has to use: [[000 Intro -- deductive auto-coding]].
How can we improve rigour and even reproducibility when using AI in social science? This chapter suggests some answers.
There are many different ways to collect data for causal mapping: [[000 Task 1 -- Introduction]].
Here is version of the final text.
Here are some thoughts from a couple of years ago when genAI first hit us, plus some thoughts about where we are going with it.
Licenses, how to cite Causal Map, and a bibliography.
🎉 Welcome to CausalMap!!