In the previous chapter Task 2 -- Introduction we looked at the main ideas of minimalist coding:
- we code links between simple propositions which we do not necessarily conceive of as variables which can take different states at different times.
- we let factor labels do most of the work, so we start off by distinguishing between, say
wealthandpoverty. - we don't use separate metadata columns for factors. In fact, we don't even have a table for factors at all: our coding results only in a table of links.
That approach has general applicability. In this chapter we get down to some specific suggested conventions about how to make this approach more useful for answering concrete questions. These suggestions are all implemented in the Causal Map app, but of course other approaches are possible.
All the pioneers of causal mapping used different algorithms to simplify, query and synthesise their data.
In most use cases, the data contains many "co-terminal links": links with the same cause and effect. We call these "bundles" of links. It is often bewildering to show all of those links in a map so by default we condense them all into a single visual arrow, usually printing the number of sources and/or citations on the arrow and making the width depend on the number of citations.
You can use the special separator ; to create nested factor labels, like this:
Causal mapping most often looks for linearity first, while of course being on the lookout for feedback loops and circular shapes. Whereas most systems approaches do the opposite.
In the first part of this Guide we have dealt only with undifferentiated links which simply say “C causally influences/influenced E” or more precisely “Source S claims/believes that C causally influences/influenced E.” We call this “barebones style” causal mapping. There is nothing more to this kind of causal map than links between factors. No other features are used.
In this section, we’ll look at “Context” as it appears to us at Causal Map when we do actual (qualitative) causal mapping: taking causal claims which real-life stakeholders actually make and trying to encode them in as systematic a way as possible. We believe that doing qualitative causal mapping is a really good testbed for theoretical ideas in evaluation and social science: do they fit with what people actually say?
Causal mapping doesn’t usually deal with the kind of non-causal themes which are the focus of ordinary QDA (like in NVivo!). However sometimes it can be really useful to be able to simply note the presence of something without any causal connection.
How can we compare different groups like districts, gender or questionnaire sections, within maps and tables?