🌻 Working Papers
Chapter contents.

This chapter is a set of working papers about causal mapping as qualitative evidence management: we code reported causal influence claims from text as a links table with provenance, then analyse the resulting evidence base through explicit transforms and queries.

The overall aim is to keep the core representation minimal and auditable, while still supporting powerful downstream analysis (filter pipelines, standardisation/recoding, coverage/fit diagnostics), including workflows that use LLMs as low-level assistants for extraction and labelling.

Core papers (start here)#

Pages in this Chapter
Minimalist coding for causal mapping

Intended audience: evaluators / applied qualitative researchers who want a teachable causal coding protocol, and AI/NLP readers who want a simple, auditable target representation of causal content in text.

Combining opposites, sentiment and despite-claims

Instead we take a piece-by-piece approach:

Causal mapping as causal QDA

Unique contribution (what this paper adds):

A simple measure of the goodness of fit of a causal theory to a text corpus

See also: [[000 Working Papers ((working-papers))]]; [[005 Minimalist coding for causal mapping]]; [[900 Magnetisation]].

Magnetisation

Intended audience: people who have done open-ended (often in‑vivo) causal coding and need to standardise factor vocabularies for readable maps/tables without destroying provenance.

Conversational AI – Analysing Central Bank speeches

See also: [[000 Working Papers ((working-papers))]]; [[005 Minimalist coding for causal mapping]]; [[900 Magnetisation]]; [[040 Causal mapping as causal QDA]].