(Remnant et al., 2025)

Source: book chapter draft in content/000 Articles/020 !! health book chapter.md.

• Purpose

• Position QuIP + causal mapping as a credible, cost-effective way to elicit and analyse perceived drivers/barriers in complex interventions (including health services evaluations).

• Data collection stance

• QuIP focuses on changes that matter to respondents, and the perceived causes of those changes.
• Goal-free / blindfolded questioning is used to reduce pro-project bias; unprompted mention is treated as important evidence.

• Not designed to estimate effect sizes; complements (rather than replaces) quantitative inference and other theory-based approaches.

• Coding stance (“natively causal”)

• Coding is not thematic tags that are linked later; coding is pairs/chains of cause→effect factors (“causal nuggets”).
• Coding is parsimonious: only causal claims are coded; non-causal descriptive text is not.

• Inductive label harmonisation across sources is expected; analyst should manage positionality and avoid over-fitting to prior ToC.

• Use

• Compare empirical causal maps against ToCs; compare groups (e.g. men/women; staff cadres) and pathways.

• Keep a traceable link from visual summaries back to underlying quotes for verification/peer review.

• Relationship to realist ideas

• Affinity to mechanism/context thinking (multiple pathways), but with broader open capture rather than only a few “hotspots”.