Summary#

This extension is about using factor labels to carve out a useful subgraph of your causal map.

Like most extensions, it is best thought of as:

1. A filter (a rule that takes one links table and returns another), plus
2. An interpretation rule (what it means to say we are “focusing on” or “excluding” factors).

There are two closely related operations:

• Focus: keep the causal neighbourhood around one or more “target” factors (their upstream causes and/or downstream consequences).
• Exclude: remove unwanted factors (and therefore remove any links that touch them).

Unlike label-rewrite transforms (collapse synonyms, remove bracket text, zoom hierarchies, combine opposites), focusing/excluding does not rename factors. It decides which parts of the existing graph you want to see and analyse.

How to think about it (practitioner-friendly)#

Focus = “show me the neighbourhood around this factor”#

You choose one or more target factors (by label search), then choose:

• how far to look upstream (causes), and
• how far to look downstream (consequences).

The result is a sub-map containing only the links that sit on those upstream/downstream chains.

Focusing is a good way to understand a factor as both:

• an outcome (what leads to it?), and
• an influence (what follows from it?),

without having to interpret the entire map at once.

Tip: In interview-style data, chains longer than ~4 steps are uncommon. Large step counts can create hard-to-interpret “hairballs”.

Source tracing = “only keep paths that appear within a single source”#

Sometimes you want coherent within-source narratives rather than a pathway stitched together across respondents.

With source tracing on, the focus result becomes more conservative: it keeps only links that lie on at least one upstream/downstream path that can be realised within a single source.

Exclude = “remove these factors (and anything touching them)”#

Exclude is subtractive: you specify one or more unwanted factor patterns, and the app removes:

• the matching factors, and
• any links that touch them (as cause or effect).

Interpretation cautions#

Order matters#

If you apply label-rewrite transforms earlier (collapse, zoom, remove brackets, combine opposites), then focusing/excluding targets are interpreted in terms of the rewritten labels.

Focus is a reading strategy (not a claim about reality)#

Focusing is a way to make a large map interpretable. It does not claim “only this neighbourhood is relevant”; it claims “within N steps, what do sources connect to this factor?”

Relationship to “collapse” (different goal)#

• Use collapse/label-rewrite when you want to treat several labels as the same concept while keeping the surrounding structure visible.
• Use focus when you want to keep the original labels but restrict attention to the local causal neighbourhood of a concept.

Formal notes (optional)#

If you want the precise (link-based) rule, here is the intended definition.

Let \(F\) be the set of focused factor labels, and let \(U\) and \(D\) be the upstream/downstream step limits.

• Keep a link \(x \rightarrow y\) if it lies on any directed path of length \(\le U\) that ends at a factor in \(F\), or any directed path of length \(\le D\) that starts at a factor in \(F\).
• Do not add extra “cross-links” between surviving factors; keep only links that are actually part of the selected paths.

For exclude, let \(E\) be the excluded factor set; remove all links \(x \rightarrow y\) where \(x \in E\) or \(y \in E\).

Examples (contrasts) from the app#

A single-theme focus (one-step neighbourhood)#

Bookmark #982 is a simple example of focusing on one theme and looking at its immediate neighbourhood.

Upstream focus with a single-source constraint (“source tracing”)#

These two bookmarks are both “upstream influences on wellbeing” views, but one requires within-source narrative coherence:

• Without source tracing: bookmark #270

• With source tracing: bookmark #534. Notice how it is more conservative.