This extension is about using factor labels to unify many “different-looking” factors into one.

• Collapse factor labels: define one or more search terms, and any factor label that matches is rewritten to a single shared label (a “bucket”). This is a quick way to treat near-synonyms and variants as one concept (e.g. money / income / salary).
• Exclude text in brackets is really the same idea: it’s just a fixed rule for rewriting labels. For example Education (primary school) becomes Education, so all bracketed variants collapse onto the same base label.

Like the other transform filters, this is a temporary rewrite used for analysis and presentation; it doesn’t change your underlying coding.

In the current app, these text matches are case-insensitive (case doesn’t matter).

All the corresponding factors are collapsed into one factor which is now labelled with the search term.

Multiple search terms, combining results#

You can put more than one search term.

Multiple search terms with separate results#

Or if you want to keep the results separate. The factors which match the filter are shown with a thicker purple border.

Multiple search terms with OR and separate results#

The previous strategy won’t work if you want two separate buckets, like (Food OR Diet) on the one hand and (Income OR Money) on the other.

In the current app, the clean way to do this is to define separate buckets (so each match collapses to the term it matched), rather than relying on case-sensitivity tricks.

Because matching is case-insensitive in the current app, you don’t need separate terms just to capture Health vs health.

Excluding brackets (same idea, fixed rule)#

If you often code specific labels like Education (primary school) and Education (secondary school), excluding bracket text is just a convenient way to collapse them both to Education without having to maintain a list of search terms.