PDF ⛶
🌻 Recoding labels temporarily
28 Apr 2026

Recoding labels temporarily#

Sometimes you want to improve your factor labels (cause and effect text) without changing the original data. You might want to:

The app supports this with the Label Set widget in the toolbar below the Sources bar. Treat label sets like different versions of the same document: one default version, plus as many named experimental versions as you want.

How to use it#

  1. Open the Label Set widget under the Sources bar. Type a new name for your experimental set, for example v1 or cleanup, and confirm. The app fills a new pair of columns (cause_v1, effect_v1) for every link from what you were just looking at, then switches you to that set.

  2. Edit or recode within the experimental set. From now on, anything you do (manual edits in the Factors panel or Links table, AI Recode, Answers, search/replace, Bulk Edit) writes to the experimental columns. The default labels stay untouched.

  3. Flip between sets with the same widget. Pick default to see the original labels; pick your experimental set to see the experimental labels. The map and tables refresh to match. Useful for quick before-and-after comparisons.

  4. When you're happy with one experimental set, open the ⋯ menu in the widget (only visible when a non-default set is selected) and choose Promote to default. That copies the current experimental set into the default columns. Your other named versions stay in the project; nothing is lost.

Summary#

Use the Label Set widget to spin up named experimental versions of your labels, work on them without touching the default, switch between versions with one click, and promote the winner to default when you're done.

See also#