Shape and recode without changing the source

Make the dataset usable without losing its history.

Merge split fields, relabel columns, build nets and derived variables, and refresh recurring waves in the same project. Recode response options, combine multi-select columns, create calculated variables, and apply numeric factors as non-destructive layers on the source.

The original source remains there to inspect.

Review what changed, revert individual transformations, and keep the original import available for audit. Every merge and recode remains traceable when a client, reviewer, or compliance team asks how a derived variable was built.

Keep the logic when the next wave arrives.

Wave refreshes append new data without rebuilding derived logic. The shaped dataset can move directly into tables, statistical analysis, and delivery—without another export or working copy.

Give every change a reason and a place.

A research project rarely arrives in exactly the form the next question requires. A combined category, a recoded scale, or a derived score can be useful for one analysis and inappropriate for another. Keeping those changes as visible project layers lets the team use the form it needs while retaining the source needed to explain, revisit, or revise the decision later.