Definitions
State what a code means in this project instead of relying on a memorable label.
macOS · Qualitative analysis · Work in progress
QualText
Qualitative analysis, close to the text.
QualText begins with a simple objection: qualitative analysis is not the same thing as asking an AI to summarize a folder of interviews. The work lies in reading closely, deciding what a passage means, developing and revising codes, writing memos, comparing cases, looking for contradictions, and constructing claims that can be traced back to evidence.
Established qualitative analysis packages can support this work, but they have accumulated breadth, complexity, cloud services, and cost that can be difficult to justify for someone undertaking only a few serious projects each year. At the other end of the market, lightweight AI tools promise immediate themes while making the interpretive path harder to inspect.
QualText is intended to occupy the space between them: a focused, local-first workbench built around text and a living codebook, with source context, memos, comparison, negative cases, claims, and an audit trail kept in view.
The application must work fully without AI. Optional assistance may eventually help locate candidate passages or compare code definitions, but suggestions are not findings. The researcher’s decisions remain the substantive analytical events.
The analytical spine
The source is the ground truth. The codebook is the developing analytical model.
In QualText, the codebook is not simply a list of labels added to passages. It is where concepts are defined, distinguished, tested, revised, and carried forward into memos, comparisons, and eventual claims. Every analytical conclusion should remain traceable through that structure and back to the original material.
Boundaries
Record inclusion and exclusion rules so similar concepts can be distinguished.
Evidence
Keep examples and counterexamples close enough to test the developing interpretation.
Revision
Preserve rationale, version history, and unresolved edge cases as understanding changes.
Connections
Link codes to memos, comparisons, negative cases, claims, and supporting evidence.
The codebook does not need to arrive finished. It may begin with provisional language, then grow sharper through use, disagreement, revision, and abandoned interpretations.
The working path
Coding is the beginning, not the result.
QualText should support the whole interpretive path without hiding how the researcher moved through it. At every stage, the work should be able to return to source context.
-
01
Source and provenance
Read the material in context.
-
02
Living codebook
Develop and test concepts.
-
03
Memos and comparison
Work out what the patterns mean.
-
04
Claims and evidence
Make the reasoning defensible.
A deliberate boundary
Focused on interpretation, not ingestion.
Transcription, OCR, translation, survey collection, and media processing are already specialist tasks. QualText should accept their outputs cleanly, preserve where the material came from, and begin when the text is ready to be interpreted.
It is not intended to become another all-purpose CAQDAS suite, research repository, presentation dashboard, or one-click theme generator. Its job is narrower: help a researcher turn analyzable text into a defensible interpretive structure.
Current status
Defining the method before drawing the interface.
QualText is presently in research and product definition. I am specifying the method, product boundary, project structure, and core workflow before building an interface. There is no download, release date, or final pricing yet. This page will follow the project as those decisions become a working macOS application.