Qualitative Coding

Qualitative coding is the transformation of open-ended responses to a survey into quantitative measurements.


Theory

Approaches to Coding

There are fundamentally two approaches to coding:

In practice, a mix of the two is most common. The latter becomes more difficult with larger coding teams.

Qualitative data analysis (QDA) software (e.g. NVivo, ATLAS.ti, Dedoose, MAXQDA, QDAMiner, Quickos) helps with the coding process.

Approaches to Writing Codebooks

A flat codebook contains discrete, separate codes. Code 1, code 2, code 3, and so on.

A hierarchical codebook contains codes that overlap or that represent more or less granularity. Code 1, code 1.1, code 1.2, code 2, and so on. A code of 1.1 can, in some circumstances, be treated as a code of 1. Such a codebook is generally less usable and can lead to inconsistency between coders.

Note that using a flat codebook does not imply exclusivity of codes. Qualitative coding is inherently about measuring relationships and trends, so every meaningful data point is connected in multiple ways. If a response is only notable in one dimension, it likely isn't meaningful.


Coding Styles

Primary/Secondary coding means that code-able sentiments are recorded in order of their appearance. If a response includes two sentences that correspond to a code A and a code B respectively, then the primary code is A and the secondary code is B. The naming of these codes does not imply weight or importance.

All Categories coding means that every applicable code is marked. This creates data that is comparable to a multi-select question.


CategoryRicottone

Statistics/QualitativeCoding (last edited 2025-01-10 15:59:56 by DominicRicottone)