IS-09 Quantitising and qualitising (data transformation)
Converts one data type into the other, for example by counting coded themes, and produces a single-type dataset of frequencies or profiles.
What it does
Converts one data type into the other, for example by counting coded themes, so that the two can be compared directly. Artefact: a single-type dataset. Read the frequencies or profiles produced.
When to use it
Use when you need to compare coded text with numeric data directly, or to turn numeric patterns into profiles that can be read.
When it does not fit
Fits when coding is reliable. Misuse: reading theme counts from a non-random sample as prevalence.
Result format
The result is a single-type dataset in which the coded themes have been converted into counts, with one row per case or group and one column per theme. It reports the frequencies or profiles produced and states whether the sample supports reading the counts as prevalence.
Requirements on the material
| Property | Requirement type | Requirement |
|---|---|---|
| Content type | Always applies | requires=text; requires=tabular |
| Unit of analysis | Always applies | requires=document; requires=record |
| Number of sources | Always applies | requires=two or more |
| Appraises source credibility | Applies in some cases | no |
Not restricted on: Structure, Variable types, Time structure, How the sample was chosen, Level of detail available, Whether sources can be joined, Amount of material, Known data problems, Producer interest, Source dependence, Origin, Document genre, Language, Personal information, Security classification, Licence.
| Minimum material | none |
|---|---|
| Additional sources needed | none |
| Not for (rules) | none |
| Not for (text) | Coding is unreliable, or theme counts from a non-random sample are read as prevalence. |
Choosing and ordering
| Kind of technique | Primary technique, runs in step 5.3 |
|---|---|
| Phase tag | analyse |
| Other phase tags | none |
| Kind of finding | pattern |
| Question answered | what; how much |
| Strongest statement it supports | describe |
| Needs as input |
|
| Makes as output |
|
| Must exist first | None |
| Loop target | proceed with limit stated |
Library checks by stakes
Each stakes level includes the checks of the level below it. Checks that need a person apply at high stakes only, so the process leaves them out at lower stakes (decision 38). Repeatability: deterministic.
| Stakes | Check | Person involved | Counts as a different method |
|---|---|---|---|
| exploratory | input validation | none | no |
| decision-informing | input validation | none | no |
| decision-informing | spot check | optional | no |
| high-stakes | input validation | none | no |
| high-stakes | spot check | optional | no |
| high-stakes | independent review | needed | no |
| high-stakes | provenance log | none | no |
Failure modes and limits
| Model suitability | model plus code |
|---|---|
| Typical model failure | Counts themes in prose without code, and presents counts from a purposive sample as population shares. |
| Error sources | input data; coder or rater judgement; data transformation |
| Certainty limits | Certainty drops with unreliable coding and with samples that do not represent the population (indirectness and risk of bias). |
Related techniques
- should follow CK-01 Inter-coder agreement statistics (percentage agreement; Cohen's kappa; weighted kappa; Scott's pi; Fleiss' kappa; Krippendorff's alpha; intercoder reliability testing): coding must be shown to be reliable before it is counted
Evidence and tag basis
Unconfirmed. The source names the approach under interpretation and reporting integration and does not describe it, so this description is the note author's. (Fetters, Curry and Creswell 2013, via core.ac.uk)
Tag basis: input profile inference; selection inference; sequencing inference; robustness inference. All tag values are inference unless a source is named.