Apply Grounded Theory (Glaser and Strauss) to build theory inductively from qualitative data through open, axial, and selective coding.
SkillScore92Maintained
Automatic
◆Research pack
Last verified 8/28/2026
What it changes
Passed 23 of 25 test cases with the skill, and 16 of 25 without it.
Without skill64%
With skill92%
Checks it passed that the model alone missed
Formats theoretical propositions as a numbered list.
Uses level-3 markdown headings for major output sections.
Includes the level-3 markdown heading `### Core Category`.
and 11 more in the full report
8 fixed15 passed either way1 still failing1 broke
Pass-rate lift
+28 pp
Tokens
+8%
Agent turns
0%
Response time
-25%
Model gemini-3.6-flashMeasured 8/28/2026v123 of 25 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 8 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 8
Prompt
I have gathered transcript excerpts from 15 ICU nurses discussing how they cope with daily workplace trauma. Please analyze this qualitative data using Grounded Theory to construct an explanatory model. I need the output structured to show: 1) the central core category and its operational definition, 2) a table detailing categories along with their key properties, variation dimensions, and link to the core phenomenon, 3) the full coding paradigm outlining causal factors, context, mediating conditions, action strategies, and final outcomes, 4) numbered theoretical statements linking the concepts, and 5) evidence regarding saturation including source counts and the point where new properties stopped appearing.
What was checked, without and with the skill
Checked
Without skill
With skill
Includes the level-3 markdown heading `### Category Structure`.
Missed
Passed
Formats the Coding Paradigm section using bullet points.
Passed
Passed
Includes the point of saturation as a bullet item under Saturation Evidence.
Missed
Passed
case-01 · gemini-3.6-flash · 8/28/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md97 linessingle file
Grounded Theory (Glaser & Strauss)
Overview
Grounded Theory is a systematic methodology for constructing theory that is grounded in qualitative data. Through iterative cycles of data collection, coding, and comparison, the researcher develops concepts and categories that ultimately form an explanatory theory. The method was originally developed by Glaser and Strauss (1967) and later diverged into Glaserian (emergent) and Straussian (structured) variants.
When to Use
Building new theory when existing theories are inadequate or absent
Exploring processes, interactions, or experiences in under-studied domains
Generating substantive theory tied to a specific context
When the research question asks "what is going on here?" rather than testing a hypothesis
When NOT to Use
When testing or verifying an existing theory (use deductive methods)
When the research question requires measurement of frequency or magnitude
When the researcher cannot commit to iterative data collection (theoretical sampling requires flexibility)
Assumptions
IRON LAW: In grounded theory, theory EMERGES from data — imposing a
pre-existing framework violates the methodology's core principle. If
you begin with a hypothesis and seek confirmation, you are NOT doing
grounded theory.
Key assumptions:
Theory must be grounded in systematic data collection and analysis
Data collection and analysis occur simultaneously and iteratively
Theoretical sampling guides where to collect data next based on emerging categories
Theoretical saturation — not sample size — determines when to stop collecting data
Methodology
Step 1: Open Coding
Break data into discrete incidents, events, or ideas. Assign initial codes (labels) to each segment. Use in-vivo codes (participants' own words) where possible. Generate as many codes as the data warrant — do not filter prematurely.