# Education Variable Identification

> Use when identifying independent variables, dependent variables, mediators, moderators, controls, constructs, indicators, instruments, and operational definitions from education research topics, papers, PDFs, Zotero libraries, survey plans, or literature review notes. Supports Elicit-style extraction, PDF RAG, and structured variable tables.

- Skill: `ilog3/education-variable-identification` (Agent Skill)
- Install (CLI): `npx skillmds@latest add ilog3/education-variable-identification`
- Raw SKILL.md: https://api.skillmd.com/api/skills/ilog3/education-variable-identification/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: ilog3 (https://skillmd.com/u/ilog3)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/ilog3/education-variable-identification

---


# Education Variable Identification

## Goal

Transform a topic, research question, or literature set into a defensible variable/construct model for education research.

## Inputs

- Research question or hypothesis
- Topic and population
- PDFs, abstracts, Zotero export, or paper list if available
- Intended method: survey, experiment, quasi-experiment, learning analytics, classroom observation, qualitative coding, mixed methods

## Workflow

1. Parse the research question into constructs.
2. Classify each construct:
   - independent variable
   - dependent variable
   - mediator
   - moderator
   - control variable
   - grouping variable
   - qualitative theme/phenomenon
3. Search literature for how constructs are measured.
4. Extract instruments, dimensions, sample items, and reported reliability if available.
5. Build a conceptual model.
6. Produce operational definitions and data collection suggestions.

## Tool Calls

### Zotero Export

Ask the user to export selected Zotero items as BibTeX/RIS/CSV, or use Zotero local data if available.

Zotero download:

```text
https://www.zotero.org/download/
```

Suggested Zotero workflow:

1. Create collection for the research topic.
2. Export collection as BibTeX or CSV.
3. Use abstracts/PDFs to extract variables and instruments.

### PDF Text Extraction

Use local tools if PDFs are provided:

```bash
pdftotext input.pdf output.txt
```

or:

```bash
python -m pymupdf gettext input.pdf
```

Then search the text:

```bash
rg -n "variable|measure|instrument|scale|reliability|Cronbach|mediator|moderator|control" output.txt
```

### Semantic Scholar / OpenAlex

Search for measurement studies:

```bash
curl "https://api.semanticscholar.org/graph/v1/paper/search?query=academic%20self-efficacy%20scale%20students%20education&limit=10&fields=title,year,abstract,citationCount,authors"
```

```bash
curl "https://api.openalex.org/works?search=academic%20self-efficacy%20scale%20education&per-page=10&mailto=YOUR_EMAIL"
```

### Elicit-Style Extraction Fields

When using Elicit or an LLM extraction table, use these columns:

```text
study design | sample | independent variables | dependent variables | mediators | moderators | controls | instruments | reliability | key findings | limitations
```

## Output Format

Return:

| Construct | Variable Role | Operational Definition | Measurement / Instrument | Data Source | Literature Support | Notes |
|---|---|---|---|---|---|---|

Then include:

- Conceptual model in text or Mermaid
- Hypotheses aligned to variables
- Measurement risks
- Missing variables to consider

## Mermaid Model Template

```mermaid
flowchart LR
  IV["Independent Variable"] --> MED["Mediator"]
  MED --> DV["Dependent Variable"]
  MOD["Moderator"] -. moderates .-> IV
  CTRL["Controls"] --> DV
```

## Quality Rules

- Do not invent a validated scale. If unknown, say "measurement needs confirmation".
- Separate theoretical constructs from observable indicators.
- For education research, always consider controls such as grade, gender, prior achievement, SES, teacher experience, school type, baseline score, and technology access where relevant.
- If qualitative, output themes/phenomena instead of forcing variables.