Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Research Design & Methods (amj-methods) · brycewang-stanford | DecimalAI

researchby brycewang-stanford

Research Design & Methods (amj-methods)

brycewang-stanford/amj-methods

Source brycewang-stanford/awesome-journal-skills ↗License MIT

Use when the research design and method are the bottleneck for an Academy of Management Journal (AMJ) manuscript — matching design (archival, survey, experiment, multi-method, field) and level of analysis to the theoretical question. Designs the study; it does not run the estimation or validity checks (amj-data-analysis).

SkillScore100Maintained

Automatic

Research pack

Last verified 8/15/2026

What it changes

Passed 23 of 23 test cases with the skill, and 21 of 23 without it.

Without skill91%

With skill100%

Checks it passed that the model alone missed

Recommends pre-testing the new scale and performing confirmatory factor analysis to demonstrate construct validity
Recommends clarifying how qualitative interviews uncover underlying organizational mechanisms while quantitative patent data tests generalizability

2 fixed21 passed either way0 still failing0 broke

Pass-rate lift: +9 pp
Tokens: +29%
Agent turns: 0%
Response time: -18%

Model gemini-3.6-flashMeasured 8/15/2026v123 of 23 cases passed with skill

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/brycewang-stanford-amj-methods/SKILL.md https://app.decimal.ai/s/brycewang-stanford-amj-methods/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add brycewang-stanford-amj-methods, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/brycewang-stanford-amj-methods/ and pins it in .decimal/skills.lock

decimalai skills add brycewang-stanford-amj-methods

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/brycewang-stanford-amj-methods/SKILL.md https://app.decimal.ai/s/brycewang-stanford-amj-methods/SKILL.md

Raw file: /s/brycewang-stanford-amj-methods/SKILL.md ↗ · machine-readable summary at /s/brycewang-stanford-amj-methods.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 2 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 2

Prompt

To measure 'algorithmic anxiety', we created 4 custom items on the fly and calculated their mean score without pre-testing or running factor analysis. Is this sufficient for testing our hypotheses?

What was checked, without and with the skill
Checked	Without skill	With skill
Recommends pre-testing the new scale and performing confirmatory factor analysis to demonstrate construct validity	Missed	Passed

case-09 · gemini-3.6-flash · 8/15/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md82 linessingle file

Research Design & Methods (amj-methods)

When to trigger

The design may not match the theory's level, timing, or causal claim
Data are single-source, single-wave, and self-reported (common-method bias risk)
The theory is causal but the design is cross-sectional/correlational
Constructs lack established, validated measures
A reviewer says "the design cannot test this hypothesis" or "endogeneity is unaddressed"

Match the design to the question

AMJ explicitly welcomes all empirical methods — qualitative, quantitative, field, laboratory, meta-analytic, and mixed. The bar is fit and rigor, not a single preferred method, and qualitative designs are held to an equally demanding standard (the Eisenhardt multiple-case approach and the Gioia methodology for grounded qualitative rigor are the field's reference points).

Theoretical claim	Design that earns it
Causal effect of a manipulable cause	Experiment (lab/field/online), or natural experiment
Process unfolding over time	Multi-wave panel; longitudinal/lagged design
Firm/strategy outcomes from archival cause	Panel archival with fixed effects + endogeneity strategy
Cross-level mechanism (e.g., team→indiv.)	Multilevel/nested data with HLM-appropriate structure
Rich, novel, or contested phenomenon	Qualitative or multi-method (often paired with a study 2)

A two-study design (e.g., field study for generalizability + experiment for causal mechanism) is a common AMJ strength — it answers both internal and external validity.

Designing against the threats AMJ cares about

Common-method bias (CMB): separate sources for predictor and outcome; temporal separation across waves; objective/archival outcomes where possible. Procedural remedies beat statistical fixes (the Podsakoff et al. guidance is the standard reference). Plan this before collecting data.
Endogeneity (archival): anticipate omitted variables, reverse causality, and selection. Plan an identification strategy (instrument, natural experiment, panel fixed effects, difference-in-differences, Heckman/2SLS, propensity matching) and the assumptions each requires.
Measurement: use validated multi-item scales; pilot new measures; plan a CFA. State the level at which each construct is measured and how cross-level data are aggregated (with justification: ICC, r_wg, aggregation theory).
Sampling and power: justify the sampling frame, response rate, and statistical power for the focal and interaction effects (interactions need more power).

Level-of-analysis discipline

State the level for theory, measurement, and analysis, and keep them aligned. If theory is at the team level but data are individual, justify aggregation; if effects are cross-level, the analysis must model the nesting (do not run OLS on nested data).

Execution bridge (StatsPAI / Stata MCP)

For the empirical / causal lane, estimate and audit rather than only specify. Full map: execution-with-mcp. AMJ is empirical management — panel, multilevel, DiD, IV, and field/lab experiments; the chain below serves that lane, while grounded-theory / qualitative work uses its own standards.

Track record

Benchmark

23 / 23

test cases passed, vs 21 / 23 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

brycewang-stanford / awesome-journal-skills

Academy-of-Management-Journal-Skills/skills/amj-methods/SKILL.md

View on GitHub ↗

License MITImported 8/3/2026Version v1

Other skills for the same job

Published skills whose descriptions closely match this one, with their headline benchmark lift.

brycewang-stanford/amj-data-analysisUse when running and reporting the statistical analysis for an Academy of Management Journal (AMJ) manuscript — measurement validity, common-method bias, the right estimator (HLM, SEM, panel, experiments), endogeneity, and robustness. Executes and reports the analysis; it does not design the study (amj-methods) or frame the contribution (amj-contribution-framing).Close match+14 pp pass-rate lift

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

Research Design & Methods (amj-methods)

When to trigger

Match the design to the question

Designing against the threats AMJ cares about

Level-of-analysis discipline

Execution bridge (StatsPAI / Stata MCP)

Track record

Benchmark

Adoption

SkillSafety

Source

Other skills for the same job

Related skills