| name | jcr-methods |
| description | Use when choosing or stress-testing the research design for a Journal of Consumer Research (JCR) manuscript — multi-study behavioral experiments, interpretive Consumer Culture Theory (CCT) fieldwork, mixed designs, or a Registered Report — so the evidence matches the conceptual claim. Designs the studies; it does not analyze them (jcr-data-analysis). |
Methods & Design (jcr-methods)
When to trigger
- You have a mechanism but are unsure how to test it
- Deciding between a behavioral-experiments paper and a CCT fieldwork paper
- A reviewer asks whether your design can actually support the process claim
- You are weighing a Registered Report for a confirmatory question
JCR is methodologically pluralistic by mandate
JCR states no single preferred method; the bar is a clear conceptual contribution supported by appropriate empirical evidence. In practice two flagship traditions coexist under one masthead, and you should commit to one design logic (or a principled mix):
- Theory-driven behavioral experimentation (the dominant tradition): multiple lab and online experiments that isolate a psychological process and its boundary conditions.
- Interpretive / Consumer Culture Theory (CCT): ethnography, depth interviews, phenomenology, or netnography that theorizes the sociocultural meanings of consumption.
The journal also publishes quantitative/modeling and methodological work. Choose the design the conceptual claim demands, not the one you find convenient.
Designing the multi-study experimental package
- Process evidence: plan studies that establish the effect, then mediation (measured or, more convincingly, moderation-of-process / manipulated mediator), then boundary conditions that the theory predicts.
- Internal validity: random assignment; manipulation checks and attention checks; pretested stimuli; counterbalancing; rule out demand and confounds by design.
- Robustness across studies: vary populations, stimuli, and operationalizations so the effect is not stimulus-bound; a convergent multi-study package is the JCR norm.
- Power & samples: a priori power analysis; specify and justify sample sizes and exclusion rules in advance. Overflow stimuli, full instruments, and additional replication studies belong in the web appendix (max 40 MB, excluded from the 60-page cap).
Designing interpretive / CCT work
- Justify site, informant selection, and immersion; show the data are rich enough to support conceptual claims.
- Plan for trustworthiness: triangulation, prolonged engagement, member checks, and an audit trail rather than p-values.
- Theorize as you go: the design should enable moving from thick description to second-order constructs.
Transparency is a design decision, not an afterthought
JCR's transparency regime shapes the design from the start: a Data Collection Statement is required for all submissions (Step 6), data/materials posting is required at invited revision unless exempt, and replication code must be provided. Build clean materials, preregistration where appropriate, and a repository plan (OSF / Harvard Dataverse / Qualitative Data Repository / ResearchBox) into the design. For confirmatory questions, consider a Registered Report (full review before final data collection; must be JCR-worthy regardless of outcome).
Execution bridge (StatsPAI / Stata MCP)
For the empirical / causal lane, estimate and audit rather than only specify. Full
map: execution-with-mcp. JCR is predominantly lab experiments; randomization-based inference and the many-outcome family-wise correction (romano_wolf) are the decisive tools.
detect_design → recommend → fit with as_handle=true → audit_result to
enumerate the checks the design owes.- Panel / staggered DiD:
callaway_santanna / sun_abraham + bacon_decomposition
honest_did_from_result. IV: effective_f_test + anderson_rubin_ci. RDD:
rdrobust + mccrary_test.
- Experiments: randomization-based inference and
romano_wolf for the many-outcome
family-wise correction reviewers expect.
Match the toolchain to the reviewer pool, and report the effect size the venue
wants. A run end-to-end (synthetic data, real returns) is in the
JF execution walkthrough.
Checklist
Anti-patterns
- A single study asked to carry a process claim.
- "Mediation" inferred from a measured mediator without manipulating the process.
- Stimulus-bound effects (one scenario, one product) generalized broadly.
- CCT design with too little immersion to support conceptual claims.
- Treating data/materials posting as a post-acceptance chore.
Output format
【Design logic】experiments / CCT / mixed / Registered Report
【Study chain】effect → process → boundary (or CCT framework)
【Validity safeguards】randomization / checks / pretests / trustworthiness
【Power & samples】a priori N, exclusions
【Transparency plan】repository + code + Data Collection Statement
【Web appendix】overflow stimuli / extra studies (≤40 MB)
【Next step】jcr-data-analysis
