Methods & Research Design (jbv-methods)
When to trigger
- You are choosing a design and unsure which fits a venture-creation question
- The design may not let you observe the entrepreneurial process you theorize
- A reviewer flags selection into entrepreneurship, survivorship bias, or sample novelty
- You are weighing qualitative process work, experiments, archival venture panels, or mixed methods
JBV is methodologically pluralistic — but theory-first
JBV prizes a clear, substantive theoretical contribution to entrepreneurship over any single method. Well-executed qualitative, conceptual, quantitative, and mixed-method studies are equally welcomed; the unifying demand is that the work advance theory about the entrepreneurial phenomenon, not merely apply a method. Choose the design that can actually test or build your theory:
| Entrepreneurial question / claim |
Fitting design |
| Process of how ventures emerge / sensemaking |
Inductive qualitative (Gioia, process), longitudinal case studies |
| Entrepreneurial judgment / cognition under uncertainty |
Experiments, conjoint/policy-capturing, vignettes (Prolific/lab) |
| Antecedents/consequences across many ventures |
Archival venture panels (Crunchbase, PitchBook, GEM, KFS, PSED) |
| Founding choice, exit, IPO, failure |
Survival/event-history; selection models |
| Financing signals (VC, crowdfunding) |
Field/natural experiments, panel with funding events |
| Theory-building plus generalization |
Mixed methods (qual to build, quant to test) |
Design issues specific to the entrepreneurial setting
- Selection into entrepreneurship: founders self-select; model the choice (Heckman/Roy) or use design-based identification rather than ignoring it.
- Survivorship: new-venture samples lose failed ventures fast; a sample of survivors silently conditions on success. Design to observe pre-founding and failed ventures (PSED-style nascent panels, registry data).
- Novel datasets: JBV values new entrepreneurship data; document construction, coverage, and the sampling frame transparently — novelty is an asset only if the frame is defensible.
- Temporal precedence: to claim antecedents/mechanisms, the design must order them in time (longitudinal, pre/post a shock, staged measurement).
Co-submission of method artifacts
The Editorial Manager workflow lets you attach a MethodsX article (detailed protocol) or Data in Brief descriptor on the "Attach files" page — useful for novel measures, hand-coded datasets, or experimental protocols.
Execution bridge (StatsPAI / Stata MCP)
For the empirical / causal lane, estimate and audit rather than only specify. Full
map: execution-with-mcp. JBV studies founders and ventures where selection / survivorship threatens every claim; lead with identification and selection-correction tooling.
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
Design referee-pushback patterns and the JBV-specific fix
| Pushback at the design stage |
JBV-specific fix |
| "Cross-sectional, yet you theorize a venturing process." |
Re-design with staged measurement or a pre/post shock; a single wave cannot carry a process claim. |
| "The construct is general-management ability relabeled." |
Anchor the manipulation/measure to a venture primitive and show discriminant validity from it. |
| "Your sample is whoever survived to be observed." |
Push the frame upstream to nascent/registry data capturing pre-founding and failed ventures. |
| "A novel hand-coded dataset — why trust the frame?" |
Document construction, inter-coder reliability, and coverage; consider a MethodsX/Data in Brief co-submission. |
Worked micro-example (illustrative)
A hypothetical JBV study claims perceived environmental uncertainty causes founders to evaluate ambiguous opportunities more favorably. Design reasoning:
- Why an experiment: the claim is causal and about entrepreneurial judgment, so a vignette experiment (240 founders, illustrative) manipulates uncertainty while holding the opportunity fixed — isolating the targeted cognition.
- Construct guard: the vignette describes a new-venture opportunity under Knightian uncertainty, not a generic managerial decision, so a referee cannot recast it as ordinary risk-taking.
- Triangulation: pair it with an archival panel where a plausibly exogenous uncertainty shock shifts real founding/financing — answering "lab artificiality" by design.
- Temporal precedence: in the archival arm, measure uncertainty before the founding outcome so the antecedent claim is ordered in time.
Calibration anchors (hedged)
- JBV is method-pluralist: a deep inductive study and a clean field experiment clear the same bar if both advance entrepreneurship theory. The dividing line is theory payoff, not method prestige.
- The desk-reject-adjacent failure is the survivor-only, single-wave, generic-construct combination; design against all three up front.
- Artifact options (MethodsX, Data in Brief) can change; confirm against the journal's current author guidelines.
Anti-patterns
- Method-led, theory-light — a clean identification with no entrepreneurship-theory payoff.
- Survivor-only sample treated as representative of venture creation.
- Cross-sectional self-report used to claim a dynamic entrepreneurial process.
- Convenience startup sample with an undocumented frame.
Output format
【Question→design fit】design chosen + why it fits the entrepreneurial claim ...
【Selection】founding-choice strategy ...
【Survivorship/attrition】sampling-frame handling ...
【Data novelty】construction + coverage + frame ...
【Temporal precedence】how ordered in time ...
【Artifacts】MethodsX / Data in Brief? ...
【Next step】jbv-data-analysis
Source: brycewang-stanford/Awesome-Journal-Skills → Journal-of-Business-Venturing-Skills/skills/jbv-methods/SKILL.md
1---2name: jbv-methods3description: Use when designing or defending the research design for a Journal of Business Venturing (JBV) manuscript — matching a methodologically pluralistic but theory-first approach (venture panels, experiments, qualitative process work, mixed methods) to an entrepreneurial question, and handling selection, survival, and novel-dataset issues. Designs the study; it does not estimate it (jbv-data-analysis) or frame the contribution (jbv-contribution-framing).4---5
6
7# Methods & Research Design (jbv-methods)
8
9## When to trigger
10
11- You are choosing a design and unsure which fits a venture-creation question
12- The design may not let you observe the entrepreneurial process you theorize
13- A reviewer flags selection into entrepreneurship, survivorship bias, or sample novelty
14- You are weighing qualitative process work, experiments, archival venture panels, or mixed methods
15
16## JBV is methodologically pluralistic — but theory-first
17
18JBV prizes a clear, substantive **theoretical contribution to entrepreneurship** over any single method. Well-executed qualitative, conceptual, quantitative, and mixed-method studies are equally welcomed; the unifying demand is that the work advance theory about the entrepreneurial phenomenon, not merely apply a method. Choose the design that can actually test or build your theory:
19
20| Entrepreneurial question / claim | Fitting design |
21|----------------------------------------------------------|-------------------------------------------------------------------|
22| Process of how ventures emerge / sensemaking | Inductive qualitative (Gioia, process), longitudinal case studies |
23| Entrepreneurial judgment / cognition under uncertainty | Experiments, conjoint/policy-capturing, vignettes (Prolific/lab) |
24| Antecedents/consequences across many ventures | Archival venture panels (Crunchbase, PitchBook, GEM, KFS, PSED) |
25| Founding choice, exit, IPO, failure | Survival/event-history; selection models |
26| Financing signals (VC, crowdfunding) | Field/natural experiments, panel with funding events |
27| Theory-building plus generalization | Mixed methods (qual to build, quant to test) |
28
29## Design issues specific to the entrepreneurial setting
30
31- **Selection into entrepreneurship**: founders self-select; model the choice (Heckman/Roy) or use design-based identification rather than ignoring it.
32- **Survivorship**: new-venture samples lose failed ventures fast; a sample of survivors silently conditions on success. Design to observe pre-founding and failed ventures (PSED-style nascent panels, registry data).
33- **Novel datasets**: JBV values new entrepreneurship data; document construction, coverage, and the sampling frame transparently — novelty is an asset only if the frame is defensible.
34- **Temporal precedence**: to claim antecedents/mechanisms, the design must order them in time (longitudinal, pre/post a shock, staged measurement).
35
36## Co-submission of method artifacts
37
38The Editorial Manager workflow lets you attach a **MethodsX** article (detailed protocol) or **Data in Brief** descriptor on the "Attach files" page — useful for novel measures, hand-coded datasets, or experimental protocols.
39
40## Execution bridge (StatsPAI / Stata MCP)
41
42For the **empirical / causal lane**, estimate and audit rather than only specify. Full
43map: [`execution-with-mcp`](../../../shared-resources/empirical-methods/execution-with-mcp.md). JBV studies founders and ventures where selection / survivorship threatens every claim; lead with identification and selection-correction tooling.
44
45- `detect_design` → `recommend` → fit with `as_handle=true` → `audit_result` to
46 enumerate the checks the design owes.
47- **Panel / staggered DiD:** `callaway_santanna` / `sun_abraham` + `bacon_decomposition`
48 + `honest_did_from_result`. **IV:** `effective_f_test` + `anderson_rubin_ci`. **RDD:**
49 `rdrobust` + `mccrary_test`.
50- **Experiments:** randomization-based inference and `romano_wolf` for the many-outcome
51 family-wise correction reviewers expect.
52
53Match the toolchain to the **reviewer pool**, and report the effect size the venue
54wants. A run end-to-end (synthetic data, real returns) is in the
55[JF execution walkthrough](../../../Journal-of-Finance-Skills/resources/worked-examples/02-execution-walkthrough.md).
56## Checklist
57
58- [ ] Design can actually observe/test the theorized entrepreneurial mechanism
59- [ ] Selection into founding addressed (modeled or design-based)
60- [ ] Survivorship/attrition handled in the sampling frame, not just statistically later
61- [ ] Novel dataset's construction, coverage, and frame documented
62- [ ] Temporal precedence supports the antecedent/mechanism claim
63- [ ] Qual studies: trustworthiness plan (data structure, audit trail, quotations)
64- [ ] Considered MethodsX / Data in Brief co-submission for protocols/data
65
66## Design referee-pushback patterns and the JBV-specific fix
67
68| Pushback at the design stage | JBV-specific fix |
69|-----------------------------------------------------------|-------------------------------------------------------------------------------------|
70| "Cross-sectional, yet you theorize a venturing *process*."| Re-design with staged measurement or a pre/post shock; a single wave cannot carry a process claim.|
71| "The construct is general-management ability relabeled." | Anchor the manipulation/measure to a venture primitive and show discriminant validity from it.|
72| "Your sample is whoever survived to be observed." | Push the frame upstream to nascent/registry data capturing pre-founding and failed ventures.|
73| "A novel hand-coded dataset — why trust the frame?" | Document construction, inter-coder reliability, and coverage; consider a MethodsX/Data in Brief co-submission.|
74
75## Worked micro-example (illustrative)
76
77A hypothetical JBV study claims *perceived environmental uncertainty* causes founders to evaluate ambiguous opportunities more favorably. Design reasoning:
78
79- **Why an experiment**: the claim is causal and about entrepreneurial *judgment*, so a vignette experiment (240 founders, illustrative) manipulates uncertainty while holding the opportunity fixed — isolating the targeted cognition.
80- **Construct guard**: the vignette describes a *new-venture* opportunity under Knightian uncertainty, not a generic managerial decision, so a referee cannot recast it as ordinary risk-taking.
81- **Triangulation**: pair it with an archival panel where a plausibly exogenous uncertainty shock shifts real founding/financing — answering "lab artificiality" by design.
82- **Temporal precedence**: in the archival arm, measure uncertainty *before* the founding outcome so the antecedent claim is ordered in time.
83
84## Calibration anchors (hedged)
85
86- JBV is method-pluralist: a deep inductive study and a clean field experiment clear the *same* bar if both advance entrepreneurship theory. The dividing line is **theory payoff**, not method prestige.
87- The desk-reject-adjacent failure is the **survivor-only, single-wave, generic-construct** combination; design against all three up front.
88- Artifact options (MethodsX, Data in Brief) can change; **confirm against the journal's current author guidelines**.
89
90## Anti-patterns
91
92- **Method-led, theory-light** — a clean identification with no entrepreneurship-theory payoff.
93- **Survivor-only sample** treated as representative of venture creation.
94- **Cross-sectional self-report** used to claim a dynamic entrepreneurial process.
95- **Convenience startup sample** with an undocumented frame.
96
97## Output format
98
99```
100【Question→design fit】design chosen + why it fits the entrepreneurial claim ...
101【Selection】founding-choice strategy ...
102【Survivorship/attrition】sampling-frame handling ...
103【Data novelty】construction + coverage + frame ...
104【Temporal precedence】how ordered in time ...
105【Artifacts】MethodsX / Data in Brief? ...
106【Next step】jbv-data-analysis
107```
108
109---
110
111**Source:** [`brycewang-stanford/Awesome-Journal-Skills`](https://github.com/brycewang-stanford/Awesome-Journal-Skills) → `Journal-of-Business-Venturing-Skills/skills/jbv-methods/SKILL.md`
thedixitjain