Topic Selection & JAR Fit (jar-topic-selection)
When to trigger
- You have a dataset or a setting but not yet a question that matters for accounting
- You are unsure whether the idea fits JAR versus TAR, JAE, RAST, or CAR
- The question is "is X correlated with Y?" with no clear stakes for accounting
- You are deciding whether to target the Ray Ball JAR Annual Conference or the Registered Reports track
What JAR is looking for
JAR publishes original research in all areas of accounting and accounting-related topics, drawing on finance, economics, statistics, psychology, and sociology. The journal's gravitational center is empirical-archival capital-markets research in the Ball-Brown tradition: how accounting information is produced, disclosed, audited, and used by investors, analysts, creditors, regulators, and managers. A JAR-fit question almost always (a) concerns an accounting/disclosure/auditing/standard-setting phenomenon, (b) lives in a setting with a credible source of identifying variation (a standard change, a regulation, a discontinuity, a shock), and (c) has clear economic stakes — it changes how we think information flows into prices, contracts, or decisions.
The fit test (run before collecting data)
- Accounting question? Is the dependent or independent variable genuinely about accounting information (earnings, disclosure, audit quality, comparability, recognition vs. disclosure, enforcement)? A pure finance or pure economics question belongs elsewhere.
- Setting with leverage? Is there a shock, rule change, threshold, or staggered adoption you can exploit? "We run a panel regression of Y on X" with no identifying variation is a weak JAR start.
- Economic magnitude? Would a referee see the estimate as economically (not just statistically) meaningful?
- Marginal contribution? Does it move beyond a known result, an out-of-sample replication, or a mechanical mediator?
- Data/code feasibility? Because JAR requires posted data and code, confirm you can build a reproducible package from licensed sources (you usually cannot redistribute raw WRDS data).
Sourcing strong JAR questions
- Regulatory and standard-setting shocks (FASB/IASB adoptions, SEC rules, PCAOB inspections) as natural experiments.
- New or under-used disclosure/text data (EDGAR filings, comment letters, conference calls).
- Tensions in theory (e.g., recognition vs. disclosure, transparency vs. proprietary cost) that a clean setting can adjudicate.
- For higher-outcome-risk questions needing new data collection (surveys, field experiments), consider Registered Reports — in-principle acceptance protects a well-designed null.
Checklist
Anti-patterns
- Data-dredging: starting from "what's in WRDS" rather than a question.
- Finance in disguise: a returns paper with no accounting-information content.
- No identification: a cross-sectional correlation framed as an effect.
- Mechanical relations: regressing a variable on its own construction.
- Replication-only: re-confirming a settled result with newer data and no new economics.
Output format
【Question】one sentence (about accounting information)
【Setting & identifying variation】shock / rule / threshold / staggered adoption / none-yet
【Economic stakes】why the magnitude matters
【JAR fit vs. TAR/JAE/RAST/CAR】fit / redirect because ...
【Track】Regular Manuscript / Registered Report / Ray Ball Conference candidate
【Data-and-code feasibility】reproducible package possible? sources ...
【Next step】jar-theory-development
Source: brycewang-stanford/Awesome-Journal-Skills → Journal-of-Accounting-Research-Skills/skills/jar-topic-selection/SKILL.md
1---2name: jar-topic-selection3description: Use when choosing or sharpening a research question for a Journal of Accounting Research (JAR) manuscript — testing whether it is a real accounting question with a credible setting, JAR fit, and identification potential before any data work begins. Locks the question; it does not build the mechanism (jar-theory-development) or design the test (jar-methods).4---567# Topic Selection & JAR Fit (jar-topic-selection)89## When to trigger1011- You have a dataset or a setting but not yet a question that matters for accounting12- You are unsure whether the idea fits JAR versus TAR, JAE, RAST, or CAR13- The question is "is X correlated with Y?" with no clear stakes for accounting14- You are deciding whether to target the **Ray Ball JAR Annual Conference** or the **Registered Reports** track1516## What JAR is looking for1718JAR publishes original research in all areas of accounting and accounting-related topics, drawing on finance, economics, statistics, psychology, and sociology. The journal's gravitational center is **empirical-archival capital-markets** research in the **Ball-Brown** tradition: how accounting information is produced, disclosed, audited, and used by investors, analysts, creditors, regulators, and managers. A JAR-fit question almost always (a) concerns an accounting/disclosure/auditing/standard-setting phenomenon, (b) lives in a setting with a **credible source of identifying variation** (a standard change, a regulation, a discontinuity, a shock), and (c) has clear **economic stakes** — it changes how we think information flows into prices, contracts, or decisions.1920## The fit test (run before collecting data)2122- **Accounting question?** Is the dependent or independent variable genuinely about accounting information (earnings, disclosure, audit quality, comparability, recognition vs. disclosure, enforcement)? A pure finance or pure economics question belongs elsewhere.23- **Setting with leverage?** Is there a shock, rule change, threshold, or staggered adoption you can exploit? "We run a panel regression of Y on X" with no identifying variation is a weak JAR start.24- **Economic magnitude?** Would a referee see the estimate as economically (not just statistically) meaningful?25- **Marginal contribution?** Does it move beyond a known result, an out-of-sample replication, or a mechanical mediator?26- **Data/code feasibility?** Because JAR **requires** posted data and code, confirm you can build a reproducible package from licensed sources (you usually cannot redistribute raw WRDS data).2728## Sourcing strong JAR questions2930- Regulatory and standard-setting shocks (FASB/IASB adoptions, SEC rules, PCAOB inspections) as natural experiments.31- New or under-used disclosure/text data (EDGAR filings, comment letters, conference calls).32- Tensions in theory (e.g., recognition vs. disclosure, transparency vs. proprietary cost) that a clean setting can adjudicate.33- For **higher-outcome-risk** questions needing **new data collection** (surveys, field experiments), consider Registered Reports — in-principle acceptance protects a well-designed null.3435## Checklist3637- [ ] The question is genuinely about accounting information, not just finance/economics38- [ ] A credible identifying setting (shock/rule/threshold/staggered adoption) is in sight39- [ ] The expected estimate is economically meaningful, not only significant40- [ ] The marginal contribution over existing accounting work is articulable in one sentence41- [ ] A reproducible data-and-code package is feasible from accessible sources42- [ ] Venue fit confirmed: JAR vs. TAR/JAE/RAST/CAR43- [ ] If new-data/high-risk: Registered Reports considered; if conference-relevant: Ray Ball eligibility checked4445## Anti-patterns4647- **Data-dredging**: starting from "what's in WRDS" rather than a question.48- **Finance in disguise**: a returns paper with no accounting-information content.49- **No identification**: a cross-sectional correlation framed as an effect.50- **Mechanical relations**: regressing a variable on its own construction.51- **Replication-only**: re-confirming a settled result with newer data and no new economics.5253## Output format5455```56【Question】one sentence (about accounting information)57【Setting & identifying variation】shock / rule / threshold / staggered adoption / none-yet58【Economic stakes】why the magnitude matters59【JAR fit vs. TAR/JAE/RAST/CAR】fit / redirect because ...60【Track】Regular Manuscript / Registered Report / Ray Ball Conference candidate61【Data-and-code feasibility】reproducible package possible? sources ...62【Next step】jar-theory-development63```6465---6667**Source:** [`brycewang-stanford/Awesome-Journal-Skills`](https://github.com/brycewang-stanford/Awesome-Journal-Skills) → `Journal-of-Accounting-Research-Skills/skills/jar-topic-selection/SKILL.md`
thedixitjain