Tables & Figures for JOM (jom-tables-figures)
When to trigger
- Tables are cluttered, inconsistent, or not self-explanatory
- You need the standard empirical-OM exhibit set (descriptives, correlations, models)
- An interaction/contingency effect needs a readable plot
- A field/case/intervention study needs a data-structure or process figure
- You are reconciling exhibits to APA style for Wiley
The empirical-OM exhibit set
A typical JOM empirical paper carries:
-
Sample/measure table — constructs, items, sources, reliabilities (survey) or variable definitions and data sources (archival).
-
Descriptives & correlations — means, SDs, full correlation matrix; reliabilities on the diagonal for survey constructs. Define every operational variable.
-
Results tables — nested regression / FE / SEM / count / survival models in columns, with coefficients, standard errors (note the clustering), fit/diagnostics, and N. Report effect sizes, not just significance.
-
Interaction/contingency plot — simple slopes with significance regions for any moderation (contingency effects are central to OM).
-
Mechanism/process figure — the hypothesized model; for field/case/IBR, a Gioia-style data structure or a process/intervention timeline.
House-style rules
-
APA conventions for tables, figures, notes, and references (JOM uses APA; Wiley applies final styling at proof). Keep formatting consistent — at first submission any consistent style is accepted, journal style preferred.
- Number tables/figures; give each a stand-alone title and a complete note (estimator, SE type, significance thresholds, N, units).
- Manuscript body is double-spaced, single-column, 12-point, one-inch margins, numbered pages, no running headers/footers; place exhibits per the author guidelines.
- Every exhibit must be readable without the text and must report the operational units (days, defects per million, inventory turns, on-time %).
Self-explanation test
A reviewer should grasp each table/figure from its title, note, and labels alone. Spell out abbreviations, state the estimator and clustering in the note, and never show a coefficient without an SE and an N.
Execution bridge (StatsPAI / Stata MCP)
Generate exhibits from the fitted result, not by retyping numbers (the usual source of
body-vs-appendix drift). Full map: execution-with-mcp. JOM is empirical operations / supply-chain — survey and archival panel data; foreground endogeneity of operational choices and clustered / multilevel inference.
-
Tables:
etable (multi-model columns) or did_summary_to_latex straight from the
result_id.
-
Figures:
plot_from_result / enhanced_event_study_plot / event_study_table —
axis units and the SE/clustering note baked in.
-
Every note names the estimator + clustering and states the effect size in
interpretable units.
See a full fitted-result → exhibit chain in the JF execution walkthrough.
Anti-patterns
- A correlation table with no reliabilities (survey) or undefined operational variables.
- Reporting stars but no effect sizes / operational magnitude.
- An interaction described in text but not plotted.
- Exhibits that restate the text instead of carrying evidence.
- Inconsistent citation/number style across exhibits.
Exhibit-completeness table reviewers check
Each empirical-OM exhibit has elements without which a reviewer cannot verify the claim. The map below is a practical completeness check; APA and Wiley styling specifics should be confirmed against current author guidelines.
| Exhibit |
Must contain |
Common reviewer complaint |
| Measures/sample table |
Constructs, items, sources, reliabilities (survey) or variable definitions and data sources (archival) |
Operational variables left undefined |
| Descriptives + correlations |
Means, SDs, full correlation matrix, reliabilities on the diagonal |
No reliabilities; no operational units |
| Results table |
Coefficients, SEs with clustering noted, fit/diagnostics, N, effect sizes |
Stars without effect sizes or operational magnitude |
| Interaction plot |
Simple slopes with significance regions |
Moderation described in text but never plotted |
| Mechanism/process figure |
Hypothesized model, or Gioia data structure / intervention timeline |
Figure restates text instead of carrying evidence |
Desk-reject and return triggers on exhibits
- A correlation table with no reliabilities (survey) or with undefined operational variables.
- A coefficient shown without a standard error and an N.
- Significance reported but no operational magnitude (days, defects per million, inventory turns, on-time percentage).
- Inconsistent number/citation style across tables, signaling rushed preparation.
Worked vignette: turning a result into a self-explanatory table
A behavioral-OM experiment finds a fatigue manipulation raises order errors, more so under high time pressure (illustrative). The results table reports the main effect (b = 0.42, SE = 0.11) and the interaction (b = 0.27, SE = 0.09), with N = 180, robust SEs noted, and a partial eta-squared column so magnitude is visible. The note states the estimator, the SE type, the thresholds, and that the outcome is errors per 100 orders — an operational unit. A reader who never sees the body can still grasp the story from the title, note, and labels. The companion plot shows simple slopes with significance regions, satisfying the contingency-effect expectation native to OM.
Exhibit objections reviewers raise, with the fix
-
"The table is not self-explanatory." Add a complete note (estimator, SE clustering, thresholds, N, operational units) and stand-alone title so it reads without the text.
-
"You report significance but not magnitude." Add effect sizes and translate them into operational consequences.
Output format
【Exhibit set】measures / descriptives+corr / models / interaction / mechanism
【Result table】estimator, SE clustering, effect sizes, N present? ...
【Interaction plot】simple slopes + regions? ...
【Process/data-structure figure】(field/case/IBR) ...
【APA consistency】...
【Next step】jom-writing-style
