Paper submission checklist

Proof-read the text

• All results mentioned in the text are shown in an exhibit (table or figure). If that is not the case, make sure this is intentional and no matching exhibit should be included.
• The value of all results mentioned in the text match the values in the corresponding exhibit (if one is referenced).
• All claims in the texts are substantiated:
  • Create a separate doc and paste in each factual claim.
  • For each claim, check that it is worded accurately and precisely.
  • For each claim, find evidence.
  • For most claims add a reference (see below how to check references). Claims which could be considered ‘general knowledge’ may not need to be cited.
• Abstract and introduction include a summary of results.
• Introduction and literature review take no more than five double-spaced pages (or there is a good reason why this is not the case.
• The data is described in the paper, including provenance, number of observations and descriptive statistics.
  • If any data was dropped, the criteria to do so are clearly described.
• Identification strategy is described in a manner that is intelligible to a non-economist.
• Empirical procedures are described in enough detail that someone could replicate the work.
• All words are spelled correctly (particularly names and foreign words).
• No sentences are unfinished and all comments have been removed.
• Names are used consistently throughout the text (for example, the same round of data collection is not referred to as both “midline” and “follow-up 1”; not referring to the same characterist as “gender” and “sex” in different places).
• Voice and person are used consistently throughout the text (royal we vs passive voice).
• Quotation marks are displayed properly (particularly when using LaTeX).
• Use of empty adjectives and adverbs is minimal.
• Word choice and sentence construction are not unnecessarily complex.
• All text inside tables was also checked.
• Someone outside the team has also proof-read the text.

Check journal guidelines

• Look at the guidelines for the journal you are submitting to. They usually have a style file that you can plug into LaTeX.
• Check if there is a word count or character limit.
• Make sure the references follow the citation style indicated by the journal.
• Follow the journal’s guidelines for captioning, titling, and formatting exhibits.
• Check if the journal requires any additional documentation, such as CONSORT diagram, data deposit, IRB details, etc.
• Check guidelines on abstract, required sections, and section ordering.
• Include the required number of keywords .
• Check if the journal requires a list of JEL codes to be included.
• Check the file formats required for the main manuscript (tex, PDF, Word).
• Check if there are specific requirements for exhibits (such as file format and resolution).
• Check guidelines for disclosure, funding, and conflict of interest statements.

More on citation styles JEL classification system

Review results and summary statistics tables

• All columns and rows clearly labeled.
  • Labels are understandable to someone unfamiliar with the project.
  • It is clear how values should be interpreted (for example, it is clear whether a higher value for an index is desirable or not).
• There are no unintentional line breaks and no text is cut-off.
• Formatting is consistent across all tables.
• The number of decimal cases shown makes sense for all the values displayed.
• Values in the table are consistent with the content they represent (for example, dummy/categorical variables should only take values between zero and one, variables like income and age should never be negative).
• Table titles are self-explanatory.
• All tables include basic information such as sample size and R*.
• The table notes contain all the necessary information for it to be understood as a self-standing exhibit, including
  • Variable definitions describing level of winsorization, imputation, definition of indexes, logs, etc.
  • Sample description (and explanation of why the number of observations is different across columns if that is the case).
  • Description of symbols used to indicate different levels of significance.
  • Description of how missing values and zeros are treated.

Regression tables

• The magnitude of effects is clear (consider adding the sample or control mean).
• Test results for any conclusions mentioned in the text are included (such as equality between coefficients and significance of effects in subgroups).
• Only interpretable coefficients are shown (for example, for regressions with fixed effects, the constant is not included).
• Tables show standard errors rather than t-statistics.
• Methods used to calculate standard errors are described in the table notes.
• Control variables or fixed effects used are shown in the table or described in the botes.
• Weighting method used is described in the notes.
• The estimation procedure used is clearly indicated in the table, title or notes (e.g. OLS, probit, logit).

More on reviewing regression outputs

Review graphs and figures

• All axes are clearly labeled.
• All the units in the graph are clearly defined.
• Axes units match the data.
• No information that is distracting or unnecessary to the graph’s message in present.
• Color distinction will be visible if printed in black and white.
• Color distinction is not affected by color blindness.
• Resolution is high enough for printing.
• Graph notes include all the necessary information for it to be understood as a self-standing exhibit
• Harmonize decimal places (if present) across all graphs/ figures
• Make sure image sizes are consistent.

Chrome extension to simulate color blindness

Check author’s names and affiliation

• All names are spelled correctly.
• Affiliation is up to date.
• A corresponding author has been indicated.
• The email for the corresponding author is up to date.

Check numbering of sections and exhibits (tables and graphs)

• All exhibits are numbered and cited in numeric sequence.
• All exhibits present in the paper are referenced in the text.
• Appendices are numbered differently from core sections.

Check bibliography

• All references cited in the text, tables and notes are included in the bibliography.
• All references listed in the bibliography are cited in the paper.
• If using a numbered citation style, all references are numbered in the same order as they are cited in the text.

Reference management software: Zotero