Preparing your article for submission
General guidelines
Political Analysis welcomes files submitted in Microsoft Word and TeX/LaTeX formats, however, during the review process it is recommended that LaTeX users submit an Adobe PDF version of their manuscript. Manuscripts should be double-spaced, in 12-point font, include page numbers, and not include line numbers (line numbers will be added by the online submission system). Footnotes should be placed at the bottom of the appropriate page (not endnotes). If using our Overleaf template, please do not modify the formatting settings.
Authors should always strive to format their manuscript with the eventual reader in mind. Manuscripts that are overly long for their contribution, which are formatted inappropriately, which are unreadable or illegible, or which are not otherwise appropriate for review will in most cases be returned to the authors for reformatting (though in some cases the Editors may close the file without external review for some unreadable, illegible, or overly lengthy submissions). An Overleaf LaTeX template for Political Analysis is available, which can support authors in formatting their manuscripts prior to submission.
Throughout the review process, authors should in most cases attempt to position tables, figures and other graphical materials as they should appear embedded in the context of the text. Placement in the text makes it easier for reviewers and editors to read a manuscript. Authors are cautioned, however, that they should never make a table, figure or other graphic so small when embedded in the manuscript’s text that it cannot easily be read by a reader using a small screen (for example a tablet or laptop computer). Upon conditional acceptance, authors are required to submit tables and figures in separate files. If figures contain embedded text or labels, authors should use Verdana or Arial fonts (in that order of preference) where possible.
Materials that are not part of the original source file but which will be important for reviewers to have access to (for example, graphics that are referenced from within a LaTeX source file, appendices, computer code, or other supplementary materials) must be uploaded during the submissions process as “Supplementary Material”. Note that supplementary materials are different from the replication materials required for the Political Analysis Dataverse.
Numbers and variables. Numbers in the articles and tables should be reported with no more precision than they merit. Careful thought, not computer packages or the need to align tables, should govern how many significant digits are reported. Remember that significant digits are not the same thing as the total number of digits reported. Do not report more significant digits than the standard errors suggest.
Variables should be scaled so as to make the reporting of results as straightforward as possible.
All uncertain numbers should be reported with some indication of how uncertain they are. In general, this is best conveyed by confidence intervals or standard errors.
Equations. The journal adheres to certain mathematical and notational standards. Attention to these standards at initial submission will make it easier to deal with conditionally accepted articles. Clarity of the mathematics, tables, and figures is as important a part of the presentation as is verbal clarity.
The author should strive to make the mathematical presentation as clear as possible. Different subareas use different types of notation; authors should strive to use the clearest notation consistent with their particular subarea. Choice of notation, like choice of word, is the prerogative of the author. As with choices about English sentences, choices about mathematical form should be made so as to clarify the presentation.
Authors of technical works should bear in mind that Political Analysis has a varied readership. Try to avoid formulae and specialized terminology in the introduction. Write crisply but clearly; Political Analysis will provide the space for you to explain any technical results. Good mathematical writing calls for the extra effort involved in revising and reworking the manuscript until it will be clear to most if not all of our readers. For a good test of the "readability" of a paper, the comments of a colleague in another specialty should alert you to problems in comprehension that our heterogeneous group of readers might face.
It should be remembered that equations are part of the text and that equal signs function as verbs. Thus, equations should contain appropriate punctuation. All symbols used in equations must be clearly defined.
The author should choose a notation that makes the argument easier to follow. In particular, a consistent notation should be chosen to represent different types of mathematical objects (e.g., matrices, scalars, and vectors). Authors are advised to adhere to the best conventions of notation unless there is a good reason not to do so. (If possible authors should attempt to use a markup language rather than hard coding typesetting specifics.)
The publisher will handle things like page breaking, white space, etc. Since we will strive for some consistency of notation, it would be better for authors to use constructs like \vector instead of \overarrow. Authors using LaTeX should use the amsmath package.
Equations should be numbered consecutively, with subnumbering (e.g., Equations 5a and 5b) used as appropriate. Appendix equations should be labelled A1, etc. Do not number equations by section.
Manuscript Components
Authors should submit one file (Microsoft Word or Adobe PDF) that includes the complete manuscript with components in the following sequence: title page, abstract, text (with embedded tables and figures), funding, acknowledgements, data availability statement, conflicts of interest, figure legend, references. Supplementary material for online-only publication should be submitted separately from the main manuscript file and should include the author names on the first page.
Title page. On the first page of your manuscript, include the title of your manuscript, each author’s names and affiliation/institution, and contact information for the designated corresponding author. Note: Please do not include an “Author’s Note” on your title page with any of the requested components (such as “Acknowledgements”) as all sections should be written within the manuscript text.
Abstract and Keywords. Abstracts are required for both Research Articles and PA Letters. The abstract (200 words or fewer) should be placed on the title page or the second page of the manuscript. Authors will also input the abstract into ScholarOne Manuscripts during the submission process. Should the abstract change, ensure it is updated in both the online system and the manuscript. The abstract will be included in the invitation for reviewers.
For further guidance on how to prepare your Abstracts and Keywords, please refer to these guidelines.
Text. For Research Articles, all sections should be numbered, with the Introduction being Section 1. While in general only one level of numbering will be needed, as in all cases, section and subsection numbering should be used to clarify argument, not to provide typographical consistency. Articles without sections should not number the introduction.
Tables and figures. Tables and Figures should be clear, easily legible, and quickly understood by the reader; those that require lengthy notes or text descriptions so that readers can understand the material presented therein are in need of simplification and redesign. That is, Tables and Figures should stand alone, and not require the reader to reference the text at all.
Thus, Tables and Figures should minimally contain:
- A title explaining the material concisely and clearly, with information about the outcome variable of other meaningful quantity of interest described.
- Information on the sample time period and number of observations included in the graphic.
- A note or notes that describe clearly what different cell entries or graphed material represents.
- Meaningful variable names or labels, which clearly indicate meaning.
- Clear and documented units of measurement.
- Legends and captions that provide additional information when necessary.
All numbers reported in a Table or Figure that contain estimation uncertainty should report that estimation uncertainty.
Authors should avoid the use of Tables or Figures that span pages. Multiple panels should only be used when there is important information that needs to be compared across the panels. When multiple panels are used in a Figure, the axes must be on identical scales for all panels so that the reader can easily compare the information across panels. Also, authors should avoid Tables or Figures that are so wide that they cannot be viewed in portrait mode.
Tables
Numbers in the text of articles and in tables should be reported with no more precision than they are measured and are substantively meaningful. There generally should not be more than 2 digits reported, unless there is good reason to do so. Variables should be scaled so that they have units that are meaningful to the reader, and so that they produce results that are simple to understand. Variables that have vastly different units should be rescaled so that estimations produced are on similar scales.
In general, the uncertainty of numerical estimates is best conveyed by confidence intervals or standard errors (or complete likelihood functions or posterior distributions). Regardless of whether the manuscript uses conventional null hypothesis testing, tables should not report p-values for tests of the null hypothesis that each coefficient is zero. However, in the case of design based experiments and related procedures (permutation tests, asymptotic approximations, etc.) p-values are appropriate and may be supplied. The use of “stars” to report different levels of statistical significance is also not acceptable, and manuscripts with these on tables will be desk rejected. When discussing statistical reliability in the text, the author may state that statistical reliability for a given estimate crosses a conventional level, stating which one (the arbitrary standard values of 0.05, 0.01, 0.001), although the alternatives above are preferred.
When model coefficients are not easily interpretable by the reader, other more understandable quantities should be produced for the reader along with their estimation uncertainty. These include marginal effects, confidence or credible bounds in figures, or other measures of statistical reliability that can readily understood by readers. The manuscript should clearly state how these quantities of interest were produced or estimated, and the manuscript should focus the discussion on the derived and understandable quantities rather than the less interpretable original estimates.
All tables should provide appropriate summary statistics in meaningful units, not unitless quantities that range from 0 to 1, -1 to 1, etc. For example, quantities like the root mean squared error accompanied by the standard deviation of the outcome variable for regression results, or the expected percentage correctly predicted for limited dependent variable models. If the author wishes to present and discuss a unitless summary statistic (like the percentage of cases correctly predicted), they should also provide for readers a comparison (like the percentage of cases correctly predicted by a null model). Furthermore, all manuscripts should strive to provide for the reader the necessary materials so that they can independently judge the quality of the reported results.
In many cases, authors will find that there is more material they wish to report than will easily fit into a single manuscript of the typical length. In those cases (e.g., appendices, additional tables or figures, reports of robustness checks, and detailed discussion of data sources and manipulations), authors are strongly encouraged to produce supplementary materials for online-only publication. See the Appendices and supplementary materials section below for more information.
Figures
For electronic transmission of papers to the editor and referees during the peer review process, all figures should be embedded in the electronic file. This is easily accomplished in Adobe Acrobat or Postscript; those submitting in other formats may submit figures as separate files if needed.
Authors should only put multiple panels into a single figure when doing so helps the reader understand an argument in the manuscript, typically in situations where the author wants the reader to compare results across the panels in the figure. In situations involving figures with multiple panels, authors must make sure that the vertical and horizontal axes of the panels are comparable across all panels of the figure.
All figures should be produced in portrait (not landscape). Authors can submit for review figures in color or in black and white. For manuscripts accepted for publication, all color figures will be published in color online at no cost.
Upon acceptance, all figures and tables should be supplied as separate files. If figures contain embedded text or labels, please use Verdana or Arial fonts (in that order of preference) where possible.
Halftone images must be saved at 300 dpi at approximately the final size. Line drawings should be saved at 1000 dpi, or 1200 dpi if very fine line weights have been used. Combination figures must be saved at a minimum of 600 dpi. Cambridge Journals recommends that only TIFF, EPS or PDF formats are used for non-dynamic electronic artwork.
For more detailed guidance on the preparation of illustrations, pictures and graphs in electronic format, see here.
Permissions. In order to reproduce any third party material, including figures or tables, in an article authors must obtain permission from the copyright holder and be compliant with any requirements the copyright holder may have pertaining to this reuse. Guidance on how to do that can be found here.
Funding. Details of all funding sources for the work in question should be given in a separate section entitled "Funding" above the "Acknowledgements" section. The following rules should be followed: the full official funding agency name should be given, i.e. "National Institutes of Health," not "NIH"; grant numbers should be given in parentheses; multiple grant numbers should be separated by a comma; agencies should be separated by a semi-colon; no extra wording like "Funding for this work was provided by ..." should be used; where individuals need to be specified for certain sources of funding the following text should be added after the relevant agency or grant number "to [author initials]." An example is given here: National Institutes of Health (CB5453961 to C.S., DB645473 to M.H.); Funding Agency (hfygr667789).
- The sentence should begin: ‘This work was supported by …’
- The full official funding agency name should be given, i.e. ‘National Institutes of Health’, not ‘NIH’ Grant numbers should be given in brackets as follows: ‘[grant number xxxx]’
- Multiple grant numbers should be separated by a comma as follows: ‘[grant numbers xxxx, yyyy]’
- Agencies should be separated by a semi-colon (plus ‘and’ before the last funding agency)
- Where individuals need to be specified for certain sources of funding the following text should be added after the relevant agency or grant number 'to [author initials]'.
An example is given here: ‘This work was supported by the National Institutes of Health [AA123456 to C.S., BB765432 to M.H.]; and the Alcohol & Education Research Council [hfygr667789].
Acknowledgements. Authors should include in this section any non-funder acknowledgements that they wish to include.
Data Availability Statement. Authors should include this section at the end of their manuscript (before References) and update it upon conditional acceptance with their Harvard Dataverse citation (provided to them when they upload replication materials).
Example: Replication code for this article is available at Barnes and Solomon (2020) at https://doi.org/10.7910/DVN/B8JYZW.
Once replication is approved, authors may need to update this statement again to include Code Ocean information and/or updated Dataverse citation. The PA Editorial Office will provide instructions at this time.
Example: "Replication code for this article has been published in Code Ocean, a computational reproducibility platform that enables users to run the code, and can be viewed interactively at https://doi.org/10.24433/CO.0469487.v1. A preservation copy of the same code and data can also be accessed via Dataverse at https://doi.org/10.7910/DVN/B8JYZW (Barnes and Solomon 2020).
Competing Interests. When submitting a manuscript through ScholarOne Manuscripts, the corresponding author will be asked to disclose competing interests of all authors. This declaration will be subject to editorial review and may be published in the article. Authors without competing interests should state that there are none to disclose.
Competing interests are situations that could be perceived to exert an undue influence on the content or publication of an author’s work. They may include, but are not limited to, financial, professional, contractual or personal relationships or situations.
If the manuscript has multiple authors, the author submitting must include competing interest declarations relevant to all contributing authors.
Example wording for a declaration is as follows: “Competing interests: Author A is employed at organisation B, Author C is on the Board of company E and is a member of organisation F. Author G has received grants from company H.” If no competing interests exist, the declaration should state “Competing interests: The author(s) declare none”.
Research with Human Subjects. Papers that use data that required review by the author or author’s Institute Review Board (IRB) should state in the “Research with Human Subjects” section the relevant details regarding IRB approval for the research project.
Citations and references. Political Analysis uses the Chicago Manual of Style 16th ed., author-date system. All sources are cited within the text in parentheses by author’s last name and date of publication, with page numbers as appropriate: (Smith 2007, 65). Authors should consult the Chicago Manual of Style for Reference list citations. Please consider the gender balance of your citation list and when considering contributing to your topic. Below are examples of some common reference types.
Journal Article
Jacoby, J., D. Speller, and C. Kohn. 1974. “Brand Choice Behavior as a Function of Information Load.” Journal of Marketing Research 11 (1): 63–69.
Book
Gerber, A., and D. Green. 2012. Field Experiments: Design, Analysis and Interpretation. New York: W. W. Norton Publishing.
Book Chapter
Lau, R. R. 1995. “Information Search during an Election Campaign: Introducing a Process Tracing Methodology for Political Scientists.” In Political Judgment: Structure and Process, edited by M. Lodge and K. McGraw, 179–206. Ann Arbor, MI: University of Michigan Press.
Newspaper/Magazine/Online Article
Cary, B. 2008. “For the Brain, Remembering Is Like Reliving.” New York Times, September 4. http://www.nytimes.com/2008/09/05/science/05brain.html.
Presented Paper
Teplin, L. A., G. M. McClelland, K. M. Abram, and J. J. Washburn. 2005. “Early Violent Death in Delinquent Youth: A Prospective Longitudinal Study.” Paper presented at the Annual Meeting of the American Psychology-Law Society, La Jolla, CA.
Working Paper
Dyer, L., and J. Ericksen. 1980. “Complexity-Based Agile Enterprises: Putting Self-Organizing Emergence to Work.” CAHRS Working Paper 08-01, School of Industrial and Labor Relations, Center for Advanced Human Resource Studies, Cornell University, Ithaca, NY. http://digitalcommons.ilr.cornell.edu/cahrswp/473.
Computer Program/Software
Borenstein, M., L. Hedges, J. Higgins, and H. Rothstein. 2005. Comprehensive Meta-Analysis, version 2. Englewood, NJ. http://www.meta-analysis.com.
Those who use LaTeX and BibTeX will find that either the harvard or natbib packages will serve them well. Users of harvard should use the apsr option. All users of BibTeX should use the chicago.bst bibliography style. While reference list formatting is only relevant after an article has been accepted, authors can simplify matters by using either the harvard or natbib package and chicago.bst as they begin writing.
Data citation. All of the data (including original and archival data) used in a paper or letter must be appropriately cited. Citations to data must include information that will make it easy for readers to find the original data sources, and for those original sources to be consistently identified in the future. Data citations should not appear in the paper’s author note, acknowledgements, text, footnotes, tables, figures, or supplementary materials. Rather, data citations must appear in the paper’s reference list, and contain the name or title of the dataset, the author or authors, any version information, the date of creation of the version used in the paper, and most importantly a persistent data identifier (for example a DOI).
Some examples:
Bullock, W., K. Imai, and J. Shapiro, "Replication data for: Statistical analysis of endorsement experiments: Measuring support for militant groups in Pakistan", http://hdl.handle.net/1902.1/14840 V5 [Version], September 5, 2011.
Monogan, J., "Replication data for: A Case for Registering Studies of Political Outcomes: An Application in the 2010 House Elections", http://hdl.handle.net/1902.1/16470 V6 [Version], June 3, 2013.
These would be cited in the text of the paper as Bullock et al. (2011) and Monogan (2013) respectively.
Appendices and supplementary materials. The initial submission may contain any material that might help the reviewers as an Appendix to the manuscript. In most cases this would be computer code. For example, papers that make extensive use of Monte Carlo simulations might provide the simulation code in an Appendix. Any such material should be clearly documented. If the manuscript moves successfully through the review process, most materials like these would not be published in the pages of the journal but would instead be provided online as Dataverse replication materials.
Material that is not essential to understanding or supporting a manuscript, but which may nonetheless be relevant or interesting to readers, may be submitted as supplementary material. Supplementary material will be published online alongside your article, but will not be published in the pages of the journal. Types of supplementary material may include, but are not limited to, appendices, additional tables or figures, datasets, videos, and sound files.
Supplementary materials will not be typeset or copyedited, so should be supplied exactly as they are to appear online. Please see our general guidance on supplementary materials for further information.
Where relevant we encourage authors to publish additional qualitative or quantitative research outputs in an appropriate repository, and cite these in manuscripts.
