Don't Format Manuscripts

Journals should use a generic submission format until papers are accepted.

"Dear Dr. Scaramouche, your manuscript has now been reviewed. Based on the comments made by the referees, I decided to reject your paper for publication in our journal. Although I realize you will be disappointed by this decision, I nonetheless hope that the comments made by the referees will be helpful for you to resubmit your ms to another journal."

Researchers are evaluated on the quantity of papers and the "quality" of the journals in which they publish, so there is pressure to submit to the "most prestigious" publication possible. Partly for this reason, the chances of seeing one's paper accepted at first submission are meager, and nobody is spared the high rejection rate. Of course, rejected papers are rarely junk science that do not deserve publication, so most of them are simply resubmitted. Resubmission involves...

Mathematicians and physicists have circumvented the problem by using LaTeX software, which uses automatic templates provided by most journals in those disciplines. But, as it requires a basic working knowledge of computer programming, nonprogramming scientists seldom use LaTeX. For the majority of scientists, resubmission is a substantial time loss.

We surveyed journal editors to assess LaTeX usage in different scientific disciplines and to estimate the number of submissions until manuscript acceptance.

We randomly selected ten journals from each of 13 different scientific disciplines of the ISI Web of Knowledge database (Table 1). For each journal, we recorded the guideline words count (proxy of the time spent formatting) and the number of manuscripts published in 2007. Additionally, we sent a simple questionnaire to editors inquiring about the number of manuscripts submitted (to calculate rejection rate) with or without LaTeX for 2007. Editors of 54 journals responded to our survey. Rejection rate was available for 35 journals.

We did not detect any differences in the rejection rate among scientific disciplines (P = 0.08). A higher rate of submission using LaTeX was found for Mathematics, Statistics, and Physics (Table 1). The guidelines for authors were longer for non-LaTEX-using disciplines (P < 0.001). And non-LaTEX-users have to deal with one supplementary guideline page (mean = 625 words).

Based on a rejection rate of 60%, an article has to be submitted four times before it has >95% chance of being accepted. The time devoted to formatting thus is not negligible.

  LaTeX rate SE N Guideline words count SE N
Mathematics 96.9 3.1 4 604.4 158.5 10
Statistics and Probability 89.1 4.0 6 1208.6 259.4 10
Physics 74.0 11.7 3 2912.1 970.3 10
Computer Sciences 45.8 40.9 2 1354.3 195.3 10
Astronomy and Astrophysics 35.1 21.4 5 3171.4 856.8 10
Engineering 1.0 1.0 2 1429.8 169.5 10
Geosciences 0.8 0.6 6 2284.6 439.0 10
Ecology 0.4 0.4 4 3212.4 1160.7 10
Chemistry 0.3 0.3 3 1739.2 222.1 10
Biology 0.0 0.0 4 1879.7 402.7 10
Medicine 0.0   1 2490.8 362.3 10
Psychology 0.0 0.0 7 1646.5 267.6 10
Sport Sciences 0.0 0.0 6 1663.7 509.4 10
Mean/Total 26.8 5.6 53 1979.6 162.3 130
Table 1. Summary statistics (mean ± SE) of the use of LaTeX in science disciplines (% of submitted papers) and the number of words contained in the guidelines for authors. The red line indicates the separation between LaTeXusing and non-LaTeX-using disciplines.

Use of LaTeX is not widespread in the scientific community: Researchers from disciplines requiring little programming skills obviously won't switch software.

We challenge the need for formatting an article before acceptance. The assessment of the scientific value of a paper doesn't require meticulous formatting. Editors should be satisfied with a readable, generic submission. Authors would format the article after acceptance. This is a simple way to save time and unnecessary effort: We call upon journal editors and publishers to take action.

F. Brischoux and P. Legagneux, Centre d'Etudes Biologiques de Chizé, CEBC-CNRS UPR 1934, 79360 Villiers en Bois, France

F. Brischoux now at: School of Biological Sciences, Heydon—Laurence Building A08, The University of Sydney, NSW 2006, Australia (

P. Legagneux now at: Departement de Biologie & Centre d'Etudes Nordiques, Pavillon Vachon, 1045 avenue de la Medecine, Universite Laval, Quebec, Quebec, G1V 0A6, CANADA (

Correction: The original article said physicians largely use LaTeX software – it meant to say physicists. We regret the error, which has been corrected.

Interested in reading more?

Magaizne Cover

Become a Member of

Receive full access to digital editions of The Scientist, as well as TS Digest, feature stories, more than 35 years of archives, and much more!
Already a member?