Writing Will Be Archived Forever

Published Date: 02 Nov 2017

Although writing papers is hard work, writing a good paper is one of the most satisfying and long-lasting accomplishments of scientific research. A well written scientific paper must fulfill two objectives. First, it must clearly and completely describe the procedures that were followed and the results that were obtained. Second, it must place these results in perspective by relating them to the existing state of knowledge and by interpreting their significance for future study.

Keep in mind that your writing will be archived forever and that people (and you!) will refer to it for many, many years. Once published, you can never change, revise, or retract your words. You will therefore want the paper to reflect your absolute best effort. Ten, forty years from now, you want the paper to continue to provide you with a sense of satisfaction and pride.

If you don’t have much experience writing, read Writing Scientific Papers by Tim North, which is available in the Resource section of our Intranet server. You may also want to read Better Writing Skills by Tim North, available in the same place. Please do not distribute copies of these documents outside the group. A number of other resources are available in the library and Eric’s office.

The procedure outlined in this document will help us develop good writing skills and a uniform style, and streamline the process of writing papers. As you follow the procedure in this document, the manuscript will go through the following stages:

Outline — getting the idea and the message right

Draft 1 — getting the logic and structure right

Draft 2 — logic and structure being approved by Eric

Draft 3 — getting the language and details right

Draft 4 — final editing by Eric

Submitted — peer review

Accepted — in production

Published — your work eternalized!

We will use our Web site to keep track of the status of a manuscript from outline to publication. In stages 1–5 the manuscript is visible only to group members. By default the published paper is visible to anyone on the Internet in stage 8. You select the paper’s visibility in stages 6 and 7.

To keep the publication process moving along, it is necessary to set a realistic time-line for all steps. The lead-author must schedule meetings ahead of time to make sure the manuscript doesn’t sit on anyone’s desk for a long time. It is the lead author’s responsibility to keep the manuscript (and the people involved) on track. If the lead author leaves the group, then someone else must be assigned as "coordinator" and be put in charge of keeping the process going.

When to publish?

The publication process begins with a complete and final set of data that tell a compelling message ("Observation B depends on A", "These data confirm/disprove C", "New method to accomplish D", etc.). It doesn’t pay to begin planning a publication if you still have any data pending. New data may invalidate your previous data and if that happens part (or all) of the effort invested in the development of a paper is wasted. Typically, upcoming publications should first be discussed in subgroup meetings. If you feel you have enough material to warrant a publication, bring the topic up at a subgroup meeting.

In general we should strike a balance between: 1) waiting too long to publish data by interpreting "complete" as meaning "everything there is to be known about the topic" and 2) working on too many (small) publications at the same time.

Before taking on any new publication we should evaluate its cost/benefit ratio: is the amount of work required to develop the publication in line with the potential payoff? Potential payoff includes: 1) the need to document claims or accomplishments,

2) establish awareness of our work and 3) professional development (development of writing skills, publication list, contribution to thesis).

Where to publish?

Our goal should be to publish in peer-reviewed journals with the largest possible readership and impact. Magazine articles and chapters in books that are not conference proceedings have the potential of broad exposure and a wide readership even though they are not peer-reviewed. In principle we will not contribute to conference proceedings, technical digests, conference programs that require extensive write-ups, or other non-peer reviewed texts. As a rule of thumb, we simply ignore requests for such publications. Exceptions to this rule are cases where we feel the payoff (see above) would be exceptional or where giving a presentation absolutely depends on contributing anything more than a short, text-only abstract.

Authorship

Refer to the guidelines in Appendix A to determine authorship. Authorship must be decided at the very beginning of the development of the paper and will always be discussed (and approved) by the whole group during presentation of the outline at a group meeting. Also, all papers must have an acknowledgement paragraph or section that details the contribution of every author to the paper. An example of such a paragraph follows.

"Several people contributed to the work described in this paper. [INITIALS] conceived of the basic idea for this work. [INITIALS], [INITIALS] and [INITIALS] designed and carried out the experiments, and analyzed the results. E.M. supervised the research and the development of the manuscript. [INITIALS] wrote the first draft of the manuscript; all authors subsequently took part in the revision process and approved the final copy of the manuscript. [NAME] and [NAME] provided feedback on the manuscript throughout its development. The research described in this paper was supported by [AGENCY 1] under contract [CONTRACT NUMBER] and [AGENCY 2] under contract [CONTRACT NUMBER]."

Papers initiated outside the group

Occasionally we are asked to contribute to papers that are initiated by authors outside the group. In such a case the same guidelines to authorship apply: if we do not satisfy all authorship requirements, we should decline the offer to be included as authors (our contributions can be acknowledged). Most importantly, before accepting authorship on a paper initiated outside the group we must be certain that 1) the quality of the paper will meet our standards, and 2) we will be able to provide input in the development of the paper and review (and approve) the final copy. For such papers it will generally not be feasible to follow the procedure described here. Typically, the paper will be in a form equivalent to "Draft 4" by the time a copy is provided to us. In such a case, the authors from the group will edit the manuscript as if it were an internally generated manuscript in the Draft 4 stage.

Formatting

Throughout the development of a paper, use the templates (outline, paper, figures) available on the Intranet. Adhering to these templates will ensure uniformity, consistency, and efficiency.

If the publisher requires a different template, we will convert the publication to the required format just before submission, once the manuscript is in final form.

The process

Developing a publication for submission involves five stages, each with its own milestone. The publication queue of our Intranet server should always be set to indicate the current stage of the manuscript. In the Draft 2 and 4 stages the manuscript is in Eric’s hands; during all other stages it is in the lead-author’s hands.

Stage 1: Outline — getting the idea and the message right

Preparation of Outline — Lead author with input from other potential coauthors

Form a ‘thesis statement’ — a short (1–3 sentence) statement of what you want the reader to understand as the most important point in your paper. This statement may end up appearing in the abstract or introduction.

Choose a format: letter, full publication, or review

Decide who you are writing for & aim at that level; determine a target journal.

Outline the basic structure of the paper (Appendix B). Use the standard outline (download from Intranet) to begin developing this structure. Ideally you should have a reasonable paragraph-by-paragraph outline of the entire paper, with each paragraph represented by a few key words.

Develop a set of figures to be included in the paper, using our standard templates and procedures (Appendix C). These figures serve to illustrate the procedure and the main results and/or conclusions. Combine the figures in a single pdf file.

Once the above material is put together, discuss (and possibly refine) the material with Eric. When that is done, upload the material to the publication queue and schedule a time to present the thesis statement, figures, and outline at a group meeting.

Discussion/Acceptance of Outline — Whole group

At a group meeting the lead author presents the thesis statement, figures, and outline (in that order). The group evaluates the material on the following criteria:

Correctness and relevance of thesis statement

Completeness of data

Validity of analysis and of conclusion(s)

Appropriateness of publication format: letter, full publication, or review

Completeness and relevance of figures

Workload vs. contributions to (sub)group goals

After discussion of the outline, the following information is decided:

An ordered list of one or more appropriate journals

An author list (see Appendix A)

A minimum of two people who are neither authors nor involved in the research project to provide feedback on a first draft (henceforth the ‘critics’)

By default, the lead author becomes the coordinator for the manuscript. The coordinator should make an entry in the publication queue on the intranet server. It is this person’s responsibility to keep the manuscript moving through the queue and, ultimately, to submit the paper for publication.

Stage 2: Draft 1 — getting the logic and structure right

Development of first draft — All authors

Use the standard publication template in Word (download from intranet) to turn the outline entries into paragraphs. Please adhere to the following format guidelines:

Check the submission guidelines for the target journal so you know the limits on words, figures, tables.

Keep your manuscript double-spaced until publication. For camera-ready publications, don’t worry about formatting until the submission stage. The footer of each page should provide the page number and the header should show the status and version (e.g., ‘Draft 1.n’, n = version), and the date of that version.

References, tables, and figure captions go on separate pages. Tables go on separate pages, one to a page, with the appropriate caption below the table.

Use EndNote to collect a library of references for the paper.

Do not paste any figures in the Word document. Keep the figures in a separate pdf file, labeled as on the Illustrator plot template.

After completing the draft, check the overall structure following the guidelines below. For each of the following points, you should go once through the entire manuscript.

Read only the first and last sentences of each paragraph. The first sentence is the topic sentence, defining the content of the paragraph. The last sentence should close the paragraph and possibly lead into the next. If the last sentence bears no relation to the first, then the paragraph is poorly structured — you have drifted away from the subject. You should be able to recognize the structure from the outline in this first/last sentence reading; if not, you need to restructure. If necessary rearrange the breaks in the paragraphs by logical divisions.

After this overall structure check, make sure that the text within each paragraph agrees with the topic sentence of the paragraph.

Use the evaluation form (Appendix D) to score your draft; rewrite any sections that do not meet the stated guidelines.

At this stage, also, all figures must be in near-final state. Once all of this is done, upload the first draft to the web site, notify the authors and critics of the availability of the first draft and arrange to meet with each author and critic individually to discuss the draft.

Editing of first draft — Authors and critics

Each member of the author/critic team checks the draft for structure and logic (no extensive "wordsmithing" yet) by:

reading the manuscript

scoring it using the evaluation form (Appendix D)

marking up the manuscript with any (major) corrections or suggestions

N.B.: by doing a thorough job and providing good feedback, the authors and critics can greatly enhance the quality of the draft and significantly reduce the time that the whole team spends in meetings developing the next draft (see below).

The coordinator/lead author then meets with each team member individually and incorporates their feedback. The manuscript should be thoroughly edited, by reading it out loud to make sure the sentences are complete and all statements make sense. It is the coordinator’s duty to do a good job so as to raise the quality and level of the paper as much as possible and minimize everyone else’s work. Once this is done, the manuscript reaches Draft 2 stage. The coordinator uploads the latest version of the manuscript, changes the status to Draft 2 and brings a printout of the manuscript to Eric.

Stage 3: Draft 2 — approval of logic and structure by Eric

Eric reads the manuscript to validate the overall approach and structure and returns it — preferably within a week — to the lead author with possible suggestions for improvement. This step changes the status of the manuscript to Draft 3.

Stage 4: Draft 3 — getting the language and details right

Development of second draft — Authors and critics

The lead author addresses Eric’s comments and circulates the revised manuscript to all authors and critics. Everyone carefully proofs the manuscript using the guidelines in Appendix E, and marks up a copy with any corrections or suggestions. The entire team then meets to go over the manuscript, revise it paragraph-by-paragraph, sentence-by-sentence, and fine-tune the language using the guidelines in Appendix E. Once everyone on the team is satisfied, the coordinator prints out the final copy, uploads it to the Web site, changes the status to Draft 4 and brings a printout to Eric.

Stage 5: Draft 4 — final editing

Finalizing the paper — (Authors and) Eric

Options for bringing the manuscript in final form are:

Eric edits the manuscript directly

Eric returns the manuscript to the author(s) with comments/requests for change

Eric schedules meetings with the author(s) to incorporate changes

Regardless of the procedure followed, it is the coordinator’s responsibility to schedule meetings to work on the manuscript and keep the manuscript moving.

Final proofing and approval

When the manuscript has been finalized, it should be proofread one more time by all authors, all critics, preferably one or two group members who have not yet seen the paper, and possibly by people outside the group. All authors must approve the final version of the manuscript before submission.

Submission of manuscript

Once all authors have approved the final version, the coordinator formats the manuscript according to the guidelines of the journal or publisher and submits the paper for publication. In general Eric should be the corresponding author as his address is the least likely to change.

Appendix A — Authorship

Authorship on scientific publications can be a delicate issue. Because of the nature of our research, our publications are almost always multi-authored. The designation and order of authorship communicates the relative contribution of the authors to the research and to the publication of the results. While it is Eric’s responsibility to ensure appropriate recognition and order of authorship, it is important that we all agree on some general guidelines.

In principle, contributions in any of the following areas could warrant authorship: 1) conception and design, 2) drafting of paper, 3) critical revision of the paper/proposal for intellectual content, 4) analysis and interpretation of data, or 5) statistical expertise and or analysis. The American Physical Society states rather summarily "Authorship should be limited to those who have made a significant contribution to the concept, design, execution and interpretation of the research study." The Council of Science Editors provides a much more substantial guideline: "All persons designated as authors should qualify for authorship. Each author should have participated sufficiently in the work to take public responsibility for the content." That last sentence is crucial.

The Council also provides the following very explicit criterion: "Authorship credit should be based only on substantial contributions to

(a) either the conception and design or the analysis and interpretation of data and to

(b) drafting the article or revising it critically for important intellectual content and on

(c) final approval of the version to be published.

All three conditions must be met." So, all authors should be able to check off each of the three statements above and be able/willing to take public responsibility for the content of the paper.

The order of authorship should be a joint decision of the coauthors. In principle, those who contribute most should appear first; those who contribute least should appear last. Because the order can be assigned in different ways, its meaning cannot be inferred accurately unless it is stated explicitly. For example, the order of authorship can be explained in a footnote.

To make the process fair and open we will: 1) discuss authorship during the discussion of the outline and 2) always add a paragraph detailing each author’s contribution to the paper. The order of authorship is left to the coauthors. If anyone has any misgivings about either authorship or the order thereof and there is no opportunity to resolve the issue in a meeting, then these misgivings should be communicated to Eric, so he can try to find a resolution.

Appendix B — Paper Outline (Stage 1)

black: sections and parts; grey: instructions

use the template from the Intranet or cut and paste the text below.

❑ Title

1. Use fewest possible number of words

2. Put topic words near start of the title

3. Include searchable key words in title

4. Suit title to audience

❑ Author list

Authorship should be limited to those who have made a significant contribution to the concept, design, execution and interpretation of the research study. Authorship credit should be based only on substantial contributions to

(a) either the conception and design or the analysis and interpretation of data and to

(b) drafting the article or revising it critically for important intellectual content and on

(c) final approval of the version to be published.

❑ Abstract

The abstract is a very short summary of your paper. The abstract should

1. stand on its own

2. not be overly detailed or technical

3. address what the article is about, whom it targets and what key goals have been reached

4. be concise and readable.

In essence the abstract is a very brief version of the introduction that follows it.

❑ context: what you did, why you did it

❑ major methods

❑ major results

❑ major conclusions

❑ Introduction

The introduction is a summing up of the entire paper in more detail than the abstract. The introduction might be the only part of the paper read by some, so it has to captivate and stand on its own.

❑ nature and scope of the problem

❑ why the work is important

❑ review relevant literature, provide rationale

❑ brief description and justification of experimental methods

❑ major results and conclusions

❑ Experimental

This section should provide enough detail so someone else can repeat the experiments. Describe procedures and experimental setup, do not include any results.

❑ Results

Present results clearly and without interpreting them. Analysis of results can be included in this section, but any interpretation needs to be moved to the next section.

❑ Discussion

Interpret the individual results by discussing their meaning. Connect to your expectations and results in the literature. Discuss possible experimental error. Address any problems in your results (outlying or unexpected data, variations in data, etc.)

❑ Conclusions

Synthesize the discussion into one or more conclusions; tie everything together and give a specific recommendation, if appropriate.

Appendix C — Preparing Graphics (Stages 1–2)

The illustrations we make should adhere to strict uniformity standards. By doing so we can use these illustrations not only in our papers, but also in our presentations and posters and be assured that there is a unified look, even if the illustrations originated with different publications and/or people in the group. All illustrations will have the same size, use the same fonts, and be of identical style.

You should prepare all illustrations in Adobe Illustrator and save them as Illustrator pdf files. Use Acrobat to combine all the Illustrator pdf files into one file, which you can upload and eventually append to the manuscript.

There are three types of illustrations:

Plots

Diagrams (e.g., schematics of setups)

Photographs

Templates for all three of these are on the Intranet, together with instructions for use. The easiest way to learn how to use these templates, however, is to follow along with someone who has become expert at using them.

The bounding box of all illustrations is 420 x 315 points. Typically the publisher will reduce this format to single column width. Working at this large size, however, allows you to work with greater precision than if you were to make it at the final resolution. Use only the Helvetica Font. The default size is 16 pt, although some labels are best at 18 pt. Stroke weights should always be 1 pt, except for curves and lines in graphs, which should be 2 pt. Symbols in graphs should be 8 pt in diameter. Pictures (photographs) should be at a minimum resolution of 225 dpi — this ensures that they will be at least at 300 dpi when reduced to single column format.

Appendix D — Structure Evaluation Form (Stage 2)

Use the table below to score manuscripts

(2 = conforms to standards, 1 = needs improvement, 0 = needs reworking)

Section

Criterion

Comments

Score

Title

Descriptive and appropriate?

Topic word near beginning?

Contains searchable keywords?

Abstract

Clear, concise, and complete description of main message?

States what was done and why?

Presents major methods used?

States major results?

States major conclusions?

Introduction

Sufficient background information so purpose of study is clear?

Hypothesis and/or objective(s) of study clearly stated?

Adequate review of literature?

Experimental

Are experimental methods clearly described?

Is the research design appropriate?

Results

Clear description of results?

Figures and/or tables referred to in text?

All relevant data presented?

All data presented relevant?

Section free from interpretation? (interpretation goes in discussion)

Results adequately analyzed?

Discussion

All relevant data discussed?

No additional data presented? (if so, move to Results)

Data justifies conclusions and/or interpretations?

Facts, calculations, and interpretations correct?

Interpretations and/or conclusions compared with published work?

Conclusion

Does end reinforce main message and connect work to big picture?