Notes on writing

This is a collection of notes on writing, mostly based on experience, and not in any particular order. It is a work in progress.

Writing is a skill that takes practice and study. Writing a thesis, paper, or professional report, is significantly different from writing a report for a course. The objective is no longer just to pass or get a good mark. The objective is excellence, and that takes a lot more effort and time. Allow plenty of time for writing, it always takes longer than you think. Always consider – if you were on the receiving end, would you think it was well written?

Literature review

  1. The literature review of a paper, report or thesis provides context for your own work. It defines the current state of the art related to your problem area.
  2. The most common error in writing a literature review is to omit any summary and conclusions. It is essential to provide an overview at the end, to summarize in your own words the state of the art, limitations, and gaps.
  3. The summary and conclusions serve as a link between what others have done, and what you have done – and why you have done it in that particular way.
  4. Do not refer directly to comments by a well-known authority. Although they may be relevant, they have not been rigorously scrutinized by the peer review process, so whenever possible, refer to the published literature, not to “according to Dr. X….”.

Theses

  1. A thesis should be a stand-alone record of your research. Therefore, it should contain data used, software, any algorithms developed etc. These can be in appendices or archived files established with DOI numbers. Your findings should be reproducible by future researchers based on the completeness of records you leave in your thesis.
  2. If your intention to publish papers about your research results (this should always be the case) it is more efficient to make use of the paper style thesis than the traditional document style thesis. The paper style contains the manuscripts, but still needs additional information to provide context before and after the papers. There will be some repetition if more than one paper is prepared, but that is normal.

Papers

  1. When submitting a manuscript to a journal, it is typically written in single column double spaced text, not in two column. The journal will usually specify the format for the manuscript (along with suggested reviewers, highlights, and other requirements). Make sure you get these right.
  2. Papers are very concise and densely written. Editing must be done down to the right word.
  3. It’s important to select the right journal for publication. Check the objectives of the journal to make sure the topic is a good fit.
  4. Check the types of papers published in some recent issues of your target journal to get an idea of style, length, whether your topic is a good fit to the journal.
  5. Check impact factor if available – you want to have wide readership and have your paper published in a well-respected journal to build up your own credentials.
  6. Check the characteristics of other papers in the selected journal for length, formatting etc. There are usually guides for authors on the journal web site. The editors don’t want to waste anybody’s time so usually provide guidance on these aspects.
  7. The objectives of the paper must be very clear. Normally they should be stated in the first section, after the motivation or background has been described.
  8. To keep the paper short, the focus must be sharp. If there are too many topics, consider splitting into multiple papers.
  9. The Abstract should be a short summary of the paper i.e. what the problem is, approach, results, conclusions (or similar).
  10. The paper should end with conclusions, not a summary.
  11. In the discussion section, be honest about the strengths and weaknesses. If you aren’t, a reviewer might be more critical. Readers will appreciate this honesty as well. Some conjecture is okay, but it’s also okay to state that reasons for x are currently unknown.
  12. Papers are about what you have done, as opposed to what you will do next.
  13. Don’t make recommendations for things that should be done as you would in a report. Again, the focus is on what has been done, not what will be done if you get another chance.
  14. When preparing Figures, many journals charge for colour in the printed journal but not in the online pdf version. Keep this in mind as lots of colour Figures could be expensive. It is a good idea to design your Figures with monochrome printing in mind anyway (use dashed, solid lines, different symbols that can be easily distinguished), since most people still use monochrome laser printers (although this is changing) when printing pdf’s.
  15. Proof read repeatedly. Come back to the report after a break of a few days if time permits. It is surprising what you will see with fresh eyes.
  16. If possible, get someone else to proof read your paper. Recognize that people have different writing styles, but well-written papers can be written in more than one way. Whatever the style, quality is important, and that is what you want your proof reader to help ensure.
  17. When reading reviews, do not take the criticism personally. Remember that reviewers have spend many hour of their own time for free to provide that advice. It is in their interest to help maintain the quality of the publications that are eventually printed and most do a very good job of that.
  18. It is your responsibility to address each and every point of the reviewers. Most reviewers provide numbered comments making this process easy.
  19. You must provide a convincing response to the editor of the journal in order to have your paper published. This usually includes rewriting sections of the paper, usually to provide more detailed explanation for clarification purposes. Explain exactly how you have responded and where in the text the modification can be found.
  20. If the reviewer comments have been particularly helpful, acknowledge that fact in the acknowledgements section of the paper. Read some papers to see how common this is!

Reports

  1. In the first paragraph or close to it, make the purpose of the report clear.
  2. Next, provide context of the conditions under which the report was written or terms of reference, including a list of information that was made available. If this is not clear, then somebody reading the report several months or years later might question why you did or did not make use of certain information.
  3. Divide the report into logical sections, and subsections. Lead the reader through the material in a logical manner.
  4. The structure of reports varies more than in papers. Papers tend to follow a well-used overall flow of Introduction->Methodology->Results->Discussion->Conclusions. Report structure depends much more on the subject being reported on. Because of this, it is very important to START by planning the overall structure, then filling in the details.
  5. It is good modern practice to have an Executive Summary at the beginning of a report, placed after the title page and before the Table of Contents. This should be a concise summary of what was done, the main points and conclusions.
  6. The Summary or Conclusions section at the end of a report should be more detailed than the Executive Summary. It is best not to simply copy the Summary or Conclusions into the Executive Summary section. Make them different.
  7. Break up pages of prose using bullet lists, tables, figures etc. to vary the appearance.
  8. In engineering reports, bullet lists are your friend. They concisely highlight a series of related points.
  9. It is also good practice to highlight important points or words using italics.
  10. Reports should be as short as possible, but no shorter.
  11. Place lengthy detailed explanations of secondary issues in an Appendix. Keep the flow of the main report focussed on the main purpose. This helps maintain engagement of the reader but provides additional information if the reader wishes to make use of it.

Citations

  1. Facts that can be considered as common knowledge do not need to be cited.
  2. Referring to interpretation of facts that are common knowledge does need to be cited (since somebody else made that interpretation).
  3. If you are not sure how “common knowledge” something is, provide a citation: When in doubt, cite.
  4. Paraphrasing must also be cited, since you are just rewording published work of others. Nothing wrong with this as long as it is cited.
  5. Direct quotation must obviously be cited.
  6. Try to use up-to-date citations. Reviewers might question how current your work is if the most recent references are several years old.
  7. Journals use their own preferred citation style, so when submitting to a particular journal, make sure you check the citation style required.

Use of Figures from Other Papers & Reports

Almost all published work is subject to copyright, and use of figures or text from such publications requires that permissions be obtained. Most journals have information on how this permission is to be obtained, but you cannot simply cite the work – written permission must be obtained.

A useful site with general information about using copyrighted material can be found here.

Links to pages describing how to obtain permissions from common publishers:

Elsevier

Springer

Reviewing your manuscripts

  1. When writing, reviewing is an on-going process. Continually re-read what you have written.
  2. Be your own worst enemy, so others don’t have to be.
  3. Are you proud of what you have written, and will readers be impressed? If not, then re-write.
  4. In undergraduate courses, there is limited time and students hand in whatever they can manage in the time available. For a thesis or paper, the process is different. You need to take the time to prepare the document to the highest standard.
  5. If you do not submit work of a high standard, it could be rejected. At worst, it will be accepted and a low quality document with your name on it will be published for all to see.
  6. Get someone not related to your work to review it at some stage. Fresh eyes can pick up problems that you are too close to see.

Miscelaneous

  1. Numbers and units should be separated by a space, e.g. 35 MPa, not 35MPa.
  2. Whatever voice you write with, be consistent. E.g., if you use “we found” or “it was found” use throughout. Also, use consistent tense.
  3. Use present tense for descriptions in Figures e.g. do not say “the graph indicated…”.
  4. Be very careful with singular vs plural items and the corresponding verb. Plural nouns take plural verbs.
  5. Data can be both singular or plural, and both forms are now accepted. However, some publications have specific requirements and may list the correct use in a style guide – check it carefully if there is one. Data is a complicated word, so for more information, check online.
  6. Do not use nonscientific terms such as: thing, huge, amazing etc.