Writing Paper

2.672 Writing a technical paper

Prof. Wai Cheng Department of Mechanical Engineering, MIT

Elements of a technical paper

  • Title
  • Abstract
  • Introduction
  • Theoretical background
  • Apparatus and procedure
  • Results and discussion
  • Conclusions
  • Nomenclature (optional)
  • References
  • Appendices (optional)

2.672 writing

  • Individual reporting
    • You share ideas, data and results with your group members. You need to write your own project report though.
  • Writing style
    • Avoid subjective comments, and use of personal pronouns; use passive voice instead
    • Write directly, avoid words that are not useful such as:
      • “In order to”
      • “the purpose of this experiment is to”
    • No. of significant figures in numbers should reflect accuracy of measurement

Paper versus report

  • Report
    • Usually for internal use
    • Document details of the project for archival purpose
      • Include details: drawings, computer programs
  • Paper
    • For external audience
    • Succinct and to the point
      • Do not need to write about routine details such as calibration procedure etc.
      • Do not put computer program listing in a paper

Title

  • The title is what draws reader to your work
    • Reveal the topic of the paper
    • Should include key words about the project
  • Make the title interesting and attractive

Abstract

  • Brief condensation of the paper
    • 150 – 200 words
    • Do not explain why the study is done in the abstract
  • What was done?
  • How was it done?
  • What are the significant results

Introduction

  • Introduces subject
    • Background and context
    • Articulate the need for the study
    • Clearly define the problem (purpose of the investigation)
    • Briefly outline overall approach

Theoretical background

(Should use a more meaningful heading than theoretical background)

  • Physical explanation of the phenomena involved
  • Develop governing equations
    • Assumptions: support them quantitatively
      • e.g. laminar flow — give Re
    • Describe model development
      • Give the key equations only; put details in the appendix
  • Connect your theory to your project
    • What theoretical values are to be compared to experimental results?

Apparatus and procedure

  • Overview of operation
    • Refer to a schematic to explain the operation
    • Give dimension of apparatus
      • Relationship between the laboratory device and the real device
    • What are being measured?
      • Do not need to give details about transducers if they are common devices
  • Experimental matrix
    • What are the variables and what range has been covered?

Results and discussion

Results

  • Use figures
  • Describe the direct observation first (e.g. pressure vs. time)
    • Point out the features and the physics behind them
  • Show how do the results change when you change the variables of the experiment
    • Magnitude (up or down; by how much?)
    • Trends and scaling laws
      • linear, exponential, …

Discussion

  • Does theory produce the same features as the observations?
  • Plot theoretical values on the same graph as the experimental results
    • Explain the differences
  • Make the results (theory and experiment) useful

Conclusions

  • Summary of your finding
  • Pronounce your judgment
    • What are the key parameters?
    • How are the results related to these parameters?
    • How good is your model? What does it capture?
    • How does your study contribute to the objective stated in your introduction?

Appendices

  • Details that your reader may not need to follow the overall picture but are required to support your work
  • Need to have a narrative to describe any equation, table, or graph in the appendix

Further remark: graphs

  • Caption should be short but informative and comprehensive
  • Axes should be labeled
    • For dimensional quantities, units are required
    • Use symbols for data points and lines for theoretical values
    • For multiple curves, mark each curve clearly

HOW TO WRITE A SCIENTIFIC PAPER

This article originally appeared in April 1989, Volume 98, No. 4, pages 243-244, and is reproduced here in its entirety.

Writing a good scientific paper is difficult. No one ever said it is easy. After having written over 150 articles, I find it less arduous and painful than 30 years ago, but it is still difficult. So, if you find that the next paper you write flows freely from your fingertips with painless ease, tear it up. It is probably no good.

First, why write a scientific paper? It is because of the audience you can reach with what you believe to be an important message. In presenting your data and conclusions through discussions with your colleagues, you can reach dozens. At a national meeting, you can reach hundreds. Only through publication in an international journal can you reach thousands.

A great advantage in publishing in a journal is that good journals have a process of critical review. The editor chooses two or more international experts in the specific area of your paper who meticulously go over the paper, make written detailed criticisms of the manuscript, and make one of three recommendations: accept for publication as is (5%), reject (50%), or return to author for revision with advice (45%). Any manuscript can be improved (even this Editorial, which has gone through five revisions with the help of nine critics). I have found that our very good authors appreciate the help and suggestions provided by reviewers just as much as the average author, and in some cases even more. You will find very few editors of major journals who routinely accept/reject all or a majority of manuscripts. Such an editor is not doing the job.

Now, to the matter of actually writing the article. There are a number of categories of articles, such as basic laboratory research, applied (clinical bearing) laboratory research, retrospective case analysis research, prospective clinical trials, and review articles. The following will as a rule apply to all the foregoing except the pure review article.

The article generally should be divided into seven clear sections, each with its specific purpose and containing only the material specific to it. There should be little cross-information among sections, because this leads to confusion engendered by fuzzy thinking. They are 1) Abstract, 2) Introduction, 3) Materials and Methods, 4) Results, 5) Discussion, 6) Conclusions, and 7) Summary.

1. Abstract. This is a single paragraph that leads the article. It states in a few sentences the purpose (intent) of the article, briefly how it was carried out, the key findings of the study, and the conclusions reached. No references are used, and no prior work is mentioned in the abstract (unless it is key to the sense of the article, e.g., a refutation).

2. Introduction. The Introduction includes a statement as to the reason(s) you did the study (confusion in the field, significant differences between your data and others', an advance in technique or analysis or result, etc.), how your results bring new light to the problem (why the reader should be interested in what you are saying), and what you are going to prove (demonstrate, improve upon, extend, refute, etc.), and why it is important to do so. The work of others should be pointed out extensively in this section by references. This will indicate the work done to date so readers are aware that your work is current. References should not be exhaustive as in a master’s thesis or doctoral dissertation; only key references should be noted and number usually one or two dozen.

3. Materials and Methods. For most authors this is the easiest place to start the writing, since it is the most straightforward. It includes the population analyzed and the reasons for it, standardization of the population for research and statistical purposes, the way(s) the population was analyzed, and the justification for doing so. The population data naturally appear here. Data units are defined appropriately and referenced. In research articles it is important to include details that enable other investigators to replicate your study (strains of animals used with age or weight, lesioning methods, methods and instrumentation used in data collection, etc.). The outcome data (results) can appear here, or in the Results section, depending upon the type of paper and the flow of the argument, but generally as a separate section.

4. Results. This section is, of course, the heart of the article. It is the reason for seeking publication. It thus must be able to totally withstand scientific scrutiny, and not simply be descriptive like the Materials and Methods section. The "n" must be significant; exclusions, if any, justified and cited; the controls adequately set up and sound; and p values distinguishing them significant. Appropriate tables must be set up, and by that I mean those necessary to document a point made in the text. Those points adequately made in the text need no tables. Here will appear experimental illustrations documenting findings, with findings clearly pointed out by arrows or labels, and crisp, concise legends.

5. Discussion. In the Discussion, a restatement of the problem defined in the Introduction is made, the appropriateness of the experimental approach as put forth in the Materials and Methods is explained, and the hypothesis stated in the Introduction is identified as supported qualitatively and quantitatively (that is, statistically) by the results. In this section, it is important in most types of papers to iterate the significance of the work performed, unless it is obvious. It is also permissible to speculate on the impact of your findings, wherever it may be appropriate.

6. Conclusions. In this section, which is seldom more than a paragraph, a statement as to what you believe you have proven or demonstrated is made. Conceptualizations and not data are appropriate to this section. Also, a part of this section is a statement as to what further work is indicated to provide further definition of the problem, and what importance there is in doing so.

7. Summary. This is the least necessary section of the group, and should be used only if a strong final statement seems needed, such as when the work presented strongly supports prior work in a controversial area, or clearly refutes prior work that has been accepted generally.

Now that the science of your scientific paper has been demonstrated, let us address the mechanics.

Each journal has a specific format in which manuscripts must be submitted. It varies somewhat from one journal to another, but the manuscript will not be accepted for review unless all the requirements are met. To find these, pick up an issue of the journal to which you wish to submit a manuscript and look for the page entitled "Instructions to Authors." This page may be printed in alternate issues, so look in another if it is not in the one at hand. Follow these instructions to the letter.

If you are not fluent in idiomatic English and you submit to a journal published in English, have your manuscript read and corrected by a physician or basic scientist who is fluent in the language. An editor may decide that a manuscript is not ready to be sent out for review because the English is such that the meaning of the author is not clear.

Finally, a word about your response to your critics (critic: Greek krites, a judge, discerner). Reviewers are, first, judges (yes/no), and second, helpers. My experience is that a new manuscript has a greater chance of acquiring the latter role than the former. That means your article will be accepted if appropriately revised. Your response to advice for revision should be to consider each point raised very carefully, revise according to the reviewers’ comments, and, in response to points with which you cannot agree, defend adequately the position you took (the reviewer may then agree with you). Ignoring criticism almost always will result in further delay (rereview) or outright rejection.

Godspeed, and remember that our journals are not only for academicians but for all those who have a message.

Brian F. McCabe, MD, Editor

Slide 1: Title Slide

  • Title: How to Write a Technical/Scientific Paper
  • Subtitle: Key Guidelines for Effective Writing
  • Presenter: [Your Name]
  • Date: [Presentation Date]

Slide 2: Purpose of Writing a Paper

  • Objective: Share findings with a broader audience.
  • Key Point: Technical papers are written to inform and contribute to knowledge in a field.
  • Importance: Reaching thousands through journal publications and enhancing professional growth

Slide 3: Essential Sections of a Paper

  • Technical Paper Structure:
    • Title: Draws attention to your work.
    • Abstract: A concise summary of objectives, methods, and findings.
    • Introduction: Sets the context and defines the problem
  • Scientific Paper Structure:
    • Materials and Methods: Details for replication.
    • Results and Discussion: Core findings and their implications

Slide 4: Writing Style and Guidelines

  • Tips:
    • Use passive voice and avoid subjective language.
    • Keep titles descriptive but concise.
    • Use precise figures that reflect measurement accuracy

Slide 5: Abstract and Introduction

  • Abstract:
    • Keep it brief (150-200 words).
    • Focus on what was done, how, and the results.
  • Introduction:
    • Explain the background and the problem.
    • State the purpose and why it matters

Slide 6: Results and Discussion

  • Results: Present findings with relevant figures and trends.
  • Discussion: Compare theoretical and experimental results, explaining discrepancies.
  • Focus: Use clear visuals and ensure consistency

Slide 7: Conclusion and Recommendations

  • Summarize: Key findings and their implications.
  • Pronounce Judgment: State what the study accomplished and future directions.
  • Next Steps: Highlight further research opportunities
Last Updated: