Writing (mit)

MASSACHUSETTS INSTITUTE OF TECHNOLOGY DEPARTMENT OF MECHANICAL ENGINEERING REPORT WRITING FOR 2.671

For our purposes, reports may be divided into three general categories Short Technical Memos: Reports of this type are little more than a summary of what was done and what was concluded. A short memo typically runs only one to two pages and is directed at someone who wants only the highlights without the details. A short memo is sometimes called an Executive Summary because top executives generally want only the salient conclusions of your work. Technical Memos: A memo of this type typically runs five to ten pages. It is intended to convey your work in fuller form, so that the reader knows why the study was performed, how it was done, what the results were and what was concluded. However, a technical memo does not supply fully the details of procedure and theory. Its focus is on the results and their implications. Full Technical Report: A technical report includes a full description of all aspects of the experiments. Theory and procedure are described well enough that another professional would understand how to repeat the measurements or calculations, even though step-by-step details are not given. A technical report is typically ten pages or longer.

ELEMENTS OF A SHORT TECHNICAL MEMO Generally, these executive summaries include only ADDRESS: To: From: and Subject:, stated clearly. These form a heading. SUMMARY: State the objectives clearly, state the procedure briefly, and summarize the results concisely. Avoid any equations, sketches of equipment, etc. You may attach one or tow figures showing your results, if needed. This section should be one to tot pages at most. RECOMMENDATIONS OR CONCLUSIONS: If it is needed, include a final short section of one to two paragraphs stating your conclusions or recommendations as concisely as possible.

ELEMENTS OF A TECHNICAL MEMO A typical technical memo might contain the following sections ADDRESS: To: From: and Subject: State this information clearly and simply. The address section forms the primary heading for your report.

1

SUMMARY Begin with a one to two paragraph (1/2 page) summary of what was done and in particular, of what was concluded. This summary should describe the question that the memo addresses and the answer to that question. The summary should not be as concise as the abstract which proceeds a technical report. INTRODUCTION OR PROCEDURE: Describe the problem, the approach taken, theoretical details, and/or the experimental method. Theory and procedures should be described only briefly, superficially if necessary. Include sketches of equipment, if needed. This section will typically be one to two pages long. RESULTS: Describe the results of the experiment in as much detail as needed for clarity. Don’t include raw data, data reductions, or uncertainty calculations here – such details should go into appendices. Discuss major error sources, the agreement of theory with experiment, and/or important curve fits. Include any important graphs here as well. The length of the results section varies, but is typically one to two pages plus figures. RECOMMENDATIONS: When needed, close with a short section (one to two paragraphs) which states your recommendations or conclusions clearly and concisely. APPENDICES: Appendices are not dumping grounds for garbage data or anything else your reader would never want to look at. An appendix contains information that an interested reader might need but which would interrupt the flow of the main text. Use appendices for derivations, for long tables of data, and for background information that readers do not need to understand your major points.

ELEMENTS OF A TECHNICAL REPORT A technical report includes more sections than a memo. The sections each serve a single specific purpose. The first section is an ABSTRACT, and is immensely helpful to your reader. It stands alone on a separate page and must be understandable without reference to outside graphs, tables, etc. One paragraph only. State problem, methods, results. Be precise and quantitative about your results, but don’t introduce symbols or equations. The rational or justification for the project should go in the INTRODUCTION. The introduction should have a brief discussion of the “real world” dimensions of your problem— the actual motivation behind your project. Then it should present the main technical issues involved in your project. Finally, the introduction should end with a general description of the report that follows— a sort of “road map” to orient the reader. Each of these parts can be a separate paragraph. The THEORY or analytical model should be discussed clearly and specifically. Explain theoretical sources of error. If you refer to other work, be sure to cite it (references) correctly (see sections on “general principles” and “mechanics”). Also make sure your number equations clearly. Basic equations go in the text; derivations go in an appendix. If you use an appendix, refer to it in the text so your reader knows it is there. Use separate appendices for separate

2

purposes (appendix A, appendix B, etc.). Note: Theory section may also follow the results section depending upon the type of report your are writing. If in doubt, consult with your instructor. DESIGN and APPARATUS may be condensed into one section of the report. Although you want your reader to be able to reproduce what you have done, this does not mean going into every minute detail. Give the pertinent details. Give the overall picture before you start describing subsystems. When you describe measurement systems, explain sources of experimental error. Software should be described briefly, so the reader understands the basic algorithm. However, you should not give the program or code. EXPERIMENTAL PROCEDURES are presented in chronological order (apparatus is usually described in spatial order). Again, give an overview before going into the details. Do not attempt to write a set of instructions for the reader. Don’t use the imperative mood of the verb. RESULTS should follow procedure. Here you will undoubtedly refer the reader to various figures. Don’t throw raw data at the reader. Data tables showing raw numerical results can go in an appendix. Figures that are part of the text may show data points; you can draw a curve through tem, but explain to your reader what you are doing. Use error bars where needed (you may need different error bars for different parts of a graph). DISCUSSION of results will compare theoretical and experimental results, as well as important problems encountered in the project. CONCLUSIONS and RECOMMENDATIONS for further work may be condensed into one section. Here you summarize what you have learned and explain the implications of your work for the real world problem. APPENDICES are not dumping grounds for garbage data or anything else your reader would never want to look at. An appendix contains information that an interested reader might need, but which would interrupt the flow of the main text. Use them for derivations, for data tables, and for background information that readers do not need to understand your major points. Appendices my also go at the end of a technical report. Figures go in the text and should be placed as close as possible to the point where they are mentioned. It is probably easiest to use a separate page for a figure rather than trying to fit it onto a page of written text. Make sure that figure pages have adequate margins (at least one-anda-quarter inch all around), including captions and axis labels. Make sure the figure has a complete and specific title to go with its number. A title like “Case One” is useless. The title (which might be as long as a brief paragraph) should make the figure understandable by itself; the reader should not have to refer to the text to get the general point. For more information on figures, see the sections on “General Principles”, “Mechanics” and “Graphs” below. Use appropriate scales and make the choice consciously: Linear, log-log, or semi-log.

3

For a long report a table of contents with headings and sub-headings will again be an immense help to your reader. Be sure to use plenty of both. Use one of the standard numbering systems for your headings and sub-headings (1, 1.1, 1.2; 2, 2.1, 2.2 etc.) Your title should be chosen carefully. Make it as complete and specific as possible, but avoid unnecessary words.

TECHNICAL-REPORT-FORMAT SUMMARY TITLE PAGE. Title must be informative, contain key words and reveal the topic of the report; page includes title, author, project supervisor, group members, place, data. ABSTRACT (about 150 words). A brief statement of the point of the study, what was done, how it was done, and results. Do not include general background here. TABLE OF CONTENTS. If report is ten pages or more, list all headings and subheadings with their page numbers. Don’t list the table of contents itself. INTRODUCTION. State context and background of study, state of the art, why the study needs to be done, what the problem is, purpose of present investigation, outline or overall approach. THEORETICAL ANALYSIS. Develop governing equations and state assumptions. Detailed algebra goes in appendix. A separate theory section is placed after introduction if the analytic model is the main object of the report. If the theory is of minor significance, it may be absorbed into one of the other sections as appropriate. EXPERIMENTAL PROCEDURE. Apparatus: describe, usually with an illustration and words to explain the illustration. Methods: describe what was done so another professional could duplicate the process. RESULTS. Present results, usually with a graph(s). Explain the graphs in prose. Detailed data go in Appendix. Estimate uncertainty and describe major sources of error. DISCUSSION. Compare results with theoretical expectations. Account for anything unexpected in results. CONCLUSIONS. Summarize the results in light of the original problem, assess the study in terms of the purpose stated in the introduction. RECOMMENDATIONS. If applicable, present recommendations for further work. APPENDICES. Give details of data reduction, uncertainty analysis, or other calculations here, if they are needed in the report at all.

4

REFERENCES. If needed, for unfamiliar empirical constants, theoretical analyses whose results are included, detailed descriptions of equipment and procedure.

GENERAL ISSUES IN REPORT WRITING In deciding what to include and what goes where, you will have many judgement calls. There are no cut-and-dried formulas for making these decisions: you have to think carefully about the purpose of your report and who is reading it. By all means ask your technical instructors for advice if you are not sure how to proceed. Notes on the Writing Process. In a real sense you are writing your final report from the first day of the lab. As you are doing your experimental work, you should be thinking about how the final report will look— its major sections, its major points, what goes in and what doesn’t. Before you actually start writing, summarize your ideas in a writing plan, which outlines (however roughly) your key points, key ideas, key figures. Have your technical instructor review that writing plan briefly if you have any major questions about it. Set aside a couple of hours spread over a number of days to generate a rough draft. Don’t censor yourself as you write: just keep writing. Most people find it easiest to begin with the apparatus and procedure sections, and then the results. The abstract and introduction and conclusion are often written last, after you have a better sense of what you’re trying to say in the body of the report. Others wish to write the introduction first as a guide to their analysis and experiments. Do NOT turn in this draft, unless you want to court disaster. Edit if first for organization and content, then for style and mechanics (simple language, clearly stated transitions, reasonable paragraph structure, grammar, punctuation). If you have any trouble editing yourself, take your report to the Writing Center (14N-317; calling for appointment first, at x3-3090, can save time). The Writing Center can also help with earlier stages of report writing. GENERAL PRINCIPLES 1. Organization and clarity of reasoning are the most important aspects of any report. 2. Present the whole before the parts, in both paragraphs and whole sections of the report. Describe the essence, function, purpose of a device, process or concept before going into details about the parts. 3. Summarize at the end of each section which is long and involves several approaches or procedures. 4. Emphasize important information by putting it first and eliminating superfluous details and words.

5

5. Use headings, spacing and visual aids fully. 6. Without weakening technical content, use as much plain, simple English as possible. MECHANICS 1. Use headings and sub-headings. Double space before headings. 2. Number and give an informative title to all figures (graphs, drawings, schematic diagrams, block diagrams, photos). Place the figure number and title at the bottom of the illustration. Refer to the figure in the text (see Figure 1), and place the figure on the page following the reference to it. If the figure fits lengthwise on the page, turn it counter-clockwise, so the top of the figure is at the binding side of the paper. 3. Tables have the same requirements, except that their numbers and titles occur at the top of the table. 4. Appendices must be titled and referred to in the text, not just listed in the table of contents. 5. Number equations in parentheses on the right side of the page. When referring in the text to the equation, do not use the parentheses. Put the explanations of the symbols in a list following the equation, if there are more than three definitions. GRAPHS 1. Label both axes clearly. 2. Choose linear, semi-log, or log-log coordinates as appropriate. Do not plot the logarithm of a variable on linear coordinates 0 use real logarithmic axes. 3. Choose numerical scales to include significant minimum and maximum values such as zero, one, or 100%. Tic marks may be placed at intervals of 1, 2, or 5 digits or linear axes. On logarithmic axes, tics are generally placed at … , 1, 2, 5, 10, 20, 50, … etc. Not every tic mark need have a numerical label, but major intervals should definitely be labeled. 4. When plotting data points, choose an appropriate symbol: circle, triangle, square, half-darkened square, etc. Do not plot points as dots (.). The uncertainty of a measurement should be represented using error bars or by choosing a symbol whose size is commensurate with the range of uncertainty. 5. Curves based on a theoretical or least-squares equation should be drawn as lines without showing the calculated points. Measured data may, of course, be shown on the same graph with theory. Curves being fitted to experimental data do not need to

6

pass through each data point (like a dot-to-dot cartoon). However, when a fitted curve lies outside the uncertainty range of a measurement, a discrepancy is indicated. 6. Do not use colors; instead use shapes to indicate differences. 7. Leave good margins (1.25 inch on binding edge; 1-1.25 in. elsewhere). 8. Make the title informative, something more than Height vs. Time, for example. REFERENCES References should be listed at the end of the paper and referred to by number in the text. For example: 1. Bridgman, P.W., Studies in Large Plastic Flow and Fracture, New York: McGrawHill, 1952. 2. Eirich, F.R., ed., Rheology: Theory and Applications, Vol. I, New York: Academic Press, 1956. 3. Sperry, R.W. “The Eye and the Brain,” Scientific American, (November 1959): 4852. 4. Zander, John D., “Computer Surveillance of Medicaid Claims in Tennessee,” Master’s Thesis, Vanderbilt University, 1973.

7