SUGGESTIONS FOR SCIENTIFIC WRITING

Writing is something that we are all expected to do, and many of us take this task lightly.
In fact, few are good writers, and all of us would benefit from improved writing skills.
This guide is intended to help students to prepare term papers and similar reports. Some of
the suggestions it contains reflect my own preferences. Some of my suggestions and
preferences differ from those of other instructors, but none will get you into trouble.
1. Purpose
The goal of scientific writing is accurate and efficient communication of information. The
volume of scientific and technical documents being produced today is immense, and most
of us need to read much more than is possible. A scientific document should be written in
such a way that the reader can easily find that portion of the document in which he or she is
most interested, assimilate the information, and get on to other work. Amusement is not a
goal of scientific writing.
2. Organization
Good organization of a paper is necessary to ensure that at all times the reader understands
the purpose of the sentences and paragraphs as they are read. There should be no surprises.
There is no universally accepted format for scientific papers, but some basic rules need to
be followed. For example, the standard scientific format "introduction, hypotheses,
methods, results, conclusions" is fine in some circumstances but not so good for others. The
following rules will help:
a. Every paper needs an introduction, and sometimes it is helpful to introduce major
sections of documents. An introduction may consist of a literature review, a theoretical
background statement, a discussion of the practical importance of a body of research, or a
simple statement of purpose. It may include a brief statement of the major conclusion of
the paper. Introductions to sections that are a few paragraphs long may consist of just a
single general sentence about the topic, but usually some sort of an introduction is helpful.
For the same reasons, sections longer than several paragraphs should be labeled with
headings or subheadings. They let the reader know what you are doing, where you are
going, and make it easy to find things.
b. Good organization extends down to the paragraph level. Each paragraph should
contain a single idea or topic that can be expressed as an outline heading. It is helpful to
create a detailed outline of the paper before or while you write. Once you have created the
outline, stick to it. Outlining functions in word processors are useful in this regard.
c. Try to order your material in such a way that you build an argument in a logical
sequence. Don't make a statement that is not obviously true and then justify it two
paragraphs later. Instead, start with what is known, add new information, and then state
conclusions.
d. Make sure that everything in your paper has a purpose and is not just there as filler.
For example, many papers in geography include as background information a description of
the area studied. This is fine, but if you are writing about income distribution, precipitation
distribution may not be relevant. A paper should have a few central points to make, and
everything in the paper should be important to those points. Minimize discussions of things
that you might think are interesting or important but that are not closely connected to your
central point. When you have finished writing and reach your final conclusions, go back
through your paper and cut out everything that is rot relevant.
e. A conclusions section may include a summary, discussion of practical applications,
theoretical implications, or future research needs. Not all of these are necessary, of course.
The important thing is that somewhere near the end of the paper you state the significant
outcomes-- the essential meaning-- of your work. Again, there should be no surprises. The
reader should already know what the main points are; you are merely underlining them.
3. Style
Rules regarding writing style in scientific papers should be followed nearly all of the time.
They help make ideas clear and unambiguous, if not producing great advances in literature.
If all these rules are followed 100% of the time the result will be accurate and efficient, if
monotonous. Vary sentence structure and language a little bit, but not too much.
a. Use subheadings to reinforce the organization of a paper and to make it easy to
locate topics of interest within the paper.
b. Each paragraph should contain a single, major idea which is summarized in the first
sentence of the paragraph. The remaining sentences of the paragraph are explanations,
justifications, or elaborations of the first sentence (see this paragraph for an example). If
there are two or more major ideas presented, they usually deserve two or more paragraphs.
One-sentence paragraphs should be avoided.
c. Sentences should be structured carefully. The subject of the sentence should be near
the beginning, and the verb should be nearby rather that at the end of the sentence. The
standard structure should be subject, verb, object. Avoid beginning a sentence with
conjunctions (and, but, or, yet) or with dependent clauses.
d. Write in the active sense rather than the passive wherever possible. ("The mayor
decided...." instead of "...was decided by the mayor").
e. Use precise language. Expressions such as "a number of", "very", "etc.", "and so
forth", "researchers have found", and "in the literature" do not convey much information. If
you know the number (exactly or approximately), state it. If someone has written
something that is worth mentioning in your discussion, say who the author was and give a
citation.
f. It is often tempting to use overly formal language in an attempt to sound important
or knowledgeable. This is unnecessary. For example, say "I studied", rather than "the
author investigated".
g. Use present tense when describing things that occur today. Use past tense when
describing things that occurred in the past. "I gathered data on climate. The climate is
humid."
h. Pay attention to simple grammatical rules. For example, "data" is a plural noun;
"affect" is a verb and "effect" is a noun; "hopefully" is an adverb; "site" is a noun; "-wise"
is not an acceptable suffix.
i. It is becoming increasingly fashionable and acceptable to make verbs out of nouns.
Some of these forms have made it into the dictionary and many are in official use, but they
are not the best words to use. "Research" is often used as a verb, but "study" is better ("I
studied the problem", rather than "I researched the problem"). Nouns that are frequently
but incorrectly used as verbs include impact, interface, access, and reference.
4. Citations
Citations serve three major functions. They give credit for ideas and facts where it is due,
tell the reader where to find more information, and allow you to avoid discussing
something that has already been discussed in another paper. Lack of proper citations is
probably the most common failing of undergraduate papers.
a. Whenever you use a major idea, or a specific item of information from someone
else, you must cite the source. Such citations are required for each separate use you make of
the source. If you write a sentence or two based on a particular source, the source is usually
cited at the end of those sentences. If you write a paragraph or more summarizing someone
else's work, begin that section by citing the work you are discussing. If you then move to
other topics but return to that same source later in the paper, cite it again.
b. Avoid using quotes unless the original wording is significant in some way. Your
purpose is presumably different from that of the work you are citing, and you should be
able to express an idea in a way that is most appropriate to your work. If you are discussing
a definition of a term, the requirements of a law, or a key statement by a public official then
quotation may be useful. Otherwise use your own words, and cite the source of the idea.
c. Citations are often used to help the reader find more information on a topic.
Literature reviews contain many such citations. They tell the reader what research has
already been done on a given topic.
d. Sometimes it is useful to demonstrate that you are aware of an aspect of a subject
without going into a lengthy discussion of it. For example, suppose a particular research
technique has been shown to have limitations in certain circumstances, but these limitations
were not important in your study. In this case you might state that "problems encountered
by Jones (1995) were not encountered in this study", or something to that effect. You thus
incorporate a lot of Jones' information without having to write it again.
e. Copying entire sentences from another work without using quotation marks and a
citation constitutes plagiarism. Close paraphrasing of groups of sentences followed by a
citation is plagiarism. Write it yourself!
f. Several citation styles are in use today, but one is most universally used and it
should be followed unless the intended publisher or reader calls for another style. In the
text, use either "Smith (1988) reported...." or "...has been reported (Smith, 1998)." If there
are two authors, name them both in the text, as in "(Smith and Jones, 1999)"; if there are
three or more use "et al" or "and others", as in "(Jones and others, 2003)". If there is more
than one citation for a given author in a single year, use "1997a" and "1997b". List all
references alphabetically in a bibliography at the end of the paper, as follows:
Article:
Author, date. Article title. Journal name (underlined or italicized) volume number: page
numbers. If the journal starts each issue with page 1, include the number or of the
issue.Example:
Smith, A.B., 1998. Distribution of trees in forests. Prestigious Journal 212:41-62.
Book:
Author, publication date. Title (underlined or italicized). Publisher, number of pages.
Example:
Smith, A.B., and Jones, C.D. (eds.), 1999. On Viewing the Forest. Woodland: Spruce and
Co., 246 p.
 
Internet document:
Author, date published if avialable. Web page title (shown on the browser title bar when
you are at the site and printed on the document if you print a page). Organization
sponsoring the site, complete internet address (URL), date site visited. Example:
Columbia University Press, 2002. Columbia Guide to Online Style. Columbia University
Press. http://www.columbia.edu/cu/cup/cgos/idx_basic.html. August 20, 2003.
 
5. Illustrations
Illustrations are very helpful and usually neglected. They communicate and summarize.
For most geographical papers a map is essential. Pay attention to the quality of your
graphics. The reader is usually drawn to illustrations, and thus they take on added
importance.
a. Graphs work better than tables for showing trends and relations. They should be
simple and clearly drawn, and axes should be labeled accurately.
b. Figures (maps, photographs, graphs, charts) usually require titles and captions.
Tables should have titles, and sometimes captions. Titles may either be separate from
captions (at the top of the figure or table) or may be the first parts of captions. Captions
should consist of whatever information is necessary for the reader to properly interpret the
figure or table, such as definitions of terms or symbols. Captions in books sometimes have
a few sentences describing the figure and its broader meaning, but in scientific articles this
information should be included in the text. The caption is also a good place to cite the
source of the information in the figure or table.
c. Tables should be kept small whenever possible, and they should summarize
information. Most tables should contain no more than about 30 cells, although longer
tables are appropriate in some cases. Very long tables, containing hundreds of cells, are
better summarized in shorter tables, relegated to an appendix, or omitted altogether.
d. Never include a figure or table without referring to it in the text. In most cases it is
appropriate to summarize or describe illustrations in the text, rather than to rely on the
reader to look at illustrations and understand them without comment.
e. In term papers and other documents that will not be sold or copyrighted it is
permissible to insert a photocopy of someone else's illustration, as long as a complete
citation is given.
f. Illustrations should be numbered sequentially, "Figure 1" or "Table 1". Each
illustration should be inserted in the text on the same page or the page immediately
following the first reference to it in the text. Do not group them together at the end of the
paper (unless the instructor/editor requires it), as this forces the reader to continually turn
pages, searching for illustrations.
g. Pages should be numbered.
6. Mechanics
Details of production and presentation of a paper are more important than many students
believe. A poor word-processing job can be just as distracting to the reader as bad writing.
a. Footnotes make difficult reading. If the information is important, put it in the text.
If it is not important, leave it out of the paper altogether.
b. Word processors can do all kinds of fancy things with fonts, clip-art, and similar
cosmetics. These are fine for advertising, but usually not helpful for scientific writing.
Stick to a standard 12-point font, and use bold, underlining, and italics sparingly.
c. Do not use fancy binders for term papers; they fall apart and make reading and
marking difficult. It is not necessary to use anything more than a single staple, unless the
manuscript is more than 50 pages or so. Use ordinary white paper. Similarly, don’t bother
with separate cover pages, and certainly don’t clutter the paper with fancy borders and all
the other doo-dads that come with your word-processing software. Stick to simple black
and white; use color only in illustrations and only when necessary.
d. Use the spell-checker on your word processor, but don't rely on it for proofreading
purposes. Remember: both "there" and "their" are in the dictionary. If you have a
grammar-checker on your word processor try using it once or twice. You may find that it
helps you clean up your writing.
7. Revising
In most cases papers are finished very late on the evening before the due date (or in my
case, very early in the morning of the due date). A paper should never be submitted
without first having been critically read by someone other than the writer. Also, proofread
your paper, and correct any spelling, grammatical, or other errors. Nothing makes you look
foolish faster than spelling errors.
8. Further information
The above are just a few summary recommendations. There are many style manuals in
common use, the most common of which is probably Turabian's A Manual for Writers, but
any of them will be helpful. I also recommend Hart's "Ruminations of a dyspeptic ex-
editor" (Prof. Geographer 28:225-232), and the U.S. Geological Survey's Suggestions to
Authors of the Reports of the U.S. Geological Survey. Even if you think you know how to
write, take a look at these guides. You may be surprised to find that things you thought
were correct or acceptable are really not so.