WRITING A SCIENTIFIC REPORT
by Jon Wong,
Department of Environmental Toxicology, University of California-Davis.
The material and information provided in this handout are excerpts
from the textbook How to Write and
Publish a Scientific Paper, 2nd ed.,
Robert A. Day, ISI Press, 1983, Philadelphia, PA. and from the
article by Berry, R. E.; Noyes, O. R. How to get your paper published
in Journal of Food Science, Journal of Food Science, 1993,
58 (5), 1179-1184.
Abstract
The abstract states the principal objectives and scope
of the investigation, describes the methodology employed, and
summarizes the results and principal conclusions. References
to the literature must not be cited in the abstract (except in
rare instances, such as modification of a previous published method).
Forget about all the experimental detail, omit all the references
from previous works, and avoid all the lengthy exposition of your
detailed knowledge of the specific problems investigated. Try
to limit yourself to a short description of the problem and its
solution. A well-written abstract allows the reader to identify
the basic content of the document quickly and accurately, to determine
its relevance to his/her interests, and decide whether he/she
need to read the document in its entirety. To write an effective
abstract, use brief statements of the nature of the problem, state
your hypothesis, give a general approach, and present your results
and conclusions. Limit your abstract to about half a page of
text or about 150 words.
Introduction
The purpose of the introduction is to supply sufficient background
information to allow the reader to understand and evaluate the
results of the present study without needing to refer to previous
publications on the topic. The introduction provides the rationale
for the present study. It should briefly and clearly state your
purpose in writing the paper. References must be carefully chosen
to provide the most important background information. The introduction
should contain (i) short paragraphs that describe the nature and
scope of the problem investigated, (ii) review the pertinent
literature to orient the reader, and (iii) state the method of
the investigation or reasons for the choice of a particular method.
For the introduction, give short paragraphs about the importance
and nature of the problem, briefly state what's been done before,
how this approach differs, remaining problems, your hypothesis,
and objective of the current study.
Materials and Methods
The main purpose of the Materials and Methods section is
to present enough detail that a competent worker can repeat the
experiments. Careful writing of this section is critically important
because the cornerstone of the scientific method requires that
your results be reproducible. The potential for producing the
same or similar results must exist, or your paper does not represent
good science. If there is serious doubt that your experiments
could be repeated, the reviewer will recommend rejection of your
paper, no matter how awe-inspiring your results.
For the materials, include the exact technical specifications
and quantities and source or method of preparation. It may be
necessary to list pertinent chemical and physical properties of
reagents or chemicals used. Avoid using trade names, use the
generic or chemical names instead. Because the value of your
paper and your reputation can be damaged if your results are not
reproducible, you must describe research materials with great
care.
For methods, the usual order of presentation is chronological.
For measurements, be precise as possible. Questions such as
"how" and "how much" should be precisely answered
by the author and not left for the reader to puzzle over. If
your experimental method is new, you must provide all of the needed
detail. However, if the method has been previously published,
only the literature reference should be given. If alternative
methods are employed, identify the methods or alterations, as
well as cite the reference.
Summing things up, state what you did and how you did it in a
way that a reader can repeat the study. Do not include discussion,
rationale, excuses, etc. Describe methods used for statistical
analysis. Do not include in the materials and methods why you
conducted the study, how it turned out, how you interpreted the
results, or the rationale. All of these should be in the results
and discussion section. The only purpose for the Materials
and Methods is to provide enough information so that the experiments
could be reproduced by a competent colleague.
Results and Discussion
Most authors choose to write the results and discussion together
in a single section. The choice is up to the writer (or the instructions
of the instructor or journal) whether the separation of the results
and discussion into two distinct section is necessary and appropriate.
The Results should contain two goals. First, you should give
some kind of overall description of the experiments, providing
the "big picture" without, however, repeating the experimental
details previously provided in Materials and Methods. Second,
you should present the data. The presentation of the data is
important because you must and unambiguously illustrate your results
before you can effectively discuss and present your conclusions.
As a general rule, present the results that best clarify
the important points of your discussion. If you made only a few
determinations of your results, give the results in the text.
Also, presentation of raw data such as peak areas from a chromatograph
or optical density readings from a spectrophotometer are not results.
Raw data should be presented in an appendix. If your results
are calculated from a mathematical expression, sample calculations
are presented in the appendix but the mathematical expression
and the resulting calculations are presented in the Results and
Discussion.
The Discussion is usually the most difficult part to write.
The components of a good Discussion usually contain these elements:
1. Presentation of the principles, relationships and generalizations
shown by the Results.
2. Point out any exceptions or lack of correlation and define
any unsettled points. Try to explain why those specific or unsettled
points do not correlate, and suggest what you can do to avoid
these lack of correlations if experimental errors were involved.
3. Show how your results and interpretations agree or contrast
with previously published work. Try to present reasonable explanations
why your results do not agree with comparable results by other
researchers.
4. Discuss the theoretical implications of your work as well
as practical applications. This provides the basis on why your
work is important and why your investigations are important in
context with other people's research.
5. State your conclusions as clearly as possible and summarize
your evidence for each conclusion.
In summary, what did you find? How do results relate to
those reported previously? Was the hypothesis correct? Does
it need to be changed and what are the appropriate changes? Are
all tables, graphs, and other illustrations absolutely necessary
to understand the study? Is information in the tables, graphs,
and text redundant or repetitive? Are statistical differences/significances
clear?
Conclusion
For the conclusion, did you answer the question, "So
what?" What recommendations/advice to the reader can you
make on the basis of your findings? Do you need to adapt your
hypothesis or extend it?
References and Citation
References must substantiate, relate to, or confirm the assumptions
in your study. Be thoroughly familiar with a source before using
it as a reference. There are standard abbreviations for most
journals, but if you don't know it, spell them out and be consistent
when abbreviating them. Usually, the standard format for a reference
is author's last name, first and middle initials, title of the
paper, journal of where the paper was published (in italics),
year, volume (and number if necessary), page number (beginning
to end). For example,
Nondek, L., D.R. Rodler, and J.W. Burks.
1992. Measurement of Sub-ppbv Concentrations of Aldehydes
in a Forest Atmosphere Using a New HPLC Technique, Environ.Sci.
and Technol. 26:1174-1178.
There are usually three ways to cite a paper: name and year;
alphabet-number; and citation order. The choice is usually determined
by the journal. The name and year system is simple such as "Nondek,
Rodler, and Burks (1992)" or "Nondek et al. (1992)".
The alphabet-number system is a citation by number from an alphabetized
list of references. The citation-order system is a system of
citing the references by number in the order that they appear
in the paper.
How to Design Effective Tables and Figures
An effective way to present repetitive results is to use
Tables and Figures. An effective table allows the reader to grasp
the information clear enough to make the meaning of the results
understandable without reference to a text. Tables should have
relevant data and variables but not experimental conditions in
the column. However, if an experimental condition (such as varying
the temperature) is a variable in the experiment, it can have
its own column. If experiments were performed at the same temperature,
this information should be presented in the Materials and Methods
or presented in the heading or footnote of the table. Table 1
is a good example of a complete table. The table lists the data
that the researchers performed on red blood cells for a gene defect
that causes a mutation in an enzyme, superoxide dismutase (SOD).
The various mutants (columns 1 and 4), the number of subjects
(column 2), and the enzyme activity (column 3) are clearly shown.
Appropriate units of enzyme activity and statistical information
are shown as footnotes. The control and the mutant enzyme activities
are compared and therefore, the researchers have effectively shown
that SOD activity is lower in the mutants than in the controls.
This table contains no repetitive data nor experimental conditions
in the columns.
A table may be useful in presenting operating conditions in the
Materials and Methods section for an instrument as illustrated
in Table 2. This is effective because it is clearer to read
from a table than from the text.
A table may be useful for presenting properties for chemicals
or reagents used in your experiment as shown in Table 3.
Figures are used when clarity and conciseness are required and
because certain types of data are more effective if they are presented
in figures, such as graphs. For example, Figure 1 is an effective
figure for the Results and Discussion section because the authors
want to compare a field blank and an ambient air sample to convince
the reader that any formaldehyde and acetaldehyde present in the
air sample did not originate from the reagents or solvents used
for the analysis.
Figure 2 is effective for the Materials and Methods section because
the authors want to describe a new gas chromatography set-up by
a visual diagram rather than describe the intricate details in
narrative form in the text.
Figure 3 is effective because the authors want to propose a model
based on their experimental results to understand the biogeochemical
processes of chromium redox chemistry in flooded forest wetlands.
Figures 4 and 5 are effective figures for the Materials and Methods
section because flow diagrams of the two sample preparations for
the analysis of PCDDs and PCDFs are easy to follow.
Certain types of data are more effective if they are presented
in figures, specifically graphs. Where possible, express data
in narrative form in the text and minimize the use of figures
and tables. However, if the data show pronounced trends, a graph
may be more effective in presenting your results than a table.
Figure 6 is an analytical calibration curve for cobalt and arsenic
that was developed from helium inductively coupled plasma-mass
spectrometry. Note how the axis and the figure captions are clearly
and briefly stated. Make sure you properly label the ordinate
and abscissa and that the ranges are appropriate for presentation.
Finally, Figure 7 is an effective figure because the graph shows
trends that are more obvious by visualizing rather than reading
the text. Experimental data points as well as a theoretical transformation
rates are well-defined and well-represented. Error bars are also
revealed and all headings are appropriately labeled. Graphs,
figures and tables must be neatly drawn. All figures must have
the proper captions, headings, and a brief description of what
information the figure is suppose to convey. Remember to always
label your figures and tables, numerically (i.e., Table 1 or Figure
1) and always refer to the table or figure in the text of your
paper, i.e., "The results of the transformation rates for
ONP, 4-Me, and 4-Cl are shown in Figure 7."
