M465:sciencewriting

Guidelines for Scientific Writing

Most scientific work is published in a specific format called scientific style. Although there are many small differences among science publications, a standard format has evolved so that readers know where to find specific categories of information (i.e., hypotheses, background and pertinent literature, methods, implications of results). Use of the standard format is a very important part of scientific writing. Arbitrary modifications of the format by "creative" writers can only confuse readers and lessen the paper’s clarity and impact. The best way to learn this type of writing is to read good examples of scientific writing in journals and to practice. We expect that the Results Summaries you turn in will include a short title, a methods section that briefly covers the experimental protocols that generated the results and then a results section including any figures or graphs. Below we give you some guidelines and examples for these three sections.

The Title

Compose a title that reflects the major conclusion of the paper. Remember that “the story is about who shows up first”, therefore, you should stress the answer to the experimental question over the experimental tools. Generally, the title should be less than ten words.
Compare these three examples:
"Response of patients to different doses of ice cream"
"Frequent administration of ice cream boosts patients' morale"
“Morale in hospital patients is improved by serving more ice cream”

The first example gives little information about the findings. The second is unclear: we don’t know how the ice cream was used (“applied” doesn’t necessarily mean “eaten”) or enough about the patients. It appears, in the second example, that the topic is uses of ice cream rather than about how to boost patient morale. Although the third example is more than ten words (11) and it is in passive rather than active voice, its advantages are that it’s clear that the main topic is hospital patient moral and that the variable showing positive correlation is eating ice cream. The third title, therefore, is preferred.

The Methods

Students commonly have trouble with the methods section, primarily because it is mistakenly considered to be the place to describe exactly what you did in lab, step by step. Instead, think of this section as the place to present the progress of the experiment through a detailed but brief description of the methods used to collect and to analyze your data. Present enough information so that the reader can evaluate what you did, see how you achieved your experimental goals, and could, possibly, replicate your work. Do not clutter your paper with trivial or nonessential information. This section should not be a transcription of the lab manual or a chronological description of your lab experience. It should be as succinct as possible.

Divide the Methods section into subheadings with titles that give the goal as well as the main tool used to achieve the goal of each part of the experiment, e.g, “DNA fragmentation and separation by restriction enzyme digestion and agarose gel electrophoresis” is a better subheading than “Gel Electrophoresis” or “Identifying the Gene”.

Methods are always presented in the past tense because the experiments are complete. Third person is preferred, even if it requires passive voice, because avoiding first person helps you focus on the progress of the experiment rather than on your lab day (“DNA was extracted in 70% ethanol” is preferred over, “We added 1ml of 95% ethanol to a tube containing…”).

Sample partial Methods section including subheadings:

Restriction Enzyme Digestion and separation of plasmid fragments by electrophoresis
Plasmid DNA was digested with a restriction enzyme by combining 5ng plasmid pHM64 (Table1) and 5ng plasmid (Table 1) with 1µl HindIII ( SigmalAdrich www.sigmaaldrich.com) for 20 min at 37C in High Salt Buffer (0.1M NaCl, 50 mM Tris pH 7.5, 10mM MgCl2, and 1mM DTT). The DNA fragments from the digestion were separated by agarose gel electrophoresis by applying all of the digested DNA to a 1.5% agarose gel in TGE (0.25M Tris, 1.9M Glycine, 13mM EDTA) with 0.5% SYBR safe DNA stain from Invitrogen www.invitrogen.com. DNA fragments were subjected to electrophoresis at 100V for one hour. Stained and separated DNA fragments were visualized and photographed under UV light.

If you use mathematical formulas or statistical analyses, include an explanation of those calculations or statistical tools. Explain the variables that were compared, the types of statistical tests used, and for what they were used. If you calculate a p value, include the threshold significance probability—often 0.05.

Results

The results section should begin by framing the overall investigation and its goals. The results section then leads your reader through your experiments in a narrative that describes how the data lead to the conclusions of the paper. This section should NOT consist only of figures and tables. Each paragraph should begin with a topic sentence that frames a particular experiment, addressing the biology you are exploring in it. The narrative portion is always written in the past tense because the experiments you describe are complete. There should be proper transitions between paragraphs and varied sentence structure as you describe the experiments (without excessive methods details). There must be a specific reference to every figure or table included as you emphasize the main findings of each experiment. A good template to use for the overall structure of this section is, “Why, How, Where, What, So What?”. In the example below we have indicated where in the text these questions are addressed.

""Example Results Section""
To determine whether or not cheek epithelial cells contained a nucleus (WHY?), they were stained with methyl green, which binds to acidic molecules such as DNA (WHAT?). As shown in Figure 1 (WHERE?), one round, dense, darker-staining structure of diameter 5-10mm was seen in each cell (WHAT?), suggesting that these cells do indeed have nuclei (SO WHAT?).

Often you will refer to a table/figure in the narrative part of Results by including it in parentheses after a statement that describes its contents. Such as:
Length of activity periods was inversely proportional to temperature (Table 1).
Parents of both species fed their chicks between 09:00 and 14:00 (Fig. 1).

Tables and Figures
The narrative text of Results describes your findings while the data that documents those findings is processed and presented in tables and figures. An effective table or figure should stand alone, meaning that the reader should be able to understand, generally, the experiment and the data’s meaning without reading either the Methods section or the narrative description. Unprocessed, raw data are usually not included in the body of the paper. Instead, process and present the data in a way that makes your main findings readily interpretable by someone who was not in lab. Number figures and tables separately and consecutively in the order to which you refer to them in the text.

Because it should be possible to understand the main point made by each figure or table’s data without reading the results narrative, table headings and figure captions must include all of the information needed to interpret the data presented; however you should not include in figure legends or table headings a discussion of your findings. In addition, make sure you reduce substantially (compared to the level of detail found in your methods section) the description of the data collection.

Figures:
Any graph, map, photograph, or drawing is a figure. Label the axes carefully with titles and specify the units.

Figure 1 Protein calibration curve. A range of concentrations of bovine serum albumin (BSA) were reacted with biuret reagent (alkaline copper sulfate). The resulting change to purple color was measured as Absorbance in a spectrophotometer at a wavelength of 540 nm. A540nm values were plotted against known BSA concentrations as a regression line using Microsoft Excel in order to create a standard curve for converting A540nm measurements of unknowns to protein concentrations.

Tables:
A table is any data presented in tabular form. Each column of data should have a heading that describes completely what is in that column, including the units of measurement. Some columns may need a double heading. For example, when reporting data from different species, you should specify which column of data was collected from each species. It will also be necessary to give an additional heading over all of the species’ data columns indicating what type of data are in these columns. Enter data values so that decimal points are vertically aligned, and align and space all entries for maximum clarity. Note that Table headings are positioned above the tabulated data while figure captions are found below the figure. Footnotes are used when symbols or abbreviations require explanation.

Table example:
Table 1. Means ± SE for a test of the hypothesis that food type affects flour beetle (Tribolium castaneum) population growth in a 70-d period. ANOVA, F = 22.7, df = 2, 6, p=0.002, &#945= 0.05. Tukey levels¹ not sharing the same letter are significantly different.

Type of flour	Number of Beetles Mean ± SE	Tukey Level
Corn	196 ± 13.3	B
Wheat	385 ± 17.3	A
White	132 ± 42.1	B

¹ Tukey HSD test

Writing Tips

1. Be sure to divide the text into appropriate paragraphs. Each paragraph must begin with a topic sentence that gives the paragraph’s context (connects backwards to the previous paragraph) and a concluding sentence that describes the main point of the paragraph. Remember that the end of a paragraph or a sentence is the stress position. Don’t confuse your reader by putting the most important information in a non-stress position or unimportant information in the stress position. Information should move, generally, from old to new.
2. You should, largely, write about your study in past tense; however, it is common to write about generally accepted scientific theory worked out by others in present tense. Active voice is generally more interesting than passive; although, in science writing, passive voice is sometimes preferred (particularly in the methods section) because it seems more objective and it appropriately changes the focus from the experimenter to the experiment.
3. Remember your goal is to convey information; you must write so that you cannot be misunderstood. Readers have structural expectations of a paper in scientific format and it confuses them when those expectations aren’t met.
4. Eliminate redundancy. Decide where a detailed explanation of a concept would be most helpful to your reader; limit yourself to a brief summary when mentioning that concept elsewhere in the paper.
5. Prepositions should have objects: Wrong: Samples A-F are shown in the plots they were taken from; Right: Samples A-F are shown in the plots from which they were taken.
6. The word 'data' is plural; 'datum' is singular. Use the verb with proper agreement: “data are shown” as…
7. Be sure that pronouns refer clearly to their appropriate antecedent. Example: "Sometimes nests were built in bushes, and they were often hard to find." Does 'they' refer to the nests or to the bushes?
8. Avoid anthropomorphism (giving human characteristics to non-human subjects.) Examples:
   1. A plant chooses to grow in a wet area.
   2. Honey bees have a visual appreciation for color
   3. Honey bees try to optimize costs.
   4. Wind discouraged pollinators.
9. 'Species' is both singular and plural (a 'specie' is a coin).
10. Do not use slang, colloquial language, or idiomatic phrases. “Ran a gel” is colloquial.
11. Contractions should be avoided.
12. Avoid use of qualitative and often meaningless adverbs such as almost, largely, very, most, strongly, greater, etc. Replace such terms with quantitative evidence or an example that supplies context. Example: Which is a more persuasive statement?
    1. Facebook is a wildly popular online social networking tool...
    2. Facebook is the Internet site most frequently viewed by males and females, aged 17-25, in the United States today (eMarketer, 2007). Sixty-nine percent of females and 56% of males in that age group have Facebook accounts; nearly 65% of users log on to Facebook once a day.
13. General statements like “results indicate” or “data show” are too ambiguous. Exactly which measurement or comparison or statistic leads to a conclusion?
14. Writing concisely means eliminating unnecessary words. Often, even a 'the' is unnecessary. Wordy: The areas without trees had a greater density of the defoliated plants. More Concise: Areas without trees had a greater density of defoliated plants.
    
    

Clarity in writing is a fundamental goal. Scientific writing should not be flowery or use vocabulary that is unintelligible to its target audience. Explain terms and concepts that your audience isn’t likely to know or use simpler language, as long as it’s accurate. Use the correct word. Although proteins and genes are related, they are not the same thing. Write not only to be understood, but so that you cannot possibly be misunderstood.

There are a myriad of small and large concepts to remember when structuring a paper in scientific format, especially if this style of writing is new to you. Each time you write or revise a paper you should review these guidelines. Before submitting a paper, reread it critically several times for clarity, conciseness, and grammatical errors.

An excellent discussion of structure and clarity in science writing is found in an article, “The Science of Scientific Writing” by George D. Gopen and Judith A. Swan in the American Scientist (Nov.-Dec. 1999), Vol. 78, 550-558.

Often you can answer your own questions by looking at models in published journal articles.