Principles of Writing in the Sciences

Scientific writing has two goals: to inform the reader of new developments in a specific field, and to address existing questions with new evidence.

Scientific research papers report new discoveries, applying evidence to answer questions and identify patterns. Writing in these disciplines often takes the form of peer-reviewed journal articles, literature reviews, grant proposals, case studies, and lab reports.

For example, in an environmental-science lab report, a student might analyze research results to address or clarify a particular scientific development or question:

“This study aims to identify levels of chlorine and phosphorus compounds in a three-mile stretch of the Columbia River, which is an area notable for salmon runs. An analysis of samples taken over a two-year period from various locations within the three-mile stretch revealed the persistence of high levels of phosphorous and chlorine compounds. In the study, we examine the relationship between salmon population and the persistence of these compounds.”

Scientific papers require a great deal of preliminary work, including research, field work, and experimentation. Translating that work into writing can be difficult, but academic conventions provide a common template for communicating findings clearly and effectively.

General Style

Writing in the sciences seeks to explain complex phenomena in clear, straightforward prose that minimizes authorial bias. It also includes elements of classical argument, since scientific papers are expected to contextualize, analyze, and interpret the information at hand.

Precision of Language

Lab reports, case studies, and other types of scientific writing must be precise in order to provide results that can be tested and reproduced.

Strive to use simple words and sentences. Some students try to make their work sound more intellectual by using obscure words and long, elaborate sentences. In reality, the academy values precise words and detailed descriptions that are still understandable to a lay audience. Don’t try to mimic the stereotype of dense, convoluted academic writing. Instead, write as simply and clearly as possible. Precision is a key component of clarity.

In the sciences, precision has two main applications: using concrete examples, and using clear language to describe them. Defining your parameters accurately is essential. Don’t generalize—provide exact times, measurements, quantities, and other relevant data whenever possible. Using precise, straightforward language to describe your work is also vital. This is not the time or place for flashy vocabulary words or rhetorical flourishes. Style, however, is still important: writing about the sciences doesn’t give you a pass to write sloppily.

Objectivity

The sciences aim for objectivity at every stage, from the experimental procedures to the language used in the write-up. Science writing must convince its audience that its offering an important, innovative contribution; as a result, it has an argumentative character. Combining objectivity and argumentative writing can be challenging. Scientific objectivity has two requirements: your hypothesis must be testable, and your results must be reproducible.

The importance of objectivity in the sciences limits writers’ ability to use persuasive rhetoric. However, it is still necessary to make a strong case for the importance, relevance, and applicability of your research. Argumentative writing does have a place in scientific papers, but its role is limited. You may use persuasive language in the abstract, introduction, literature review, discussion of results, and conclusion, but avoid using it when you describe your methods and present your results.

Transitions

Many students struggle to transition from one topic to the next. Transitions are well worth mastering—they are the glue that holds your ideas together. Never assume that the reader will correctly guess the relationships between different subtopics; it is your responsibility to explain these connections.

Scientific Reasoning

Keeping your chosen model in mind while you write can help ensure that your decisions and conclusions are logically consistent. Also, watch out for logic traps such as bias and faulty causality. Researchers must account for their own biases, or personal preferences, prejudices, and preconceived notions. These may include cognitive bias (irrational thinking), cultural bias (the imposition of one’s own cultural standards upon research subjects), and sampling bias (the tendency during sample collection to include some members of the intended sample more readily than others).

Overview of the IMRAD Model

The body of a scientific paper generally consists of the following sections: introduction (which may include a literature review), methods, results, and discussion.

In the natural and social sciences, the format for the body of the paper varies depending on the discipline, audience, and research methods. Generally, the body of the paper contains an introduction, a methods section, results, and discussion. This method is called IMRAD for short.

These sections are usually separate, although sometimes the results are combined with the methods. However, many instructors prefer that students maintain these divisions, since they are still learning the conventions of writing in their discipline. Most scientific journals prefer the IMRAD format, or variations of it, and even recommend that writers designate the four elements with uniform title headings.

Try to stay true to each section’s stated purpose. You can cite relevant sources in the methods, discussion, and conclusion sections, but again, save the lengthy discussion of those sources for the introduction or literature review. The results section should describe your results without discussing their significance, while the discussion section should analyze your results without reporting any new findings. Think of each section as a course served at a fancy dinner—don’t pour the soup into the salad or add leftover scraps from the entree to the dessert!

Introduction

In the first section of your paper, make a case for your new research. Explain to your reader why you chose to research this topic, problem, or issue, and why such research is needed. Explain any “gaps” in the current research on this topic, and explain how your research contributes to closing that gap.

Literature Review

While not always required, the literature review can be an important part of your introduction. It provides an overview of relevant research in your discipline. Its goal is to provide a scholarly context for your research question, and explain how your own research fits into that context. A literature review is not merely a summary of the sources you’ve found for your paper—it should synthesize the information gathered from those sources in order to demonstrate that work still needs to be done.

Explain your selection criteria early on—why did you choose each of your sources? The literature review should only refer to work that affects your particular question. Seek out a diverse range of sources. Look at primary-research reports and data sets in addition to secondary or analytical sources.

Methods

This section should explain how you collected and evaluated your data. Use the past tense, and use precise language. Explain why you chose your methods and how they compare to the standard practices in your discipline. Address potential problems with your methodology, and discuss how you dealt with these problems. Classify your methods. Are they empirical or interpretive? Quantitative or qualitative?

After you support your methods of data collection or creation, defend the framework you use to analyze or interpret the data. What theoretical assumptions do you rely on?

After you provide a rationale for your methodology, explain your process in detail. If you are vague or unclear in describing your methods, your reader will have reason to doubt your results. Furthermore, scientific research should present reproducible (i.e., repeatable) results. It will be impossible for other researchers to recreate your results if they can’t determine exactly what you did. Include information about your population, sample frame, sample method, sample size, data-collection method, and data processing and analysis.

Results

When you describe your findings, do so in the past tense, using impartial language, with no attempt to analyze the significance of the findings. You will analyze your results in the next section. However, it is perfectly acceptable to make observations about your findings. For instance, if there was an unexpectedly large gap between two data points, you should mention that the gap is unusual, but save your speculations about the reasons for the gap for the discussion section. If you find some results that don’t support your hypothesis, don’t omit them. Report incongruous results, and then address them in the discussion section. If you find that you need more background information to provide context for your results, don’t include it in the results section—go back and add it to your introduction.

Discussion

This is the place to analyze your results and explain their significance—namely, how they support (or do not support) your hypothesis. Identify patterns in the data, and explain how they correlate with what is known in the field, as well as whether they are what you expected to find. (Often, the most interesting research results are those that were not expected!) You should also make a case for further research if you feel the results warrant it.

It can be very helpful to include visual aids such as figures, charts, tables, and photos with your results. Make sure you label each of these elements, and provide supporting text that explains them thoroughly.

Royal Academy School: One of the goals of the literature review is to demonstrate familiarity with a body of knowledge.

Abstract

The abstract is the first (and, sometimes, only) part of a scientific paper people will read, so it’s essential to summarize all necessary information about your methods, results, and conclusions.

The Importance of the Abstract

The abstract of a scientific paper is often the only part that the reader sees. A well-written abstract encapsulates the content and tone of the entire paper. Since abstracts are brief (generally 300–500 words), they do not always allow for the full IMRAD structure. A specialized audience may read further if they are interested, and the abstract is your opportunity to convince them to read the rest. Additionally, the abstract of an article may be the only part that is available through electronic databases, published in conference proceedings, or read by a professional journal referee. Hence abstracts should be written with a non-specialized audience (or a very busy specialized audience) in mind.

What to Address in the Abstract

While each medium of publication may require different word counts or formats for abstracts, a good general rule is to spend one to two sentences addressing each of the following (do not use headers or use multiple paragraphs; just make sure to address each component):

Summarize Your Introduction

This is where you will introduce and summarize previous work about the topic. State the question or problem you are addressing, and describe any gaps in the existing research.

Summarize Your Methods

Next, you should explain how you set about answering the questions stated in the background. Describe your research process and the approach(es) you used to collect and analyze your data.

Summarize Your Results

Present your findings objectively, without interpreting them (yet). Results are often relayed in formal prose and visual form (charts, graphs, etc.). This helps specialized and non-specialized audiences alike grasp the content and implications of your research more thoroughly.

Summarize Your Conclusions

Here is where you finally connect your research to the topic, applying your findings to address the hypothesis you started out with. Describe the impact your research will have on the question, problem, or topic, and include a call for specific areas of further research in the field.

Introduction and Thesis

In academic writing, the introduction and thesis statement form the foundation of your paper.

Social sciences: The social sciences include academic disciplines like anthropology, sociology, psychology, and economics

The introduction can be the most challenging part of a paper, since many writers struggle with where to start. It helps to have already settled on a thesis. If you’re feeling daunted, you can sometimes write the other sections of the paper first. Then, when you’ve organized the main ideas in the body, you can work “backward” to explain your topic and thesis clearly in the first paragraph.

Present Main Ideas

The introduction to a social-science paper should succinctly present the main ideas. The goal of the introduction is to convince the reader that you have a valid answer to an important question. In order to do that, make sure your introduction covers these five points: the topic, the question, the importance of the question, your approach to the question, and your answer to the question.

Structuring Your Ideas

A popular introduction structure is the concept-funnel—begin with general information about your topic, narrow the focus and provide context, and end by distilling your paper’s specific approach. As you move from general background information to the specifics of your project, try to create a road map for your paper. Mirror the structure of the paper itself, explaining how each piece fits into the bigger picture. It is usually best to write the introduction after you have made significant progress with your research, experiment, or data analysis to ensure you have enough information to write an accurate overview.

Papers in the sciences generally aim for an objective voice and stay close to the facts. However, you have a bit more freedom at the beginning of the introduction, and you can take advantage of that freedom by finding a surprising, high-impact way to highlight your issue’s importance. Here are some effective strategies for opening a paper:

• Make a provocative or controversial statement
• State a surprising or little-known fact
• Make a case for your topic’s relevance to the reader
• Open with a relevant quote or brief anecdote
• Take a stand against something
• Stake a position for yourself within an ongoing debate
• Talk about a challenging problem or paradox

Establishing Relevance

After you engage your reader’s attention with the opening, make a case for the importance of your topic and question. Here are some questions that may help at this stage: Why did you choose this topic? Should the general public or your academic discipline be more aware of this issue, and why? Are you calling attention to an underappreciated issue, or evaluating a widely acknowledged issue in a new light? How does the issue affect you, if at all?

Thesis Statement

A thesis statement is a brief summary of your paper’s purpose and central claim. The thesis statement should be one to three sentences, depending on the complexity of your paper, and should appear in your introduction. A thesis statement in the social sciences should include your principal findings and conclusions. If writing about an experiment, it should also include your initial hypothesis. While there is no hard-and-fast rule about where to state your thesis, it usually fits naturally at or near the end of the introductory paragraph (not later than the very beginning of the second paragraph). The introduction should provide a rationale for your approach to your research question, and it will be easier to follow your reasoning if you reveal what you did before you explain why you did it.

Testability

Your thesis is only valid if it is testable. Testability is an extension of falsifiability, a principle indicating that a claim can be proven either true or false. The statement, “all Swedish people have blonde hair” is falsifiable—it could be proven false by identifying a Swede with a different hair color. For a hypothesis to be testable, it must be possible to conduct experiments that could reveal observable counterexamples. This is the equivalent of the principle in the humanities that a claim is only valid if someone could also reasonably argue against it.

Thesis Statements to Avoid

• The statement without a thesis: A statement of a fact, opinion, or topic is not a thesis. Push the thesis statement beyond the level of a topic statement, and make an argument.
• The vague thesis: If your thesis statement is too general, it will not provide a “road map” for readers.
• The “value judgment” thesis: Your argument should not assume a universal, self-evident set of values. Value-judgment-based arguments tend to have the structure “[latex]x[/latex] is bad; [latex]y[/latex] is good,” or “[latex]x[/latex] is better than [latex]y[/latex].” “Good,” “bad,” “better,” and “worse” are vague terms that do not convey enough information for academic arguments. In academic writing, it is inappropriate to assume that your reader will know exactly what you mean when you make an overly general claim. The burden of proof, and thorough explanation, is on you.
• The oversized thesis claim. There is only so much material you can cover within a page limit, so make sure your topic is focused enough that you can do it justice. Also, avoid arguments that require evidence you do not have. There are some arguments that require a great deal of research to prove—only tackle these topics if you have the time, space, and resources.

Methods

A methods section is a detailed description of how a study was researched and conducted.

IMRAD: The Methods Section

Your methods section should include a full, technical explanation of how you conducted your research and found your results. It should describe your assumptions, questions, simulations, materials, participants, and metrics.

Because the methods section is generally read by a specialized audience with an interest in the topic, it uses language that may not be easily understood by non-specialists. Technical jargon, extensive details, and a formal tone are expected.

The methods section should be as thorough as possible since the goal is to give readers all the information necessary for them to recreate your experiments. Scientific papers need a thorough description of methodology in order to prove that a project meets the criteria of scientific objectivity: a testable hypothesis and reproducible results.

Purpose of the Methods Section: Testability

Hypotheses become accepted theories only when their experimental results are reproducible. That means that if the experiment is conducted the same way every time, it should always generate the same, or similar, results. To ensure that later researchers can replicate your research, and thereby demonstrate that your results are reproducible, it is important that you explain your process very clearly and provide all of the details that would be necessary to repeat your experiment. This information must be accurate—even one mistaken measurement or typo could change the procedure and results drastically.

Example Methods Section

The following is an example of a methods section of a scientific paper:

“The study focused on a three-hundred-mile stretch of the Columbia River, which has been the stretch of the river most studied historically. Five locations were selected, each sixty miles apart. Each location had three different water samples selected for three levels of the water: surface water (3-inch depth), mid-level water (12-inch depth), and water from the bottom of the river (36-inch depth). Samples were taken three times a day, seven days a week, during a period stretching from the fourth month before until the fourth month after the year’s salmon run.”

Results

The results section of a scientific paper objectively presents the empirical data collected in a study.

Writing the Results Section

The results section is where you state the outcome of your experiments. It should include empirical data, any relevant graphics, and language about whether the thesis or hypothesis was supported. Think of the results section as the cold, hard facts.

Style

Since the goal of the scientific paper is to present facts, use a formal, objective tone when writing. Avoid adjectives and adverbs; instead use nouns and verbs. Passive voice is acceptable here: you can say “The stream was found to contain 0.27 PPM mercury,” rather than “I found that the stream contained 0.27 PPM mercury.”

Presenting Information

Using charts, graphs, and tables is an excellent way to let your results speak for themselves. Many word-processing and spreadsheet programs have tools for creating these visual aids. However, make sure you remember to title each figure, provide an accompanying description, and label all axes so that your readers can understand exactly what they’re looking at.

Was Your Hypothesis Supported?

This is the part where it is the most difficult to be objective. If you followed the scientific method, you began your research with a hypothesis. Now that you have completed your research, you have found that either your hypothesis was supported or it was not. In the results section, do not attempt to explain why or why not your hypothesis was supported. Simply say, “The results were not found to be statistically significant,” or “The results supported the hypothesis, with [latex]p <.05[/latex] significance,” or the like. Be objective—there will be time for interpretation later.

Getting Ready for the Conclusion

It will be tempting to start drawing conclusions as you write the results section. You may also want to compare your results to the outcomes of other researchers’ experiments. Resist the urge! You’ll get your chance to draw conclusions in the conclusion section. The most you should do in the results section is present the data your experiments produced. Think of it as laying the foundation for what you will conclude later on in your paper.

Discussion and Conclusion

The discussion section of a scientific paper analyzes and interprets the results of a study, while the conclusion explains implications for further research.

Discussion

The discussion section of a scientific paper should interpret the results of your research. First, briefly remind your reader of your research question and principal findings by briefly restating these points. Then explain the results themselves. Discuss how they fit (or do not fit) your hypothesis, and whether they are consistent with the results of similar research projects. Did you encounter anything surprising or idiosyncratic? If so, why is it significant? What might have caused it? Build on the research question you posed in the introduction, and the context you established in the literature review. Make a case for the meaning and significance of your findings, and support your case by connecting it to related research.

Acknowledge other possible interpretations of your results, and admit your project’s limitations. Your argument will be more convincing if you can anticipate your reader’s potential objections to your claims and address them directly in the discussion section. For example, generalizability (or how applicable a study’s results are to a more general population) is more limited with a smaller or less homogeneous sample. If your research sample is small or limited, be sure to acknowledge those limitations and address how they might have affected the results.

If your interpretation has broader implications, you can either suggest them in the discussion section or introduce them in a separate conclusion. You don’t have to write a conclusion if your points fit neatly into the discussion section, but a conclusion is helpful if you want to make suggestions that stretch beyond the scope of your project.

Conclusion

The conclusion section is not strictly necessary in the social sciences, but it can be helpful to provide a succinct summary of your work. It is also a good place to make bold speculations about the implications of your project. You should discuss, somewhere in your paper, the significance of your research for future research, public policy, personal decision-making, or other spheres of influence. But think carefully about whether you could benefit from the distancing effect of putting these implications in a separate conclusion.

The conclusion should not repeat your discussion section. It should take one to three paragraphs to restate the research question, the main results, and the meaning of those results. The conclusion then reaches beyond the suggestions you made in the body of the paper to emphasize the importance of the results and their potential consequences.

Conclusion: Conclusions have been an important part of writing for centuries.