Oct 25, 2021

How to Write the Methods Section of a Research Paper

A common piece of advice for authors preparing their first journal article for publication is to start with the methods section: just list everything that was done and go from there. While that might seem like a very practical approach to a first draft, if you do this without a clear outline and a story in mind, you can easily end up with manuscript sections that are not logically related to each other. 

Since the methods section constitutes the core of your paper, no matter when you write it, you need to use it to guide the reader carefully through your story from beginning to end without leaving questions unanswered. Missing or confusing details in this section will likely lead to early rejection of your manuscript or unnecessary back-and-forth with the reviewers until eventual publication. Here, you will find some useful tips on how to make your methods section the logical foundation of your research paper.

Not just a list of experiments and methods

While your introduction section provides the reader with the necessary background to understand your rationale and research question (and, depending on journal format and your personal preference, might already summarize the results), the methods section explains what exactly you did and how you did it. The point of this section is not to list all the boring details just for the sake of completeness. The purpose of the methods sections is to enable the reader to replicate exactly what you did, verify or corroborate your results, or maybe find that there are factors you did not consider or that are more relevant than expected. 

To make this section as easy to read as possible, you must clearly connect it to the information you provide in the introduction section before and the results section after, it needs to have a clear structure (chronologically or according to topics), and you need to present your results according to the same structure or topics later in the manuscript. There are also official guidelines and journal instructions to follow and ethical issues to avoid to ensure that your manuscript can quickly reach the publication stage.

Table of Contents:

  1. General Methods Structure: What is Your Story? 
  2. What Methods Should You Report (and Leave Out)? 
  3. Details Frequently Missing from the Methods Section
  4. More Journal Guidelines to Consider 
  5. Accurate and Appropriate Language in the Methods

General Methods Structure: What Is Your Story? 

You might have conducted a number of experiments, maybe also a pilot before the main study to determine some specific factors or a follow-up experiment to clarify unclear details later in the process. Throwing all of these into your methods section, however, might not help the reader understand how everything is connected and how useful and appropriate your methodological approach is to investigate your specific research question. You therefore need to first come up with a clear outline and decide what to report and how to present that to the reader.

The first (and very important) decision to make is whether you present your experiments chronologically (e.g., Experiment 1, Experiment 2, Experiment 3…), and guide the reader through every step of the process, or if you organize everything according to subtopics (e.g., Behavioral measures, Structural imaging markers, Functional imaging markers…). In both cases, you need to use clear subheaders for the different subsections of your methods, and, very importantly, follow the same structure or focus on the same topics/measures in the results section so that the reader can easily follow along (see the two examples below).

If you are in doubt which way of organizing your experiments is better for your study, just ask yourself the following questions:

  • Does the reader need to know the timeline of your study? 
  • Is it relevant that one experiment was conducted first, because the outcome of this experiment determined the stimuli or factors that went into the next?
  • Did the results of your first experiment leave important questions open that you addressed in an additional experiment (that was maybe not planned initially)?
  • Is the answer to all of these questions “no”? Then organizing your methods section according to topics of interest might be the more logical choice.

If you think your timeline, protocol, or setup might be confusing or difficult for the reader to grasp, consider adding a graphic, flow-diagram, decision tree, or table as a visual aid.

What Methods Should You Report (and Leave Out)?

The answer to this question is quite simple–you need to report everything that another researcher needs to know to be able to replicate your study. Just imagine yourself reading your methods section in the future and trying to set up the same experiments again without prior knowledge. You would probably need to ask questions such as:

  • Where did you conduct your experiments (e.g., in what kind of room, under what lighting or temperature conditions, if those are relevant)? 
  • What devices did you use? Are there specific settings to report?
  • What specific software (and version of that software) did you use?
  • How did you find and select your participants?
  • How did you assign participants into groups?  
  • Did you exclude participants from the analysis? Why and how?
  • Where did your reagents or antibodies come from? Can you provide a Research Resource Identifier (RRID)?
  • Did you make your stimuli yourself or did you get them from somewhere?
  • Are the stimuli you used available for other researchers?
  • What kind of questionnaires did you use? Have they been validated?
  • How did you analyze your data? What level of significance did you use?
  • Were there any technical issues and did you have to adjust protocols?

Note that for every experimental detail you provide, you need to tell the reader (briefly) why you used this type of stimulus/this group of participants/these specific amounts of reagents. If there is earlier published research reporting the same methods, cite those studies. If you did pilot experiments to determine those details, describe the procedures and the outcomes of these experiments. If you made assumptions about the suitability of something based on the literature and common practice at your institution, then explain that to the reader.

In a nutshell, established methods need to be cited, and new methods need to be clearly described and briefly justified. However, if the fact that you use a new approach or a method that is not traditionally used for the data or phenomenon you study is one of the main points of your study (and maybe already reflected in the title of your article), then you need to explain your rationale for doing so in the introduction already and discuss it in more detail in the discussion section.

Note that you also need to explain your statistical analyses at the end of your methods section. You present the results of these analyses later, in the results section of your paper, but you need to show the reader in the methods section already that your approach is either well-established or valid, even if it is new or unusual. 

When it comes to the question of what details you should leave out, the answer is equally simple ‒ everything that you would not need to replicate your study in the future. If the educational background of your participants is listed in your institutional database but is not relevant to your study outcome, then don’t include that. Other things you should not put into your methods section:

  • Background information that you already presented in the introduction section.
  • In-depth comparisons of different methods ‒ these belong in the discussion section.
  • Results, unless you summarize outcomes of pilot experiments that helped you determine factors for your main experiment.

Also, make sure your subheadings are as clear as possible, suit the structure you chose for your methods section, and are in line with the target journal guidelines. If you studied a disease intervention in human participants, then your methods section could look similar to this:

materials an methods breakdown

Since the main point of interest here are your patient-centered outcome variables, you would center your results section on these as well and choose your headers accordingly (e.g., Patient characteristics, Baseline evaluation, Outcome variable 1, Outcome variable 2, Drop-out rate). 

If, instead, you did a series of visual experiments investigating the perception of faces including a pilot experiment to create the stimuli for your actual study, you would need to structure your methods section in a very different way, maybe like this:

materials and methods breakdown

Since here the analysis and outcome of the pilot experiment are already described in the methods section (as the basis for the main experimental setup and procedure), you do not have to mention it again in the results section. Instead, you could choose the two main experiments to structure your results section (Discrimination and classification, Familiarization and adaptation), or divide the results into all your test measures and/or potential interactions you described in the methods section (e.g., Discrimination performance, Classification performance, Adaptation aftereffects, Correlation analysis).

Details Frequently Missing from the Methods Section

Manufacturer information: For laboratory or technical equipment, you need to provide the model, name of the manufacturer, and company’s location. The usual format for these details is product name (company name, city, state) for US-based manufacturers, and product name (company name, city/town, country) for companies outside the US.

Sample size and power estimation: Power and sample size estimations are measures for how many patients or participants are needed in a study in order to detect statistical significance and draw meaningful conclusions from the results. Outside of the medical field, studies are sometimesoften still conducted with a “the more the better” approach in mind, but since many journals now ask for those details, it is better to not skip this important step.

Ethical guidelines and approval: In addition to describing what you did, you also need to assure the editor and reviewers that your methods and protocols followed all relevant ethical standards and guidelines. This includes applying for approval at your local or national ethics committee, providing the name or location of that committee as well as the approval reference number you received, and, if you studied human participants, a statement that participants were informed about all relevant experimental details in advance and signed consent forms before the start of the study. For animal studies, you usually need to provide a statement that all procedures included in your research were in line with the Declaration of Helsinki. Make sure you check the target journal guidelines carefully, as these statements sometimes need to be placed at the end of the main article text rather than in the method section.

More Journal Guidelines to Consider 

Structure & word limitations: While many journals simply follow the usual style guidelines (e.g., APA for the social sciences and psychology, AMA for medical research) and let you choose the headers of your method section according to your preferred structure and focus, some have precise guidelines and strict limitations, for example on length and maximum number of subsections or header levels. Make sure you read the instructions of your target journal carefully and restructure your method section if necessary before submission. If the journal does not give you enough space to include all the details that you deem necessary, then you can usually submit additional details as “supplemental” files and refer to those in the main text where necessary.

Standardized checklists: In addition to ethical guidelines and approval, journals also often ask you to submit one of the official standardized checklists for different study types to ensure all essential details are included in your manuscript. For example, there are checklists for randomized clinical trials, CONSORT (Consolidated Standards of Reporting Trials), cohort, case-control, cross‐sectional studies, STROBE (STrengthening the Reporting of OBservational studies in Epidemiology), diagnostic accuracy, STARD (STAndards for the Reporting of Diagnostic accuracy studies), systematic reviews and meta‐analyses PRISMA (Preferred Reporting Items for Systematic reviews and Meta‐Analyses), and Case reports, CARE (CAse REport).

Make sure you check if the manuscript uses a single- or double-blind review procedure, and delete all information that might allow a reviewer to guess where the authors are located from the manuscript text if necessary. This means that your method section cannot list the name and location of your institution, the names of researchers who conducted specific tests, or the name of your institutional ethics committee.  

Accurate and Appropriate Language in the Methods

Like all sections of your research paper, your method section needs to be written in an academic tone. That means it should be formal, vague expressions and colloquial language need to be avoided, and you need to correctly cite all your sources. If you describe human participants in your method section then you should be especially careful about your choice of words. For example, “participants” sounds more respectful than “subjects”, and patient-first language, that is, “patients with cancer”, is considered more appropriate than “cancer patients” by many journals.

Passive voice is often considered the standard for research papers, but it is completely fine to mix passive and active voice, even in the method section, to make your text as clear and concise as possible. Use the simple past tense to describe what you did, and the present tense when you refer to diagrams or tables. Have a look at this article if you need more general input on which verb tenses to use in a research paper

Lastly, make sure you label all the standard tests and questionnaires you use correctly (look up the original publication when in doubt) and spell genes and proteins according to the common databases for the species you studied, such as the HUGO Gene Nomenclature Committee database for human studies.  

Visit wordvice.ai to receive free online English proofreading and editing before submitting your manuscript to journal editors.