Since the first half of the 20th century, the IMRAD (Introduction, Methods, Results, and Discussion) structure has become the dominant structure for scientific reports on original research (i.e. most journal articles, congress papers, bachelor and master theses, etc.). The four elements that make up this structure are the main ingredients for a scientific report and are preceded by an abstract and followed by conclusions. They also usually form the main headings for the successive sections of a scientific report or paper.

In a way, the IMRAD structure reflects the process of scientific discovery through the empirical cycle. Although the empirical cycle is often complex and involves many iterative feedback loops, the IMRAD structure seems to reduce it to an oversimplified, linear, and stepwise process. This is also why the IMRAD structure has been criticized in the past for being too rigid and too simplistic (e.g. Medawar, 1964). Nevertheless, the IMRAD method has been adopted by the majority of journals across a wide range of disciplines because it provides a clear and logical structure that helps the reader to browse through articles more quickly to find relevant information. Furthermore, papers and reports written according to the IMRAD method contain all required information to evaluate the quality of the research without unnecessary details.

In contrast to other scientific texts, Earth Science specific texts often contain a separate, additional chapter in between Introduction and Methods: the Background chapter. This chapter (or sub-section of the introduction, depending on the preferred style of the field or research journal) generally includes relevant background information, for example in terms of the geological or tectonic setting of the study area, that is relevant for readers to understand the context of the research presented in the text.

The title of a paper or report reflects the content of the report and is informative and short (approximately 15 words at maximum). Omit any redundant phrases, such as “a study of…” or “a report of…”. For reports, a subtitle may be added. An example of an appropriate title is “Assessment of soil erosion in Africa using remote sensing”, because it informs the reader about 1) the subject (assessment of soil erosion), 2) the location (Africa), and 3) the research method (remote sensing). In a paper, the title is placed on top of the first page, followed by a list of authors. In reports, the title is on the cover page. The cover page also includes the names of the authors (and student number, if applicable), date and location of publication, name and code of the course, and the name of the supervisor(s) (if applicable).

In the abstract you concisely summarize the context, purpose, methods, main findings, and major conclusions of the research, in this particular order⁶. It should inform the reader about the highlights of your work. Be aware that the abstract needs to announce the content of the report/paper or study, i.e. your findings, and not what the text contains⁷. The summary or abstract is not a chapter, so it is not numbered. For both research reports and papers, include a list of 5 key words that are related to your study. These key words should be mentioned at the end of the abstract.

Some research reports will contain a table of contents that consists of a list of all chapters, sections, and subsections. This is often the case for long texts, such as theses, but less common for scientific publications. Make sure you know what the preferences are for the editor/lecturer you are writing for.

The chapter and section titles in the table of contents should be short and each title should cover the contents of the section concerned. The table of contents reflects the logical order of the chapters and sections and refers to the respective page in the report where this section starts. Furthermore, extensive reports (i.e. theses) often include a complete list of figures and tables and a list of appendices with reference to the pages they appear on. Both tables and figures are numbered consecutively (do not use Roman numerals), but in research reports it is also allowed to number the tables and figures consecutively per chapter (e.g. Figure 1.1, 1.2 or Table 3.1, 3.2, .3.3, 3.4).

In the introduction section or chapter, you indicate what has been studied, why it has been studied, and, in general terms, how it has been studied. The introduction comprises the top half of the hourglass format (Fig. 1), starting broadly with the largest scale of impact of the study and the problem that exists at that scale (step 1 in Fig. 1). This is followed by an explanation of why that problem (still) exists (step 2), and it ends with the novel attempt you take at solving this problem, forming the aim of the study, or hypothesis related to a certain research question (step 3). The introduction then briefly indicates the methods you use to solve it (step 4) in an outline of the approach taken, and briefly summarizes in what direction the results will be developed without indicating what the results or conclusions will be.

Ideally you already know the relevance of your research topic before collecting your data, and you should be aware of the relevant scientific background before processing and analysing your data. This should provide you with enough information to start a rough draft of your introduction in the initial phase of your study, when you are gathering and reading literature. Nonetheless, many research projects yield unexpected results, so you need to check at the end whether you have provided all relevant information (if not, add), or whether all information you provided is indeed relevant (if not, delete). Keep in mind that you can only write the final version of your introduction after you have finished at least a complete draft of the discussion⁷. Only then will you know the findings of your research.

If the problem you present in the introduction relates to a certain study area, it can be valuable to shortly introduce the geological setting to place your problem in that context. In that case, be concise and do not include details that your audience does not (yet) need to understand your problem statement. In-depth information may follow in the Background/Geological setting chapter.

The background introduces the reader to the relevant context that is required to understand your specific study. This includes background information on your research setting (i.e. regional geology/geography etc.), as well as the relevant ‘history of thought’ (i.e. results found and conclusions drawn by previous researchers). For this reason, the title of the background chapter can also vary depending on its context (e.g. it may be called Geological setting). As you write the background chapter, think about its purpose for your audience: build up the chapter in a way that its relevance for your problem statement is clear to an audience that has no or little understanding of your field of interest.

Keep in mind that the background chapter describes the setting objectively. It does not, in contrast to the introduction, contain any arguments for the relevance or value of your work. Despite the many references to previous, relevant research in this chapter, it should not be seen as a full literature review and it should not summarize everything that is somehow related to your topic. Remember, all relevant background information (and thus, most sources) that you want to discuss in your discussion need to be covered in either your introduction or background.

The methods section or chapter, sometimes called “Materials and Methods“, describes how the study was conducted. It describes the procedures that you have taken, which materials you used to find an answer to your research question, and contains all necessary information to repeat the study. In the methods section, you show that the data were collected in a consistent, accurate, and accountable manner. This provides important information under which conditions your research outcomes have been established and, thus, in which context they can be interpreted. Although the methods section is primarily descriptive, it is also recommended to give reasons why you have chosen the methods and materials used. The methods section may contain the following subsections (if relevant):

• Field methods: description of the type of data (e.g. distance, height, thickness, direction, electrical conductivity) or samples (e.g. rock, soil, sediment, water samples) you collected in the field, how they were collected (e.g. corings), and used measurement devices (e.g. GPS, ruler, compass, EC meter; for these, provide manufacturer and device type if relevant). In the case of mapping, also describe which parameters were mapped and which system was used (e.g. classification system). Details of the sampling or observation strategy (e.g. transect sampling, observations of outcrops, random sampling, stratified sampling) should also be given here.
• Laboratory methods: description of the laboratory procedures or experiments: which parameters were measured using which method?
• Data processing: description of the statistical methods you used to analyze the measurement data.
• If available, also include information about the accuracy and precision of the acquired data, e. if it is based on independent, existing information (e.g. literature, device manuals etc.). If you have examined the accuracy and precision of the data yourself as part of your study (for example as part of a quality control), you should report only the methods here. The results of this examination should then be reported in the results section.

In the results section or chapter you describe the main outcomes of your study obtained according to the methods section in a neutral (i.e. without interpretations that may be subject to discussion) and accessible manner. Limit yourself to only those results that are relevant to answer your research question. Generally, you do not use many citations in this section as you focus on your own data, and are not yet interpreting it⁷.

Structure the description of the results by, for example, describing the primary features first, and then the secondary and tertiary features, etc. You may also order the description of your results geographically (e.g. from north to south), or chronologically (e.g. from old to young).

Present the relevant results in the form figures, tables, photographs, and maps. In the accompanying text, you elucidate the general patterns in the outcomes and give relevant quantitative information (e.g., “The upper layer is usually 2 m thicker than the approximately 30 cm thick lower layer” or “The grain size increases with depth from 2 mm near the surface to 5 cm at 1 m depth”). Be as concise and to the point as possible when you describe your data⁷, quantify as much as possible, and avoid vague descriptions.

You can start to write the results section as soon as you have a good overview of your data and have made a selection of what you want/need to include in your report. Be aware that any data you want to discuss in the discussion needs to be mentioned in the results section.

In the discussion section or chapter, you give interpretations of your results and use these to answer the questions you phrased in your introduction. You are not allowed to present newly acquired data or published data that you haven’t mentioned before in your setting or introduction. This means that everything you want to discuss needs to be mentioned in the introduction, setting, methods, or result section. Also be aware that you need to back up your interpretations with a clear, scientific reasoning: don’t try to convince the reader, but simply explain your line of reasoning. It is important that the reader can follow your reasoning pathway and has the freedom to develop their own.

A persistent misconception is that the discussion section is meant to provide an overview of all flaws and uncertainties related to your observations and findings: this only diminished the impact of the research. The uncertainties of the research are given in the results. In the discussion, you identify and discuss what can be inferred from your outcomes bearing in mind the flaws and uncertainties that you honestly identified before.

Generally, the structure of the discussion follows the lower half of the hourglass format: you start small by relating and comparing your results to each other (e.g. Do they support or contradict each other?) and then put them in a broader context of the literature (e.g. what did others find that relate to your subject?). Alternative interpretations may also be given (e.g. “The results suggest …, but could also imply …”).You should also discuss the consequences of your findings for the aim/research question/hypothesis (e.g., do the outcomes support or oppose the hypothesis? should you revise your assumptions? what information is still missing to definitely answer your research question?). Make sure you answer or at least discuss all the questions you posed in your introduction⁷ . If it is not possible to solve a question yet, or if there are multiple explanations that are all equally likely, you do not need to give a solution. However, you then must explain this to the reader⁷.

Broadening further, you identify and discuss the implications for your field of science (e.g. what new questions arose from your work) and society (e.g. how can the new findings be used to solve societal issues or to develop new technology?). Remember to link back to the research questions and common thread you introduced in your first chapter(s) (see ‘Introduction’). If you have come across discussion points that do not relate to them, consider leaving them out to create a more focused discussion chapter. Alternatively, if your results do not answer your initial research question, consider changing or rephrasing the research questions you set earlier on.

In the conclusions section you summarize the main findings and provide answers to your research question. The conclusions should logically follow from your results and discussion. Do not present new facts or results that have not been discussed in a previous section. Finally, draw wider conclusions regarding the implications for new questions that arose from your study and the implications of the results for ‘the greater picture’. This is also the main difference between the conclusion and the abstract, as the latter focuses more on the study itself, although some overlap cannot be avoided. Follow the bottom half of the hour glass and do not forget to list the ‘small’ conclusions.

It is common to include an acknowledgements section at the end of a larger research report, such as your thesis. Here, you can thank and refer to persons and institutes who have contributed to your research (e.g. by assisting fieldwork or laboratory analyses, for providing data, funding, or for feedback on an earlier draft of your report or paper). If you have been funded by a specific agency and have been assigned a grant number, this would be the place to mention that.

It is up to you who you want to include in this section but remember to acknowledge all who contributed to your research outcomes. For a thesis, it is common to acknowledge your supervisors, support staff, and funding bodies. You may also mention your research group, other students, friends and family, and other people/institutions you feel have contributed to your project. When you decide to include a personal thank you note, keep it as professional as possible. Do not make jokes and do not add frivolous acknowledgements (to your dog or stuffed unicorn).

At the end of the report or paper (after the conclusions section or chapter), you include a reference list, in which you list all literature to which you have referred in your main text. Conversely, all references in the reference list should appear in the main text. The reference list is not a chapter, so it is not numbered. Just as the summary, the reference list is ordered in alphabetical order.

Make sure you keep track of your citations during the complete writing process. Document every source you read and cite from the first literature study to avoid any missing sources, which could eventually result in plagiarism. We highly recommend you to use a reference management tool during the writing process. If you add all your references to it, you can make the reference list after you have finished the complete draft version of your report. In the case you are not using one, update the reference list throughout the writing process to avoid a high workload at the end. See the Reading scientific literature for more information on reference management. 

An appendix contains materials that would obscure the structure and message of the text if it is included in the main text. Examples include tables of all quantitative research results, examples of field observation forms, or computer code. All appendices should be given an appropriate title and numbered consecutively.