Following our last post on writing good Introduction and Abstract, we discussed what a good Materials and Methods section should look like.
- The golden rule to writing a good methods section is to ask yourself whether your reader could replicate your study based on just the information you provided. Sufficient information should be provided so anybody could repeat your experiment and produce similar results. The challenge is to write this as concisely as possible. You are not conducting a high school experiment, so don’t have to tell the readers that you washed your test tubes or recorded field data with a blue ball point pen!
- If a protocol used is well-established, simply cite the developer’s paper. Only elaborate when the default protocol is modified to suit your study (e.g. changed default software settings) or if you’re writing for an audience which is not likely to be familiar with the technique.
- Field or lab based ecology papers typically use the following subheadings:
- Study site and/or focal species
- Field data collection (e.g. survey)
- Lab data collection (e.g. soil analysis)
- Statistical analysis
- A really useful alternative or addition to the structure above is to list each aim/question/hypothesis as a subheading; use the questions or hypotheses posed in your introduction to structure your methods (& results) sections. Relate the methods back to questions that you posed – explain HOW you tackled the questions posed.
- Include a figure if it helps to explain the study design (e.g. pot arrangement in a glasshouse, block or nested plot arrangement, map of study region, theoretical graphs), if there’s no space in the main text this can be a useful piece of online supplementary information. This helps reviewers and readers check your stats are appropriate and your degrees of freedom match up with the design (yes reviewers do check this kind of thing!).
- If you are writing a modelling paper, particularly for complex simulation or agent-based models, we recommend Volker Grimm’s ODD (Overview, Design concepts, and Details) protocol (see journal article here). In fact this way of thinking about a methods section is excellent for structuring your methods story even if you’re not writing a modelling methods section
- Modellers – while you might be a native geek speaker your favourite computer language may not be the mother tongue of your readers; consider putting a flow-chart of your algorithm and pseudo-code in supplementary information and your original code should be very well annotated.