The study is finally complete; you have conducted your analyses and obtained your results. Now all that's left to do is to write up the report. Reports can be in any number of formats from papers for conferences or journals to simple presentations or posters. For this blog post, we will focus on papers; however, most of the techniques can be applied to other formats as well.
A General Overview
Depending on the work you've done previously, you would have at most half of your report completed in your preregistration: the introduction, background, and methodology. If you wrote your preregistration more informally, you'll have to clean up the language, but the general process should be transferable. From there, you can add the results based on the outcomes of your study.
From here, there are typically four more sections to include: limitations, discussion, future work, and conclusion. The limitations section includes assumptions you made to run your study or things you didn't account for. The discussion section is used to interpret the outcomes of your study, with any necessary disclaimers. The outcomes are then tied back to answering the research questions or goals proposed at the beginning of your paper. The future work section contains where you or someone else in your field or study should do as a logical next step. This can include overarching goals as well, but its best to mention those as part of the introduction or conclusion. The conclusion section helps the reader understand why your study matters. This can be thought of the same in any literature that is written, though it's more common with explanatory or argumentative works.
Preregistration and Resources
If you have a preregistration or any additional resources, they should be linked in the paper as well. If you're writing your paper using Latex, they do allow hyperlinks via `\hyperlink
`, but you need to make sure your publication venue allows them.
The preregistration link should be provided in the methodology and in the appendix. The methodology section makes the most sense to provide easy reference as to what parts of the study were confirmatory or exploratory. The appendix meanwhile is a good way to have all of your links in a single place.
The resource link published in the abstract and as part of an appendix. Most publication sites make the abstract publicly available, so your resources will not be locked behind some paywall. In the appendix, I would recommend having a 'Resource Accessibility Statement' which contains direct links to your and any third-party resources. If the resources need to be requested, provide a process or a link to a document containing information on how to do so. Resources that can't be made public or on request should be mentioned as such.
Reporting Results
Reporting the results in your paper is subjective to how the author chooses to format it. I recommend, however, that the results in your paper should have a 1-to-1 correlations to those outputted by your analysis, whether in the terminal, text file, or image. By reporting the results in the same order it appears in the paper with no additional information, it makes it easier to verify the reproducibility of the work and to understand where the reported outcomes are coming from.
If you want to report more information inside the analysis itself, provide an option that only reports what's in the paper when enabled instead. Reproducibility work should always be consistent and easy to verify.
Some Additional Thoughts
Latex over Word
I'm pretty sure this is not as common anymore, but publication sites should favor using Latex templates over Word or Google docs. The format of Latex files can be changed while the contents itself remain the same. This is especially helpful when dealing with table spacing, image sizes, or text wrapping within columns.
Because of this, you can also use your analyses to generate some Latex format of your data such that it can be easily copy-pasted into your paper. It also provides the additional benefit of using a diff checker to validate results rather than manually checking the numbers.