Guidance on the planning and reporting of experimental design and analysis
Bibliographic record
Abstract
The aim of this guidance document is to help authors plan experiments, conduct analyses and present the results of work intended for publication in the British Journal of Pharmacology (BJP). The guidance is structured to minimize the risk of generating and publishing false findings. Below, we explain the key elements of experimental planning (blinding, randomization and adequate group sizes) and how to avoid generating incorrect findings (false positives in particular). We explain how to capture the relevant design and analysis information in your manuscript so it can be examined quickly, efficiently and fairly in peer review. In accordance with previous modifications to our requirements at BJP, we have eliminated a great deal of the items set forth in the most recent guidance document (Curtis et al., 2022) that were found to be too granular or subject-specific. We also hope to dispel some myths about what BJP will consider, in a clear and helpful manner, and encourage more authors to submit their work confidently to the journal according to the vision of the editorial board (Papapetropoulos et al., 2023). The three key elements of experimental planning should be reported in the ‘Experimental design and analysis section’ of ‘Methods’. Compliance with the guidance cannot normally be adjusted after the experiments of a study are complete. However, data generated by experiments not conforming to these three elements may be reported in the paper if such data are a preliminary or minor part of your narrative. There is an easy way to decide if such data are appropriate for BJP: If the data in question can safely be excluded from your abstract and your conclusions without undermining the narrative, then you may include them in your paper. We expect referees to consider such data and not simply recommend rejection of the manuscript. In ‘Methods’, please explain which P value you have stipulated to denote statistical significance when comparing between groups, time points and so forth. This is almost always P < 0.05. When ANOVA or related multi-group statistics are employed, remember the F statistic and the variance homogeneity are the gatekeepers that determine whether you can justifiably compare individual groups with one another. Please state in ‘Methods’ that ‘post hoc tests (such as Tukey's test) were run only if F were significant (P<0.05) and there was no variance inhomogeneity’. The same requirement applies to more complex multi-group analysis: repeated measures analysis and analysis of covariance, for example. It is particularly important to follow this rubric and make this clear in ‘Methods’ as some software packages will allow a post hoc test to be run even when these conditions are not met, generating false positive results. If you planned to perform parametric analysis (t test or multiple comparison tests) but cannot because conditions are not met, you may find that a log transform generates Gaussian data that remove the variance inhomogeneity. The data may then be amenable to parametric testing. If this is not the case, then please use non-parametric statistics. Individual values or samples should not be excluded from data analysis unless the exclusion criteria have been defined in ‘Methods’ and the number of samples or values excluded per group is reported. The best place for reporting such an occurrence is the figure/table legend. If an experiment is worth doing, it is worth planning. If it is not worth planning, it may be not worth doing. Any issues concerning data analysis can be resolved once a study has been completed, but a badly designed study may be unpublishable. Here, we explain how to plan your experiments, so they incorporate randomization, blinding and independent group sizes of at least n = 5 into the design, and how to get your paper published if you cannot do this. The key points are summarized in Figure 1. It means that experiments that are not randomized or blinded or adequately powered may be included in the paper, but the author will need to explain the value of the data which must be presented without the statistical analysis that pharmacologists normally use as a pattern recognition aid. You are invited to contact the BJP consulting editor for design and analysis if you would like advice on your experimental planning — before you start your experiments. All authors contributed to the design and writing of the manuscript. The authors declare no conflicts of interest. N/A-Editorial.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.302 | 0.599 |
| Meta-epidemiology (narrow) | 0.004 | 0.008 |
| Meta-epidemiology (broad) | 0.007 | 0.005 |
| Bibliometrics | 0.010 | 0.008 |
| Science and technology studies | 0.005 | 0.010 |
| Scholarly communication | 0.013 | 0.009 |
| Open science | 0.008 | 0.005 |
| Research integrity | 0.019 | 0.021 |
| Insufficient payload (model declined to judge) | 0.162 | 0.133 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".