Research in materials that aid in diagnosis, disease prevention and treatment has always received great attention, but work in healthcare comes with great responsibility. The transition of these materials from bench to bedside requires reproducibility and reliability, which are strongly dependent on correct reporting of results, as it is in all fields of science. Along with this comes the responsible use of the word “significant”, which should be strictly avoided when correct statistical analysis is absent. Although there is criticism regarding the meaning and the use of P-values, the gold standard in scientific writing is still to present your research with all the relevant information and interpret your data taking into account the right statistical calculation.
As Editors of journals such as Advanced Biosystems and Advanced Healthcare Materials, amongst others, we want to publish your research in its best form; therefore, we have adopted the policy of correct data presentation. With the hope of supporting you in this aspect, we have compiled list of do’s and don’ts:
- For increased confidence in the data, reproduce the results in independent experiments
- Make sure to use a reasonable number of replicates, considering the variation in the nature of the experiment you are conducting
- Report details of your experiments and standardize your setup: use same sex or age of animals for in vivo studies, or cells at the same passage number for in vitro studies
- Include the correct controls, both positive and negative, in your experiments
- Wherever possible, provide quantitative data along with qualitative data
- Evaluate your experimental setup correctly and use the right statistical test: e.g. don’t use t test instead of analysis of variance (ANOVA) when you need to compare the means of more than two groups
- Include error bars in your data plots
- Consider box plots or scatter plots over bar plots to better illustrate the variance in your samples
- In figure legends, mention:
- how many replicates were used
- what the graphs and the error bars represent
- which statistical test was used
- what the significance symbol stands for
For example, data presented as mean ± SEM, n=6, P-values are calculated using one-way ANOVA with Bonferroni correction, *P<0.05, **P<0.01.
In the methods section of your manuscript, have a subsection on statistical analysis, whereby you mention:
- how data was presented
- how significance was calculated (the statistical test in its correct form with its assumptions)
- what was considered significant
- which software was used for the analysis of significance
For example, continuous variables are expressed as mean ± SEM. For normally distributed data sets with equal variances, one-way ANOVA testing followed by a Tukey post-hoc test was carried out across groups. In all cases, significance was defined as p ≤ 0.05. Statistical analysis was carried out using X Software.
We hope that this information helps in your scientific endeavors.