The need to create more credible research has long been recognised but is now beginning to gain attention, with various different studies identifying issues in reproducibility and reliability (1-5). Interestingly, these issues are
not just in neuroscience, but have been identified across multiple disciplines including economics, psychology and in education - just to name a few (6)! Some of this is to be expected, as there are natural variations across research systems, such as in the life sciences. However, common issues across all disciplines included methods in reporting and evaluating research. There are now several initiatives to tackle this and to try and create more transparent methods of research reporting and evaluation.
Find out more
about the initiatives we have signed up for to improve credible research.
Research practice issues
A question that is often asked is whether credibility is an issue. Over time, research culture has led to the generation of data and results from experiments whereby credibility is often an afterthought. Ask any researcher and they will be well aware of the dreaded
'publish or perish' nature of research; where the value of a researcher and their work is judged on their ability to publish high quality data in high impact journals, or else they face their research being overlooked and undervalued. While researchers generally strive for their data to be published in these high impact journals, these journals often only accept novel data, statistically significant results, or results which support the hypotheses, thus ruling out a vast number of studies. Unfortunately, this has also led to more unfavourable methods of ensuring research does get published.
For example,
p-hacking is now increasingly common, where researchers may trial different statistical analyses, increase sample sizes retrospectively, or misuse data analysis in other ways to ensure their results reach significance.
Another example of unfavourable research practices includes
HARKing, where the researcher
Hypothesises
After
Results are
Known, attempting to negate publication bias against null results. This type of malpractice can lead to the publication of data which is unreliable, unrepeatable, and potentially even untrue. By promoting credible research, we hope to negate these practices and advocate for research integrity, where the research undertaken is a true understanding of science.
The credibility initiative
However, while there is still a long way to go to promote credible research across neuroscience and all science disciplines, it is encouraging to think of all the progress that has been made. To tackle issues such as HARKing, initiatives such as
study pre-registration and
registered reports are new ideas to help publish the basis to a study, including the aims, hypothesis, and methods, prior to any results being known and thus focus on the outcome of the science rather than whether the results fit a certain narrative. Practices such as
open data and other
open science practices have also helped to ensure that the data has not been manipulated and does provide the published results. The creation and inclusion of these credible research practices has led to a shift in research culture promoting overall research credibility. In fact, many journals are now requesting that the data from their publications be deposited in
open data archives (7). Similarly, major funding bodies are also supporting these credible science practices by requesting any work funded by them are published in
open access journals to ensure complete transparency (8). Together, these initiatives across different platforms within neuroscience research are tackling any issues with credibility and promoting good, credible science.
