We all know that fabrication, falsification and plagiarism fall under the definition of research misconduct. However, the VIB definition of research misconduct is broader than that. It also includes practices “that seriously deviate from those that are commonly accepted within the scientific community for proposing, conducting or reporting research.”
VIB guidelines and recommendations
There is a continuum from truly correct to truly deceptive scientific research. The grey area in
between is often referred to as ‘Questionable Research Practices’ (QRP). Whether a form of QRP
will qualify as research misconduct is very much determined by the intent of the researcher.
Examples of QRP:
• Neglecting negative outcomes
• Using inappropriate statistics to support one’s hypothesis
• Inappropriate research design
• Leaving out relevant controls
• Inappropriate re-use of controls
• Removal of ‘outliers’
• Conscious bias
• Unethical experimentation
• Peer review abuse
Examples of actions that lead to incorrect data being taken up in the scientific record but are not
considered scientific misconduct
• Honest error
• Unconscious bias
Two different ways of representing good versus deceptive science and the continuum of questionable research practices in between, indicated by the circle. Adapted from a presentation by Daniel Fanelli.
A critical mindset to prevent unintentional QRP
Much, if not all, unintentional QRP can be prevented by applying very high quality standards
and a very critical mindset. Designing proper scientific experiments can be learned, and the
same holds true for ensuring the quality of their execution. Perhaps one of the most difficult forms
of QRP to avoid is unconscious bias. We all have certain hopes or expectations for the outcome of a scientific experiment we have designed. And it is very easy not to see certain signals that may be telling us that the ‘truth’ is somewhat different from what we were expecting. It is therefore all the more important to step outside the box of one’s own thinking and look at one’s data from different angles. Critical discussion with one’s colleagues and mentors will also help prevent mistakes.
Critically discuss experiments in group
When we are making decisions, the choices are definitely not all black and white. What is the
actual appropriate statistical method for analyzing the data produced by a particular experiment?
Which controls – positive and negative – should be included? Should there be different controls for
different treatment regimens? How many? What about outliers? Do they reflect a chance event that
can be ignored and can be considered a statistical anomaly? Or does the outlier tell us something relevant about the topic under study and removing it will skew the conclusion?
These are all issues that should be thoroughly discussed beforehand, preferably with a group. The more critical minds look at it, the better. Quality and honesty should be the guiding principles for the discussion. Never look the other way when someone proposes something that raises your eyebrows. Make a comment. Ask a question. The more an issue is critically discussed in group, the lower the chances someone will knowingly or unknowingly violate the principles of quality and honesty.
An appeal to your good judgement
In recent years the discussion about scientific misconduct has become much more intense. More fraud cases are out in the open and questions are asked about whether or not the frequency of scientific fraud has been increasing. Has it? Scientific fraud is a very serious offense that should not be undertaken. Period… End of discussion…
With this article we would like to raise the awareness about the fact that there is much more
than only blatant fraud and that we need to do all we can to prevent all forms of questionable
research practices. This is an appeal to intensify discussions about the design and interpretation of
experiments wherever necessary. Make the correct decisions. Be honest. Do not cut corners in the
hope that your conclusions will then become appealing enough to be accepted in a certain
scientific journal. If the whole of the scientific community would do the same effort, we could
perhaps decrease the number of scientific results that cannot be reproduced.
If you have questions about scientific integrity and questionable research practices you can contact
the VIB Research Integrity Officer, René Custers (+32 9 244 66 11, rene.custers*Replace*With*At*Sign*vib.be)