Misconduct in science

Sir, – Prof William Reville's column on fraudulent science, which notes that an increasing number of researchers have been found to falsify data to increase the likelihood of publication, is very telling regarding how the current academic scientific system works ("Fraud is now the biggest enemy of science", June 2nd).

The majority of research scientists around the world work on short-term contracts. The number of professors will typically be a small fraction of postdoctoral researchers. For many, their salaries for these contracts are dependent on their ability to win financial assistance from funding bodies. The higher your publication count, the more likely you are to succeed. You need to publish at a rate that sustains your career. It is a “publish or perish” working environment.

I wonder if a secure working environment for researchers would help to tackle the worrying trends we see in scientific publications.

Science, as a profession, never went through the same processes as the legal or medical ones – with all the benefits of things like career structure and ethics. Perhaps it should. – Yours, etc,



School of Physics ,

Trinity College Dublin,

Dublin 2.

Sir, – There is no question that the fraudulent reporting of scientific research presents a grave challenge to modern science, and no question that the problem is growing. My concern is that media reports of the phenomenon often struggle to give an accurate portrayal of the scale of the problem, and often convey the impression that the problem is equally prevalent throughout all of science.

For example, a recent Proceedings of the National Academy of Sciences (PNAS) study found that over 2,000 scientific research articles were retracted (or withdrawn) from the PubMed database of scientific literature, and that 43 per cent of these articles were retracted for reasons of fraud. This is a matter of great concern to any scientist, but few media reports of the PNAS study pointed out that the retracted articles represented about 0.01 per cent of the total number of research articles published on the same database over the same period – and thus the number of articles retracted for reasons of fraudulent reporting comprised less than 0.005 per cent of the total sample set.

I certainly agree that it is very likely that such figures represent only the tip of the iceberg. However, it seems to me that there is a pressing need to quantify such speculations. In addition, almost all such studies have focused on biomedicine and related fields. There is a great need for similar studies in other fields of science. – Yours, etc,



School of Science

and Computing,

Waterford Institute

of Technology.