From Chemometrics to Genealogy

Bernard Vandeginste

  Home :: Contact :: Syndication  :: Login
  34 Posts :: 25 Stories :: 87 Comments :: 0 Trackbacks

News

Welcome to the weblog of Bernard Vandeginste. I hope you will enjoy my articles on events (gebeurtenissen), professional items in the field of chemometrics and also on my hobby, the genealogy search of my ancesters.

Follow me on twitter

Article Categories

Archives

Post Categories

My Blog

My Blogs I read

My Chemometrics links

My Genealogy links

My Qualimetrics links

“ Een Kwestie van Geloof” (A Matter of Belief) is the header of R. de Jong's article on fraud in science, published in C2W Life Sciences (July 2009), a magazine for members of the Royal Dutch Society of Chemistry. According to Prof. Emeritus Vliegenthart quoted in this article, it appears that areas of spectacular discoveries, such as biochemistry and medicine are particularly vulnerable for fraud. Famous cases of Jan Hendrik Schoen and Catherine Verfaille demonstrate that researchers may succomb to the pressure of publishing, e.g. in order to secure their funding. But they are not the only ones. On the average one scientist out of three admits to occasionally delete outliers without valid reason. A disturbing two percent of the scientists even admits to have fabricated data. Indeed, modern data and image processing techniques allow researchers to extract information hidden in blur and submerged in non-informative data or even to impute missing values. Of course the borderline between data enhancement and manipulation is very thin and very difficult to be detected by peer review. “ And even when peers did a good job by rejecting a paper, e.g. one by Schoen, they were overruled by the editors who did not want to miss an opportunity of a break through” says Prof. de Boer. By experience I know the tension between the experimentator and data analyst to get an effect or hypothesis confirmed. Here too, the borderline between what is acceptable and what is not, is very thin. Good Laboratory Practice (GLP) was once implemented to combat fraud in routine and non-routine analysis. Scientific research is governed by the simple rule that research should be verifiable and repeatable. Therefore it may take years before a fraudulous researcher, such a Schoen is caught. Are we so naive to believe that scientists are “ clean”?   
posted on Thursday, July 23, 2009 10:46 PM

Feedback

No comments posted yet.

Post Comment

Title  
Name  
Url
Comment   

ATTENTION: the code you need to copy is CaSe SeNsItIvE and is required to prevent spam.
Enter the code you see: