Elsevier and Bad Information

December 22, 2014

Years and years ago, a unit of the Courier Journal & Louisville Times created the Business Dateline database. As far as I know, it was the first full text online database to feature corrections. The team believed that most online content contained flaws, and neither the database producers, the publishers, nor the online distributions like LexisNexis invested much effort in accuracy. How many databases followed in our footsteps? Well, not too many. At one time it was exactly zero. But people perceive information from a computer as accurate, based on studies we did at the newspaper and subsequently as part of ArnoldIT’s work.

Flash forward to our go go now. The worm, after several decades, may be turning, albeit slowly. Navigate to “Elsevier Retracting 16 Papers for Faked Peer Review.” Assuming the write up was itself accurate, I noted this passage:

We consider ourselves to have an important role in prevention. We try to put a positive tone to our education material, so it’s not a draconian “we will catch you” – it’s also about the importance of research integrity for science, the perception of science with taxpayers…there are a lot of rewards for doing this the right way.

The questions in my mind are:

  • How many errors are in the LexisNexis online file? What steps are being taken to remove the ones known to be incorrect; for example, technical papers with flawed information?
  • How will Elsevier alert its customers that some information may be inaccurate?
  • What process is in place for other Elsevier properties to correct, minimize, and eliminate errors in print and online content?

I can imagine myself in a meeting with Elsevier’s senior management. My task is to propose specific measures to ensure quality, accuracy, and timeliness in Elsevier’s products. I am not sure my suggestions will be ones that generate a great deal of enthusiasm. Hopefully, I am incorrect.

Stephen E Arnold, December 22, 2014

Comments

Comments are closed.

  • Archives

  • Recent Posts

  • Meta