Peer Review and Supported Documents
My talk with Tim Siftar (librarian at Drexel's Hagerty Library) last week got me thinking that I missed the target a bit with my blog assignment for my CHEM 242 Organic Chemistry class. I wanted to make sure that they used "credible" references so I asked them to use CiteULike, which strongly encourages them to use journal references by having specific fields to fill in like Journal Name, Year, Volume, Page, Authors, etc.
Certainly there is a tremendous improvement in the references used by students over the last term. This is a step in the right direction. They need to learn how to spot a "peer reviewed" article. But I think a skill that is even more important is the ability to assign a probability of authenticity to a document found out of context.
In chemistry, that means that every statement expressed as a fact has a reference. Every conclusion is linked to experimental data. Opinions and speculations don't need a reference - the author is the reference.
With these criteria, I think that (if done with care) blog posts of scientific research are potentially easier to authenticate than a paper in a printed journal because every statement can be supported by a hyperlink that can be immediately verified. Every conclusion can be supported by online data. It will be interesting to see how close we can get to this with the two students working in my lab and blogging about it this summer.
When the supporting information is not immediately available, peer review may not work the way many assume it does. For example, here is an article generated by the random linking of buzzwords that got accepted for publication after being reviewed by three reviewers.
I am not saying that peer review is of no value. If you work in a field for some time you know which journals are most reliable. In that capacity peer-review probably works fairly well in letting through articles of a certain specification. But how do those reviewers authenticate the manuscripts they receive for publication in those journals? Those are the skills I want my students to learn.
One of the key distinctions I want my students to make is between an apparently authoritative reference and an authenticable one. For example here is a dot-gov medical site with plenty of information but absolutely no references. Or if these sites do have references they are often all lumped at the end making any single statement effectively untraceable.
Based on these considerations I have attempted to further clarify my class blog assignment.
Certainly there is a tremendous improvement in the references used by students over the last term. This is a step in the right direction. They need to learn how to spot a "peer reviewed" article. But I think a skill that is even more important is the ability to assign a probability of authenticity to a document found out of context.
In chemistry, that means that every statement expressed as a fact has a reference. Every conclusion is linked to experimental data. Opinions and speculations don't need a reference - the author is the reference.
With these criteria, I think that (if done with care) blog posts of scientific research are potentially easier to authenticate than a paper in a printed journal because every statement can be supported by a hyperlink that can be immediately verified. Every conclusion can be supported by online data. It will be interesting to see how close we can get to this with the two students working in my lab and blogging about it this summer.
When the supporting information is not immediately available, peer review may not work the way many assume it does. For example, here is an article generated by the random linking of buzzwords that got accepted for publication after being reviewed by three reviewers.
I am not saying that peer review is of no value. If you work in a field for some time you know which journals are most reliable. In that capacity peer-review probably works fairly well in letting through articles of a certain specification. But how do those reviewers authenticate the manuscripts they receive for publication in those journals? Those are the skills I want my students to learn.
One of the key distinctions I want my students to make is between an apparently authoritative reference and an authenticable one. For example here is a dot-gov medical site with plenty of information but absolutely no references. Or if these sites do have references they are often all lumped at the end making any single statement effectively untraceable.
Based on these considerations I have attempted to further clarify my class blog assignment.
posted by Jean-Claude Bradley at
1:08 PM
0 Comments:
Post a Comment
<< Home