Five simple tips for readers of science journalism

March 13, 2007 by Lennart Kiil

1. Make sure results and conclusions presented to you have already been published in serious scientific peer reviewed journals.

Beware. In recent years it has become popular for public relations departments to send out press releases before publication has taken place in the journals.

2. Remember correlation does not necessarily imply causation.

This one is obvious, but there are still plenty of examples in science journalism showing that even some professional writers are not keenly enough aware of this.

3. Be aware of evaluative and normative language where such cannot be defended.

Science, on definition, cannot talk about how things ought to be. When you see expressions like 'too much' or 'too little' and so on, your are basically just looking at the prejudices of the author.

4. Sensational results or provocative angling?

Check sources and sponsors. There is a lot of sponsored research of high quality, but it never hurts to know who is behind.

5. Can the conclusions safely be drawn on the basis of the results? Are other and simpler conclusions possible on the same set of data?

Conclusions drawn on a specific data set often has a lot to do with the assumptions made in the study. Very often communicators of science fail to tell that part of the story.