It's the conclusion of Science Fair Week on the OhmsBlog! Yesterday's post is available here.

---

Another science fair has come and gone. Much like last year, I've decided to write some general observations about what I saw.

1. Originality - Unfortunately there were projects that lacked originality. When I say "originality," I don't mean "gee-whiz I've never thought of that before" originality. I'm referring to the creation of original content. Downloading a bunch of stuff off the internet does not make a science fair project. The best science fair projects have a core purpose, and everything that the student does to obtain conclusions to that purpose should be the student's own work. I saw lots of this last year, too.
   
2. Methodology - There were lots of methodology holes, some more significant than others. As a student it's easy to overlook something. It's useful to have some additional pairs of eyes take a look at the project to identify oversights. Confounding variables sometimes work their way into projects and they can call the results into question. Another issue was that some projects strayed too far from the three allowed project types: experiment, innovation, and study. Notice the absence of the word "demonstration" from that list. Students can demonstrate something as part of their presentation, but it should be related to the experiment/innovation/study. A demonstration on its own is not a science fair project. As another recommendation, judges want to know why a student did something as much as they want to know what and how. Every decision that is made over the course of the project needs to be justified reasonably.
   
3. Analysis - It is important to think carefully about what kind of analysis needs to be done on the data to make it applicable to the purpose. Sometimes taking an average isn't good enough.
   
4. Bibliography - Again, Wikipedia citations appeared all over the place. Where are the advisers to teach students that this is not a credible source of information for scientific enquiry? As another recommendation, I'd suggest that any claims made in a project that are not proved via the project itself must be cited.