I was a lowly graduate student when I judged my first high school science fair, wondering how I, a 22-year-old who couldn't sit through the first 3 minutes of a scientific seminar without becoming irreconcilably confused (12 years later, I'm up to 8 minutes), could be remotely qualified to judge someone else's research. And this wasn't just any high school science fair—it was the big, end-of-year science fair at the region's foremost magnet school for mathematics and science.

And then I saw the students' projects. There's no kind way to say this: The magnet must have had its poles reversed.

Let me give you an example. This is a real, actual project at the science fair, and I'm not exaggerating any aspect of it to make today's youth seem inferior. They do a terrific job of that on their own.

A brother and sister hypothesized that junk food causes bad breath. For 2 weeks, the sister ate only healthy food, and her brother ate only junk food. Every night, they'd breathe in their mother's face, and if a child had good breath, he or she would receive a smiley sticker. If a child had bad breath, she or he got a frowny sticker. At the end of the study period, the boy had more frowny stickers than the girl, so QED, junk food causes bad breath.

Oh yeah. American teenagers are definitely on their way up from 29th place worldwide in scientific literacy. Watch out, Croatia; we're coming for you.


CREDIT: Hal Mayforth
Click the image to enlarge.

Since then, I've judged science fairs at several levels, and while a few stellar projects reveal which kids will themselves sit through graduate seminars someday (nerds!), serious flaws in research methodology are pretty much ubiquitous.

"Now wait a minute," I hear some of you saying, hopefully not out loud because you're at work. "These are children you're talking about! Precious snowflake cupcakes! If anything, we should be learning from them!"

To which I say, sure, great idea. Let's learn from the kids. There's a lot we can learn from science fair projects that we can then apply to our own research.

If you want to perform science that today's youth would deem "awesome" or "gnarly" or "hella vituperative" or whatever the kids say, be sure to obey the following 20 rules, all of which I learned from science fair participants:

  1. Currently, the two hottest fields of scientific research are plant growth when watered with various sodas and the monitoring of human heart rate before and after playing video games.
  2. A number even more important than impact factor is the glitter-to-results ratio. Apparently you can compensate for lack of results by gluing extra glitter to your data. (In academic science, this is known as the Undergraduate With PowerPoint principle.)
  3. Good scientists include figures in their studies. Most figures are photos of the scientists performing the experiments, typically in the kitchen. (Imagine the benefit to our scientific literature if your next journal article included the caption, "Figure 3: Me and Brayden makin' bubbles in the sink!!!!")
  4. Numbers are hard (not to mention impossible to add or subtract without a calculator!), so don't use numbers. Record only qualitative information, which may or may not relate to the effect you wish to evaluate. For example, instead of writing that your plant's height is 6 centimeters, report that it is "kind of droopy" or "green and white."
  5. If something goes wrong when taking a measurement, good Lord, don't try it again; that could take an extra 5 minutes! Instead, report the whole debacle, preferably with a sullen, passive-aggressive attitude that suggests that science is boring and stupid anyway. This is a good lesson for dealing with funding agencies, which are boring and stupid anyway.
  6. Sample size, schmample size. Use one thing of each type. You're done.
  7. Attribute any mistake, mismeasurement, or questionable data to "experimental error." (The term, not the hilarious monthly column).
  8. Begin your work the night before your grant report is due.
  9. Remember, the most important part of your research is identifying which is the independent variable and which is the dependent variable. Most reputable scientific journals will look for this first because it's worth 5 points.
  10. Don't worry about subjectivity and the possibility of experimenter bias: The experimenter is you, and you're awesome! Have a participant ribbon.
  11. Your experiment is the most important thing in the world, and once you draw a conclusion, the world should take heed. Remember this as you write your "Results and Conclusions" section. It is impossible to overstate the importance of your work.
  12. Stretch those research-grant dollars by drawing broad, definitive conclusions from a single data point.
  13. Background research can be performed using four easy steps:
    1. Search Wikipedia.
    2. Print Wikipedia page.
    3. Staple the printout.
    4. Ensure that you do not accidentally read anything.
  14. Proper spelling and grammar are optional. But you knew this already: You're a scientist.
  15. The most impressive and scientifically sound research is always performed by parents. So call your parents. See if they've got some free time this weekend. (What, Mister Big Shot, you're too busy to call your parents?)
  16. Instead of trying something new, perform the same research that your colleagues in the next lab are working on. That way, you're likely to get the same result, and if you make the same mistakes, no one loses too many points.
  17. The best titles for scientific reports are in GIGANTIC FONTS.
  18. When questioned about your research, recite a memorized series of sentences with meanings that you're unsure of. Hey, neat, you know that one already.
  19. If your research involves something biohazardous, such as growing mold, display the mold prominently, because everyone will want to see that.
  20. If you can't think of a good topic for your research, look for books like Seven Simple First-Place R01 Grants That Earned an A+ and 25 Awesome Specific Aims That Are Sure to Get Funded.

Other than the bad breath experiment, and another in which two seniors made Jell-O (that was the whole experiment: They concluded that they could indeed make delicious Jell-O), and one more in which a third grader accidentally misspelled a section title "Resluts," the saddest experiment I've seen at a science fair actually says a lot about the problems with modern science.

A middle school student wanted to test whether darker colors absorb heat. She used food dye and water to make clear ice cubes and purple ice cubes, then roasted each with a hair dryer and timed how long it took to melt. On her first try, the clear ice cube melted faster (probably because her ice cubes weren't necessarily the same size, nor was the hair dryer held the same distance from each cube). She tried again and the same thing happened. On her third try, the purple ice cube melted faster. In her "Conclusion" section, she wrote: "It took me three tries to get the right answer." (I dare you to write that in your next Science submission.)

If there's one lesson I hope the kids take from the science fair, it's the one the ice cube kid failed to learn: A science experiment doesn't have a "right answer."

I didn't correct her. I probably should have, but my judging sheet only allowed me to give a numerical score, which is obviously an insufficient way of adequately critiquing even the most basic scientific project.

Yes, just like grant reviews.

Adam Ruben, Ph.D., is a practicing scientist and the author of Surviving Your Stupid, Stupid Decision to Go to Grad School.

10.1126/science.caredit.a1400078