How Your Kid Can Make a Good Science Fair Project
A long time ago, I proposed that we reserve the word “science” for the body of facts that scientists have discovered and “testing and discovery” as a separate term for the process scientists use to verify old facts or uncover new ones. The latter is what science fairs are supposed to be about.
The Atlantic has a piece on the many issues with the science fairs as they are conducted today. It is too harsh in its condemnation. It treats science fairs as a driver of education inequality as if rich people don’t help their kids with other aspects of their educations.
The piece, however, does identify some things that are broken. Students get no direction on what testing and discovery are and how to conduct a project. I think this is by and large because teachers don’t know themselves. A lot of university professors lack an intuitive grasp of it. I review academic papers that seem to follow the same motions of a typical science fair project. They present a hypothesis, collect some data, and find that the data is consistent with the hypothesis. But those are the most superficial aspects of testing and discovery. You can do each of those things without having found out anything.
Let’s start off with an example from the article of what not to do.
…I spent over an hour building a model of DNA out of licorice and gummy bears.
That’s an illustration, not an experiment. Illustrations have their own merits, but it doesn’t qualify as any of the three things the author’s child was instructed to do:
developing a hypothesis and conducting an experiment to test that theory, inventing something new, or researching “something specific.”
Of these options, only the first is testing and discovery and is the type of project that is the focus of this post. The second is engineering, which is awesome in its own way, but I won’t discuss it here. The third is for people who can’t fathom doing either of the first two and just want to get the whole ordeal over with by doing a book report.
Here’s an example of testing and discovery your kid has probably already done. It gets at the essence of what testing and discovery is:
Your kid wonders whether a particular room is empty or not and then opens the door to check.
The reasons this qualifies as good testing and discovery is
- Your kid was genuinely curious.
- The result was truly unknown by your kid until the observation was made.
- The observation was definitive and distinguished between the possibilities of the room either being empty or not.
That’s it. That is the heart of testing and discovery. Do not discount it.
I realize this won’t fly as a submission to the science fair. However, that’s only because no one else cares. If the room was at Area 51, you’d be in business since people already care about that. Or you can find a question no one thought to ask, but having asked it they now care. For example, “does the school principal have any communicable diseases?” People will definitely stop at your project and look at the results. Find evidence to distinguish between the multiple possibilities posed by an interesting question, and your project will be the talk of the fair.
Let’s look at a common, terrible project I’ve seen many times at science fairs and see what we can do to fix it.
You take a battery, some wire, and an iron nail. You wrap the wire around the nail, connect it to the battery, and observe whether the nail becomes magnetic. Your hypothesis is something like:
A nail, battery, and wire can be used as a magnet.
And since it works to pick up a paperclip, you declare your hypothesis verified and make your poster.
This is not testing and discovery. This project doesn’t stem from genuine curiosity. Your kid didn’t start off asking “I wonder if a nail can be made into a magnet”. You or some book said that it could, and then your kid followed the recipe.
The experiment doesn’t differentiate between possible ways the world could be. Let’s say the nail doesn’t become a magnet. Is your hypothesis now falsified? No. You haven’t learned anything from your failure.
Testing and discovery is about making multiple hypotheses about what the world could be and then conducting an experiment to find out which way the world actually is. Go ahead and build the magnet with your kid as a preliminary exercise and ask what would happen if you tried a different kind of nail. Ask what would happen if you tried it outside when it it is freezing. Ask what would happen if you tried different sizes of batteries or different kinds of wires. Does the kind of metal the wire is made out of matter? You are probing the boundaries of knowledge.
Eventually, you will get an “I’m not sure.” Either that, or you will give up and move to another topic and receive an “I’m not sure” with that.
Once you have an I’m not sure, you have two options. The ideal one is that you research the answer in books. Then, you again probe for the next knowledge boundary. If you do this enough times, you will have a project that is not only testing and discovery, but also novel research that perhaps could be published in a science journal.
The more realistic option is that you ask your child to elaborate on “I’m not sure.” What are the possibilities? She may come back with “Maybe bigger nails make better magnets” or “Maybe nails with more iron in them make better magnets”. These are good hypotheses since they are testable, and your child doesn’t already known the answer. You can then try different nails and see whether we live in Universe A where bigger nails make better magnets or Universe B where they don’t.
Does anyone else care about the results? Of course! Electromagnets have a lot of uses. If you could make a stronger one or weaker one, that could be helpful.
That’s all there is to a good science fair project. To actually win competitions, you child will need to put in more work and do more research to fill in knowledge gaps, but the essential process is the same and will remain the same more or less forever.
I think we discussed this on your previous post, but I am a big fan of rebranding most “science education” as “inquiry skills”. Testing and discovery are two such inquiry skills. I am increasingly uncomfortable with science being a discrete subject for elementary aged students. The fact is, they are — or should be — using inquiry skills throughout their day. These should be taught and employed both explicitly and implicitly. The body of knowledge can come later: most of it is rote anyway and with well-developed inquiry skills, they will be better positioned to access it on their own.
I feel similarly about social studies/history. We should be focusing on research and study skills (which, in a way, are just a specialized form of inquiry skills).
We need to stop teaching kids stuff and start teaching them how to learn.Report
Excellent post, vikram, excellent and wholeheartedly seconded comment, @kazzy
I would just add stressing the importance of teaching children how to observe and to be aware observing the world around them. With scientific inquiry, we build in appreciation of the problems and shortcomings and distortions introduced by the limits of our equipment for observing. Looking at light particles with electrons, Wm. Bell (I think,) once described as trying to understand a bowling alley by sending bowling balls across the lanes.
Our own selves are subject to observation limits and bias by the nature of our sensory organs and the shape of our thought patterns and the limits of our knowledge; sometimes the greatest difficulty is finding the right question to ask, as demonstrated by the difficult of finding an appropriate project for a good, real science project that uses scientific process.
There certainly is value in learning good modeling; I have beautiful prints of birds on this year’s calendar as a result. The act of modeling can be an act of discovery, leading to questions to investigate, as well as a way of reporting investigation. But modeling alone, is only a part of scientific method, not the sum of it.Report
You inspired me to ask an off-topic question on Sunday; I’m looking for some scientific input. Thanks, Vikram.Report
We need to stop teaching kids stuff and start teaching them how to learn.
This was, to me, the big difference between school & college. Schools taught facts and some skills (like math), colleges taught how to learn & think critically. Given the world pre-internet, I can see the rationale for this distinction.
Nowadays, not so much.Report
@oscar-gordon
I read an interesting article recently in which a teacher railed against a vision of next generation learning wherein “super teachers” transmit via video came expertly delivered lessons while on site “behavior management specialists” handle the students.
Now, I’ve got my reservations about such a future. But the way this guy (or gal? I don’t remember) spoke… it was full of bitterness about being replaced as the sole possessor of society’s knowledge. It really pissed me off. Yes, the democratization of access to knowledge isn’t without its warts, but it is a hugely positive development that all of us — teachers especially! — should welcome.
We’ve updated Bloom’s Taxonomy. If you haven’t seen it yet, I suggest Googling it. It is a really positive shift.Report
This?
http://en.wikipedia.org/wiki/Bloom%27s_taxonomy
Updated how (talk me like I’m stupid).Report
Here things are called “science and engineering” fairs. This has pluses and minuses. On the plus side, it opens things up and can leverage the popularity of the maker movement. On the minus side, it can detract from the experimental design aspect that has been the traditional emphasis. I haven’t decided which I favor — although the push to include engineering, and allow software to be featured, isn’t going to go away as long as Intel is putting up the bucks for the big prizes at the top levels. For me, the fundamental question that I don’t know the answer to is, “Is trial-and-error solution of an engineering problem, possibly hands-on or possibly through simulation, as valuable a learning tool as an experiment to determine if nail size affects magnetic field strength?”Report
@michael-cain
Trial & Error design is not as valuable as the experiment. It’s not without value, just not as valuable. However, there is no reason design has to happen by trial & error, or random simulation. There are established, well used scientific approaches to engineering design, both in simulation and prototype design. Design of Experiments itself is a fascinating field that requires a lot of critical thinking to setup well.Report
I’d rather not get into which is more valuable here. Science and engineering are both awesome. Additionally, it seems unlikely that someone who is not good at trial-and-error to get something working is likely to grow up to be a good scientist. Similarly, knowing how knowledge is developed is likely to benefit an engineer even if it’s not strictly necessary.Report
I think schools do a generally poor job of explaining to children what research actually is. I have a nephew who really struggled with middle-school “research paper” assignments that basically required him to : 1) Pick a topic; 2) Read three articles on the topic; 3) Summarize the three articles in your own words. He couldn’t understand what the underlying goal was; why he couldn’t just quote the source material outright; and where to draw the line between “research” and plagiarism. He would paraphrase complex conclusions that no child could have come up with and get dinged for plagiarism even as the assignment required paraphrasing historical facts that no child could have come up with. Eventually, he got the hang of it in a sort of Pavlovian way – this type of statement costs you points, this type doesn’t. And coupled that with the age-old skill of reading a paragraph, closing the book for a minute, and then rewriting the paragraph using dumber words. By the end, the whole exercise had depressingly little to do with research and was primarily about demonstrating that you had done X hours of reading and writing.
I don’t really know how to fix it without requiring a lot of parental engagement, but it would at least be helpful to explicitly state the goal of this kind of project and distinguish it from the way research is traditionally done.Report
My wife teaches high school AP social studies. I know that there is a big push to use primary sources, but in practice this seems to mean using pre-digested anthologies of primary source material. This is better than summarizing secondary sources, but not a great deal better.Report
That brought back painful memories. On the plus side, I guess you at least get some experience writing. But that’s about it.Report
Really great post. My personal problem with science fairs back tin the day was that I lacked the imagination and the base of knowledge to ask interesting questions, and I feel this is a fairly standard occurrence among the 15 year old set, even those with a math/science bent to their intellects. Plus, asking interesting questions becomes more difficult if your dealing with biologically, chemically, or psychological/sociologically dangerous substances – which for the last case, was all of them. (every science teacher I’ve ever had counseled explicitly or implicitly against anything that would require an IRB or its school level equivalent. Even for most straightforward types of experiments, the equipment available to a student at a regular, non-STEM magnet public high school is quite limited – and even if present, is still a subject of unknown unknowns. (i.e. even if the school has a henway in storage, most students won’t even think of asking if the school has one due to their aforementioned lack of educational background).
The thing I like about “and engineering” fairs that Michael Cain mentions is that it directly fixes another deficiency in most high school science education – the general lack of tinkering and machine building. That sort of thing, is, in my experience, often relegated to shop class, for students on a different track. Doing some welding in physics class to do some kinematic experiments is a worthwhile experience on both ends of the educational spectrum (i.e. practical and theoretical). Perhaps the success of Mythbusters and fame of the Pumpkin Chuckin’ winners have made this fusion more common since my high school days?Report
I actually do wonder whether reading this post would have been helpful when I was 15. I’m not sure I was aware then of how much we don’t know. I think I similarly felt that everything worthwhile requires highly specialized equipment. There are a lot more basic questions that don’t get asked though.
IRBs suck, but I wouldn’t write off any experiment that involved them. As a business researcher, I particularly hate IRBs because there is no possible way that any of my research could have caused anyone any issues, but I have to file the same paperwork anyway. But they are passable if you are willing to go through the paperwork.
In the fairs I’ve judged, “engineering” was a single category. It was always the largest. I think maybe they could have segmented it to be more fair to the students competing in it.Report
But that’s the catch. Asking a 15 year old to do paperwork is in fact the most realistic simulation of and best real-world training on how SCIENCE! actually operates, but it’s easily avoided if you’re just, in essence, a hobbyist, and not seeking grant money. (or grad school credit).Report
Good point. I certainly never did any behavioral projects as a result.
Someday I might write a rant about IRBs, but I think there are a lot of them floating around the internet already.Report
My favorite “Science Fairs” is still the FIRST program.Report
This project doesn’t stem from genuine curiosity.
That’s because no one gives a shit. Kids just do science fair projects because their teachers make them.Report
The point of a science fair project is not to ask an interesting question. It’s to ask any question and then correctly follow the scientific method in detail — observe, hypothesize, test. Throwing eggs off the roof and measuring how high they bounce can win the science fair *if* you do a sufficiently good job of eliminating the independent variables, recording the results, showing how they do (or don’t) support your hypothesis, and writing it all up.Report
I’d say that depends heavily on the level of the fair in question. A grade school or middle school fair could ask some basic questions & test them. A high school fair should be a fair bit more sophisticated.Report
I wanted to add a person note, as someone who won a state science fair and placed at the ISEF in high-school. The idea that rich kids are buying their way to the top was not reflective of what I saw. At least at the senior level, everyone who placed had clearly done their own work. They knew what they were talking about. And they were basically the kind of boundlessly energetic young minds you’d expect to have a couple Ivys as safety schools. From what I remember, the parents were very engaged but not particularly wealthy or intellectually snobbish (which surprised my parents who were both academics).
All that said, the dynastic element was definitely there. Most of the winners had been involved in science fairs for many years nearly all had already won several middle-school trophies. This was typically the result of strong infrastructure: either a serious science fair elective in school (as the article describes); or, more typically, mentorship programs with a local college/university. In fact, having some kind of academic mentor was the norm, and meant that – for better or worse – the fair was basically a dumping ground for academic PIs with some extra ideas lying around and no college students to implement them.
Anyway, in contrast to the linked article, I totally understand why Intel and Google are funding these fairs. Aside from good PR, it is actually a very good way to identify a tiny fraction of young students who have the motivation and infrastructure to do college-level research. It is not, however, a good way to introduce the general young student population to scientific research.Report
I think there’s a common misconception on the two generally accepted sides of the aisle that the claim in question is that the privileged rich buy their kids into success in a way that doesn’t actually involve their kids.
A lot of media is currently painting a narrative in which rich kids are pushed and shoved into being rich themselves. In this model, they have no agency. In actual life, having rich parents just means you get access to more stuff. Having parents who can afford piano lessons doesn’t mean you will actually learn to play the piano, and if you do learn, it doesn’t mean the accomplishment isn’t yours.
At least until the rich can afford to do skill uploads like in the Matrix.Report