Several years ago, I served as a judge for a school science fair. The cafeteria and gymnasium were filled wall-to-wall with mason jars of bugs and plants, model volcanoes and solar systems, bells and buzzers, trays of colored water and, inevitably, all too many experiments testing the relative strength of paper towels. (These were handy companions, though, for the experiments that involved trays of colored water.)

This experience got me thinking: What really constitutes a good science fair project? Must it be an experiment? And what role do parents have in helping their kids put one together?

My conclusion is that a good science fair project can be an experiment, but it also can be a demonstration or a report. The real criterion is that it should be a stretch, but still fun. To make that happen, parents need to offer plenty of guidance without taking over.

The main thing for parents to focus on is the unique educational experience a science fair offers. It's an opportunity for a child to work on an extended project, with special attention and involvement from teachers and parents. At enlightened schools, every participant is a winner, and every student experiences the pride of showing off his or her project to visiting parents, teachers, and classmates. There aren't many other opportunities for most children to really delve into a project and then share how much they've learned.

But a science fair can also be a source of anxiety for parents whose scientific self-confidence is shaky, and it can loom forbiddingly large to a child unused to projects that take more than an evening. I recall being much more nervous about my fourth grade science fair project than I was about my thesis defense many years later.

The key for parents is to take it easy. Science fair projects don't have to be Science-with-a-capital-S. They don't require a lot of scientific input from parents. The key for both children and parents is to start early, seek out the resources you need, and above all focus on the shared project and not on the outcome. I promise: the Nobel Prize Committee will not be watching.

Finding a topic

First, you need a topic. Your guidance is very important here. It takes more than luck to find a topic that is of interest to your child, doable, portable and yields an interesting outcome.

There are several ways to go about this. One is simply to look around the house for some question that piques your child's interest. If you start in the kitchen, you are at considerable risk of winding up with "which paper towel is the strongest" again. But, you could also resolve once and for all whether it's really true that hot water freezes faster than cold water. Or you could find out how Jell-O solidifies over time as it cools with the use of a thermometer, a stopwatch and some marbles (make sure you stir it before each test to keep the temperature uniform). Or, if your refrigerator is like mine, there's always "which foods do molds prefer?"

There's also nothing wrong with tidying up that rock collection (okay, the rock box,) and labeling each rock with its name and country of origin and an interesting factoid about its composition or use. An annotated collection (of rocks, insects, leaves or other natural items) can be a good science fair project, providing the science fair will be adequately secure or that if some items vanish it won't be too upsetting. Models, posters, or illustrated reports about topics in science and nature are also fine, as long as the materials and resources used are different and more exciting than conventional homework.

Personally, however, I favor science fair projects that are either functioning demonstrations of how things work (technology), or experiments (science).

There's an almost infinite range of possibilities within these two categories. Young botanists can study how long it takes seeds to germinate in pots maintained at different levels of moisture and warmth. Future physicists might be attracted to a project that measures how much a magnet twitches when a current is momentarily run through a nearby coil of wire, depending on how close it is. Budding engineers can explore electricity by hooking up a table-top doorbell system with a battery, a switch, some aptly-named bell wire and a low-voltage buzzer from a local hardware store. You and your child can build a buzzer from scratch, too: several current "science projects for kids" books explain how.

It's a rare attic or garage that can't cough up the substance of a "how it works" demonstration, usually a gadget forced into retirement by a smaller or more electronic replacement. Manual typewriters, hand-operable sewing machines, Instamatic cameras, loudspeakers, and dial-type thermostats are easy for a parent-child team to take apart. They can still be operated in their taken-apart state, and their workings can be understood. If you need some tips on what's going on inside, see one of those illustrated "how things work" books. Televisions, computers and other electronics are best avoided -- you can't safely plug them in when taken apart and, even if you could, they're just as mysterious on the inside as on the outside.

One of my favorite projects was, "How does a tape recorder work?" This required the sacrifice of a cassette recorder, with the plastic chipped away or the tape head wrenched up so that a strip of cassette tape could be guided across it by hand. Voices could be recorded on the tape and played back faster, slower or even backwards, and no one who tried it could ever see a cassette recorder as a mystery again.

If no science projects come to mind, you can go looking for ideas in books. (See "Pages of Project Ideas," below) Many science fair books offer general advice, but only a few describe actual projects. The problem with general-advice books is that, for better or for worse, they'll remind you of how it felt to be in science class and you'll soon feel deeply inferior if you haven't helped your child come up with a project that has a Hypothesis, Experiment and Conclusions. The Scientific Method is a great paradigm, and I'll have a few words to say about making it accessible. However, students can learn just as much by preparing a demonstration of how something works as they can from an experiment.

What's fair in a fair

For all science fair projects, including demonstrations, much of the educational benefit comes when the child must effectively explain the project in words and pictures. One way to avoid frustration is to save the posterboard for last. Plan the exhibit on paper first. Help your child draft paragraphs explaining the project, making sure that viewers get the information they will need to understand the project. Your most important role as critic is to point out what's still confusing or left unsaid. You can expect a lot of resistance to change once the marker hits the posterboard.

Now, I have to warn you that if your science fair is the kind with judges who take the whole thing very seriously, they're going to be looking for an experiment. Your child won't win with a demonstration of how those pump-spray bottles work, or how to build a bird's nest, or even an illustrated report on the introduction of exotic species into new ecosystems. But these projects are valuable, too, and part of your job as a parent is to help your child to focus on the learning, sharing and achievement, not on the awards.

Nor can you take any of these examples and make it into an experiment. You simply can't force a project into the Scientific Method if it doesn't fit. Doing so makes science seem rule-bound and formal, when it isn't really that way at all.

If you are working with something that is more experimental in nature, keep in mind that a "hypothesis" is not a guess. If you are testing what kind of popcorn pops largest, a hypothesis like, "I think the one with the yellowest kernels will pop largest," seems to children an unnecessary formality. And they are absolutely right.

But, with your help, your child can look a little deeper and create a real experiment. Then the hypothesis will be meaningful. Perhaps moisture in the kernels makes popcorn pop. That's a theory--it tries to explain why. The corresponding hypothesis would be, "Kernels with more moisture will pop larger." Now you have to get popcorn samples with different moisture content that are otherwise identical. That's called experimental design. Twelve samples stored in glass jars for twelve weeks, each left open to dry for a different number of weeks, will do the trick. (I told you to start early.) Pop the same number of kernels from each jar, measure the volume of the popped kernels, and make a graph: volume vertically, weeks of drying horizontally. Is there a trend in the graph? That's your conclusion. Now here's a "pop" quiz: how can you be sure that the kernels left in open jars longer really got drier?

A lot of judges aren't themselves very clear on the scientific method. The other kid's pointless hypothesis, "Earthworms will prefer Cap'n Crunch over Special K," may take the prize, while your child only gets an honorable mention for her excellent but hypothesis-free study relating water level in a Coke bottle to piano notes, which has a great graph correlating distance from bottle-rim to water surface with frequency of the corresponding note. Many real scientific discoveries are made through just such methodical observations that have no specific initial hypothesis.

So focus, instead, on the experience of the science project. If you can keep the time-pressure low and forget about winning, it can be a great focused "together" activity for you and your child. For that matter, why restrict it to once a year? The best scientific education your child could possibly get would be to learn to look at the world in an exploratory way. Can you levitate a magnet with other magnets? Experiment with magnets of different sizes. Is an airfoil-shaped wing essential to flight? See if toy airplanes need one. Does hot water really freeze faster than cold? Try it!

Michael Peshkin is a professor of mechanical engineering at Northwestern University and the father of two children, ages 7 and 10.

Pages of Project Ideas

The following books offer general advice on science fair projects along with many specific project topics:

The Science Fair Project Series, published by Tab Books, provide project ideas and a lot of detailed instructions. These books include Botany: 49 Science Fair Projects, Earth Science: 49 Science Fair Projects, and so on throughout the various fields of scientific inquiry, and are designed for sixth through ninth graders.

The Science Fair Series by Janice VanCleave (Wiley, 1993) offers many appropriate projects for younger children.

The Complete Science Fair Handbook by Anthony Fredricks and Isaac Asimov (Good Year Books, 1990) lists project ideas for grades 4 through 8, but offers fewer details.

Mr. Wizard's Experiments for Young Scientists by Don Herbert (Doubleday, 1990) is another useful resource.

Or call the Chicago Academy of Sciences for details on their Science Fair Workshop and lists of science fair project topics. [updated Sept-2000: For our on-site Science for Families science fair workshops contact the Registration Coordinator at (773)755-5111 or visit our web site at http://www.chias.org/education/edmfami.html For a fee, we also do offsite workshops at schools, libraries, community centers etc. Anyone interested in these workshops should contact the Manager for Science Outreach at (773)549-0606 x3069.]