There are several sections to this revamped document. It begins with some information on how to plan a science fair project, some optional thoughts for advanced students about the seven types of science you may encounter when you are thinking about your project, a huge list of ideas for projects you can do, and a bit of help in approaching just a few of the ideas.

Background: this has existed for some time as a Web page, and you may even be reading it as a Web page still, because the HTML code that has been written as an e-book is also being used as an expanded replacement for the original Web page. There is also a .PDF file which has most of the topic ideas that are listed here, but it is not quite as complete as this version because it got too complex trying to manage parallel versions. The details of where to find these alternatives can be found at the end of this Web page or e-book, whichever you are looking at.

Use the PDF file as a classroom handout (it is 22 pages of fine print in two columns to save paper), or for preference, rely on the Web or e-book version you see here. Save the trees!

How to do a science project

Preface

For example, it is likely that the only way you can find out what spiders live in your garden is to go and study them, identifying the species as you go. All the same, that is not a very exciting piece of information to have, but what if you studied a number of yards, found the number of spiders (or the number of spider species) in each yard, and then tried to see what influences any variations.

The whole point of a project is that it should be about something that interests you, it should be driven by curiosity, and it should provide some kind of theory that other people could later go and test somewhere else. And for that to happen, you are going to have to follow some of the basic rules of the scientific method. That is what changes it from a project to a science project.

This material has been written to help you get started, but it is only a beginning.
Back to the contents

Planning your project

Your logbook

Your log is written as you go. There is no need to make rough notes on bits of paper, so you can copy them into the log when you get home. Real logbooks show where they have been written in during rain storms, they have mud stains (or bloodstains), and that is OK. Look after your log, but do not stress out if it suffers some indignity. And if, for some reason, you have to use loose paper, date it, and paste it into the book.

You should keep notes of any interviews or phone calls you make, phone numbers and email addresses, because you never know when you will need to email a contact again - and computer systems can always turn nasty on you. If they are important, you may wish to paste in printed copies of e-mails, but as a rule, this is not necessary. It is worth your while noting when you get an answer, though.

The log will not be a part of your final project presentation, but it should be a part of your submission, an appendix which proves that you actually did the work, and got the results you say you did. The things you write in there should all be dated, so that the record is completely clear, and neatness is not so important as clarity. So get your book, label it, put an address or a phone number on it (it is going to be VERY valuable if it is lost), and start with a list of possible topics, and then move on to a timetable.

Back to the contents

Choosing a topic

Another way is to take something you have always wondered about, even something simple like the shape of a bubble, or why ducks (or steel boats) float, or why a balloon goes BANG! when a pin is stuck in it. The best projects are all curiosity-driven. As you read the news in a newspaper, or watch it on TV, think about some of the human problems you see, and wonder how they might be fixed. Spot a claim in a TV commercial, and wonder if it would really stand up under test.

Try putting different terms in these blanks:

What is the effect of __ a__ on __ b__?

Some of the a terms you could use might be temperature, noise, quenching, design, density, humidity, wind direction, overnight minimum temperature, music, pressure, detergent, water turbidity, acid, oxygen, hot hydrogen . . . Some of the b terms might seed germination, rusting, growth, rotting, grain size, ripening, wave frequency, bird species, flight duration, surface hardness, learning, driver fatigue . . .

Back to the contents

Literature searching for a topic

Right now, I am working on a book about sugar, and I keep running into some seriously weird and odd bits and pieces, simply because I am looking at a lot of books, and even more Web sites about sugar. There are more than 3 million Web sites that mention "sugar" in some way or another, but less than 1% of them have both the word sugar and the word cane, and about 4% mention sugar and beet. For example, I found a site that gave me recipes for fried cat, yellow jacket soup, rat roulade and frozen ramen on a stick, and just a few references to sugar. So I had to adjust the search, and you would have to do the same.

By tweaking the search a bit and looking again, you may come across the fact that sucrose in cut sugar cane is converted to fructose and glucose, so cane has to be crushed and heated within 16 hours of cutting. This immediately raises some questions: is this a problem with beet sugar? How does the cane 'know' how long it is since it was cut? The fact that heating stops the conversion tells us an enzyme is probably involved - why does sugar cane need it, and what makes it start acting?

Of course, many of the questions you wonder about will be too hard for a school level project, but if you know you are going to have to do several projects over several years, you might start with an easy sugar project, and then go to more advanced stuff in later years. I don't know if other plants have enzymes that do the same thing, but I know there is a test that will tell me if the sucrose has been breaking down.

Once you have found a topic, you will need to come back to a further literature search, but now you will be looking for information on a particular topic, rather than an idea for a topic, so you can make your search much tighter. It is a good idea to do an Advanced search with one of the better search engines like Altavista or Google.

Back to the contents

Asking for help in finding a topic

Asking for help on a project is like going for a job: you have to make a good impression. Do it right, just once, and it keeps on paying off. As a science writer, I need to write to busy people all over the world and persuade them to fill me in on their research, and where it might go. Even i have to go by the rules.

For starters, most adults will not be impressed if you email them or ring them up and say something like "I have to find out what goes mouldy faster: white bread or brown bread. What is the answer?" Note the problems with this real example: no please, no courtesy, and absolutely no sign of any effort by the student. That sort of request is usually ignored, and so are e-mails saying "send me everything you have on mammals!".

If you write and say after a few courtesies "I need to compare the rates at which white and brown bread go mouldy: how many times do you think I should run the experiment?", or "Can you tell me the best way of setting up a controlled humidity chamber?", that shows that you are not asking to have all the work done. All the same, expect to get an answer in the form of another question, or a hint, rather than the whole answer, because that is the way scientists are.

In the same way, if you write to somebody saying "I have to do a project, what do you suggest?", that leaves the person written to with no idea of level, age, type of project, or what resources the student has available. It also indicates that you have not read anything, and have not done any searches for yourself.

The sort of question that gets attention says: "I am looking at the effects of diet on flatworm regeneration, but I am worried that there may be variations in the temperature", or "I am trying to see what foods flatworms prefer. I am planning to use cheese and small pieces of liver: do you have any other suggestions?".

To get attention straight away, you need to list the ideas you have already had, because this saves your tame "expert" from wasting time telling you what you know already. You are the one who will benefit from this, so you are the one who must put in the hard yards.

Your project should include printed copies of all emails and letters, so remember that you are also writing for a public audience.

Stranger danger:
Please, as a matter of principle, it is courteous to tell people your age, school year or grade, but you must never provide personal information without your parents being involved, and even then, you need to avoid it. It will hardly ever be a problem, but sadly, you just have to be that extra bit careful, because there are a few loonies out there. Even if you are a great big bruising 17-year-old footballer, there is a standard to be set, and you need to set it for other younger people, so NO DETAILS! If somebody needs to mail you stuff, get them to send it care of your science teacher at school, or your parents at work.

Back to the contents

Coming up with a research question

To take the sand grain example, you are going to need to take a number of samples from the same beach at different times of the year - and by the way, that particular study will probably be varied by the effects of wind blowing finer grains up to the back of the beach. Maybe the answer is to take all samples from two metres below the high tide mark, and from a depth of 20 cm below the surface, or you might sample different depths from one hole, or different points down the beach.

This means that even as you are thinking up a research question, you are actually doing a lot more as well.

Back to the contents

Planning your timetable

The big problem will come when you are doing a longitudinal study, one that runs over a long time. This can make for a complicated timetable, and may mean you need to analyse data in small slabs, and write up bits, all the way through. Word processors are marvellous for this!

But you MUST have a workable timetable, and you MUST stick to it.

Back to the contents

Legal and other drawbacks

Behavioural studies of animals generally need a sample population of 30 individuals or more to be valid. As a general rule, seek advice from medical doctors or veterinarians, teachers, or parents, depending on what you are doing. If other students are to be used as test subjects you need to cover yourself by seeking expert advice. If there is any doubt at all, write a letter to the subjects' parents, explaining what you are doing and why, and get signed parent permission slips. Then, when you hand in your project, you can demonstrate that you have behaved in an ethical and proper way.

Chemistry projects will usually mean that you need to seek safety advice beforehand - and you will probably need to arrange laboratory and glassware access. Adult supervision is a very good idea!

Earth science studies: most science projects take place in the middle of the year, which means during the winter months, you should consider how cold weather will affect any field work you may need to do.

Ecology studies are interesting and easy to find, but doing the measuring of variables is more challenging. Many of these projects require field work and some may take months to complete, so think about the weather you may need to deal with (making notes in rain on a wind-swept beach in winter is no fun!). The other problems you may face: disease from contaminated water, permissions to gain access to sites of special interest, permission to take specimens.

Electrical experiments which use mains power are NOT a good idea, and if you are working with transformers, get some reliable supervision, just to make sure that there is anything wrong with your setup.

Microbiology projects can be a real problem, especially if you are working with pathogens (which generally isn't allowed!), but even if you aren't, they may grow on your culture medium. Be conservative, and get lots of expert advice.

Plant growth projects need to be done during the growing season outside or require greenhouse conditions during the winter. Plan ahead, because plants need to grow for about 1-2 months to get measurable results. You must grow many plants in test or control groups for the results to be valid. A minimum of 30 plants in each test group should be used, with as many in the control group.

Back to the contents

Making the final plans

You will also change your plans when you notice something new and exciting as you are going along, and you may suddenly discover that you have something more interesting to study. Before you make any changes, work out a timetable, and if you have time, change - there is nothing wrong with you changing your topic, but make sure it is justified.

Finding information

Doing an advanced search

Doing a project

Back to the contents

The scientific method

The first thing to be said is that there is no single scientific method, because the way you work will depend on which of the seven types of science you are doing (and most scientific work involves more than one kind of science). Don't worry, though, because they all have certain things in common.

his is a quick, light run over the main points, and anything you read here should apply in just about every case. The simple fact is that the scientific method is a way of setting out to answer interesting and worthwhile questions, and it usually begins with an observation. You may notice something unusual or unexpected, or you may see something operating and wonder how it works, or if it can be made to work better. In other words, your observation leads to a question.

Once you have an interesting question, you can ask other people what they know about the topic, gather some information, and maybe make an intelligent guess about what is going on. Often there may be several things happening at the same time, several variables that may be having an influence, so that your problem becomes one of trying to work out what part each variable plays. This is a key to the scientific method: pinning down the variables one by one, studying them one at a time, and making a logical conclusion, followed by some sensible tests.

What you test, is called a hypothesis (or 'an hypothesis' if you are being really classy). This is a question which has been reworded into a form that can be tested experimentally, taking the form of a prediction of what should happen if you have managed to work out the effect of the variable you are studying.

Next, you need to work out how you would test the hypothesis, what single thing you would change that should bring about a measurable change. You need to write this down, step by step, and you need to think carefully about the control group you will use for comparison. For example, suppose I have a hypothesis that over 20 years, listening to rock music will make people aged 35 go bald. If I watch any sample, with or without rock music, some of them will go bald in that time, and if we have a suitable control group, we will realise that the rock music was not a cause, after all.

The next step is to do the test, a number of times, if that is necessary. And most importantly, this means measuring, because a lot of effects are what we might call dose-dependent, where a bigger 'dose' causes a bigger change, and other effects may have a threshold, and if the 'dose' is less than the threshold value, there may be no effect at all. Poke a sleeping lion very gently, and nothing will happen, but if you poke it hard enough to wake it up (the threshold value) it will probably bite you, and from that point up, the lion's retaliation will probably be dependent on the does of annoyance you gave it. get the idea?

Then you need to analyse the results, which often means using statistical analysis, which we will deal with separately. And after that, you need to come to a conclusion. And that, in simple terms, is what the scientific method is about.

Back to the contents

Measuring things

Recording results

Analysing the results

Summarising results

Reaching conclusions

Presenting your project

That said, check the specifications for the contest you are entering, to see what you are expected to do, because some of the requirements may be different - or the names given to the sections.

Back to the contents

Abstract

Table of contents

Purpose

Hypothesis

In simple terms, your hypothesis is some underlying suspicion that you have about the way things are. It can come from your first observations, from reading, or in some other way, but there must be a good reason why you suspect that the hypothesis could be true.

Back to the contents

Experimental design

The variables you have been working with will probably be in your title. These are the controlled (or manipulated) variables, and the dependent (or responding) variables. There will also be a number of other variables that are held constant, and you need to specify these as well.

In your design, you should specify exactly how you are going to change the manipulated variable, and what changes you will be measuring in the dependent variable. You will also need to specify sample numbers, and details of your control group(s) where these exist, and how many variations on the manipulated variable will be applied to different experimental groups, if you are using more than one.

Your experimental design should also outline any replications that you will be using, so as to make sure that you have not got your result by chance. As well, you need to list what you will measure, including the units you will use. While NASA may continue to use 'British' units (which the British are learning not to use), nobody else does so: there is no excuse for using anything other than metric (SI units).

Back to the contents

Materials used

Procedures

Research report

When you have finished putting this together, you may find it helpful to add introductory and summary paragraphs. The introduction should indicate the structure of the section, and the final summary paragraph should emphasize the main points. Most judges will expect you to use formal language for a report like this. While you may think it is important to communicate to real people, not judges, if you have done a piece of sweet science, it is a pity to miss out on recognition because of the way you packaged the information.

Formal language means no personal pronouns, because the judges usually think it is wrong. There was a time when scientists would write "I put the crystals in a test tube and dissolved them in water", but these days, they write "the crystals were placed in aqueous solution in a 150 mm x 25 mm Pyrex test tube", and that is the style you will have to follow, which is a pity.

Back to the contents

Results

Remember that if you are presenting graphs, the actual data tables should also be available for viewing. If you have entered data tables into a spreadsheet package, you need to print out the spreadsheet.

Back to the contents

Conclusion

On the other hand, if your hypothesis was proved wrong, it may be useful to discuss why that may be. Every piece of science is part of an on-going tradition, and your comments may help the next person to answer the puzzle you set out to solve.

Back to the contents

Acknowledgements

Bibliography

Back to the contents

The seven types of science

Back to the main contents

This was written some years ago as an analysis of some of the different ways people use the word "science". The main aim was to explore the reason why scientists can never agree on what "the scientific method" is, because they are doing different sorts of science. It finished with an analysis of why "creation science" is not science at all, an analysis which I have deleted here because it is irrelevant to the completion of science projects. Find it on the Web, if you really must know how it ended originally.

Some years ago I taught about human evolution at the Australian Museum. My job was to take visiting groups of students over the evidence and show them how the clues were to be seen in all sorts of hominid bits and pieces. I even became quite good at walking like a chimpanzee.

One day I started thinking about the line attributed to Lord Rutherford, that there are two kinds of science, "Physics" and "stamp collecting". I started to list the various things that scientists do in the name of science and concluded that on most days, most scientists use three or four different modes of science.

Before I explain, consider a practical example of natural experimentation, based on that best of all Australian team sports, hockey, which we play on a green surface with a white ball, and 11 players a side.

Novice players quickly learn that a hockey ball is more likely to rise off the ground when you hit it with your front foot well behind the ball. This is called undercutting, and it can be fun when you do it but a lot less fun when somebody does it to you. If you were learning the game you might decide to explore this effect (in other words - experiment).

You would vary the distance from your foot to the ball, how hard you hit it, the position of your hands on the stick. You'd try changing the angle of the stick. Eventually you would find a method which makes the ball lift every time. (You'd probably be sent off the field soon after discovering this, but the way of the empirical scientist was never easy).

This is the popular view of science. Scientists experiment. They "prove" things, and once that's done it's a fact. And if scientists disagree, then whatever the proposition, it's not yet a "fact". This popular view of science is very mistaken. While it is true that the experimental aspects of science are very important, experiments are only one part of the greater science endeavour. In my view, science involves at least seven different "types" of activity. Each type of science can be seen as a tool which has something to offer, and each type has its limits.

Back to the start of this section

Type 1 science: orthodox science or sensu stricto science

Type 1 science is about straight descriptions of unbiased experimental results, with all the observations reported objectively and dispassionately in the literature, or so we are told. This simple view ignores the subjective choices scientists make when they choose what to investigate, and what phenomena they will explore and measure.

In considering the novice hockey player undercutting the ball, I made no mention of hitting the hockey ball left-handed. Hockey players never use the curved side of the stick and even the left-handers have to play right-handed, so a whole set of investigations has been ignored simply because "we don't do things that way".

Type 1 scientists are unlikely to investigate the influences of the paranormal on a chemical reaction. Most people would applaud this decision to ignore superstition and discount the existence of evil spirits and invisible things which make us ill. It seems rational and wholly sensible. Yet it turns out that there are "invisible" things which make us ill - bacteria and viruses come to mind, so perhaps it is not such a good idea to ignore the possibility of a trillion and one demons, all contending to influence our experimental results one way or another. Scientists will ignore the paranormal until they find some variability which cannot be explained any other way. Then they study the effect, find its cause, and always seem to show that it is really normal after all. Still, to the extent that we leave out demons, evil spirits, ESP and the efficacy of prayer from our investigations, Type 1 science is just that little bit subjective.

Most importantly, hard or Type 1 science is governed by rigid rules. We have our framework for thinking about things and we view everything in that framework. It gives us all sorts of propositions we can then test using the principle of falsifiability. On its own hard science is not very creative. It can even be fairly restrictive, but it is very, very powerful because we are encouraged to try to falsify everything.

It works like this: we can disprove any law just by finding a single exception, but no matter how many supporting instances we find, we can never prove the law is true: the next trial might produce just that single exception we have been looking for!

Suppose I claim I can stop snakes from crossing my track in the bush by tapping out coded messages with my toes as I walk along - codes which the snakes interpret as a warning. For twenty kilometres we walk and we see no snakes. Jubilantly I claim success. You, being sensible, say this is a load of twaddle. It proves nothing: sooner or later a snake will appear and where will my precious theory be then, eh? If my theory is wrong, how come we have seen no snakes? I ask.

That is the problem with falsifiability. We can never be sure the vital piece of disproving evidence is not just around the corner. And scientific theories hardly ever involve silly claims about coded signals tapped out to the snakes, so common sense cannot be used to get to the heart of the matter, to expose the fallacy, half as easily.

This leads to a problem. If nothing can be proved, then you cannot do anything, and that would mean absolutely no progress at all. So for practical purposes most scientists accept many things as "proven" or "true", based on their practical expectations. They call their handy assumptions "laws", "rules", "principles", "theories", "hypotheses", "conjectures", and so on.

Now back to falsifiability. If a scientific idea is to be of any use at all, it must give us testable assertions. It must say something like "people of Cornish descent will be more able to wiggle their ears than a random sample of humans", or "from this, we deduce that all cultures will have a creation myth which begins in water and darkness". Any idea which does not give rise to testable assertions and hypotheses is useless and unscientific. It can safely be labelled non-science and discarded.

Type 1 science gives us many testable assertions and hypotheses, once we have a theory to work with. It is just a pity that it is not more use when we are trying to develop new theories. That may be why we use the other types of science as well.

Some pieces of Type 1 science are "softer" than others. When two clever scientists wanted to know whether a fossil known as the "Taung baby" or the "Taung child" was ape or human, they gave it a CAT scan. Fossils are not bone, but rock. Yet deep within the stone that once was bone, the finest traces of structure remain. When the Taung child died, "adult" teeth were still erupting within the skull, and a cut through the skull would have shown us once and for all whether the eruption pattern was that of an ape or a human. But you cannot destroy a "one-off" specimen, so a CAT scan was the next best thing.

This non-destructive test revealed the unerupted teeth, and told us that the Taung baby was neither one thing nor the other. It was betwixt and between, a "missing link", if you like to use that term. Discovering that was hard science, fully replicable by doing the another scan later, or even by a physical slicing of the specimen. The facts were pretty much open to just one interpretation, at least for now.

Back to the start of this section

Type 2 science: interpretations placed on observations

Or to be very cautious, maybe we did not all originate there, but at least we can all trace our ancestry back to one female, who probably was born in Africa. This is an issue I will return to later, when I deal with speculation as a type of scientific activity.

A great deal of evolutionary biology looks back at what has occurred, and says things like "humans happened like this because . . .". We cannot run the experiment again. It takes too long and we do not have control over most of the variables anyhow. History and many social sciences come close to Type 2 science and that may be why the Type 1 scientists do not really like Type 2 science.

When a historian says "The Great War happened because . . .", this is the most reliable interpretation we can offer based on the evidence. If the evidence changes, our view will need to change as well. As with the evolution of humans, we can see the effects, but we do not know all of the causes. If we ran the experiment again, we might not get the "right result" this time. Type 2 science might not be repeatable.

We may be describing, after the event, a unique sequence. If we "ran the tape again", to use Stephen Jay Gould's expression from A Wonderful Life, we might get an entirely different result next time. So while we describe Type 2 events in scientific language, our results are less certainly replicable, at best. In some cases, there is absolutely no prospect of replication at all.

The textbooks used in our schools ignore this form of science, preferring to emphasis the nicely replicable nature of physical science. Interestingly, most of the "experiments" in school science books are precisely the replicable Type 1 exercises which fit the descriptions given at the start of chapter 1 in each of those books. Do the educators feel that Type 2 work is too challenging for school children, or is this a self-fulfilling prophecy? Or is it simply that every textbook carries a large legacy of comfortably familiar material which sometimes sails close to plagiarism?

Soft science is still governed by scientific rules. We have a theoretical framework in which we explore causes and effects, and we apply Ockham's Razor, choosing the simplest available explanation. We may have a certain amount of trouble in coming up with testable, falsifiable propositions, but we can do it. On its own, soft science would not be very creative, but it has more creative potential than hard science. It leads us to begin asking "what if" questions, to ponder about causes, rather than demonstrating them.

One of the most beautiful events in science must have been that moment when Raymond Dart picked the brain cast of the Taung baby out of a box of rocks. A small ape-person's skull, resting on its side, had been partly filled with mud which later solidified to give us an impression of the inside of the skull. One glance at this was all Dart needed to realise that this was the brain of an ape-like animal which walked upright just like us, and not at all like a chimpanzee. That piece of dumb rock spoke to Dart, allowing him to place an interpretation on it, making it Type 2 science, while giving his conclusion all the certainty of the very best Type 1 science.

You can really only appreciate Dart's cleverness if you have seen the original or an exact copy of it, because only then can you realise the importance of what Louis Pasteur called "the prepared mind". I have looked at first-rate Taung casts knowing what I ought to see, and it is still hard to spot. This may be one of the reasons why Dart had trouble getting people to agree with him. But there were other darker reasons why this upstart colonial, this double colonial, an Australian in South Africa, would be ignored by the British Establishment. I will return to that in a moment.

Back to the start of this section

Type 3 science: fraud

Because the intention is to defraud and convince, the fraudulent scientist must produce results which are absolutely consistent with all of the other known scientific principles, rules and discoveries. Most frauds are done for "good scientific reasons", because the fraud feels that the principle he is trying to demonstrate is more important than a mere "scientific method".

Because it pretends to give us real data, fraudulent science is often testable. By definition, fraud is extremely creative, and it is often right, (in which case it is unlikely to ever be found out), but a few frauds are wrong and these can be extremely destructive.

The "Piltdown skull" was concocted from a piece of a human skull and an orang outang's jaw. These had the tell-tale bits that would have identified their true origins broken off. The bones were chemically stained before they were planted at Piltdown. The teeth were filed flat to hide the tell-tale patterns of cusps that would have betrayed their origins. Scientific investigation eventually exposed the fraud, but the fraud sent scientists off on the wrong line of thinking for forty years. That was a wicked fraud, because the false Piltdown picture of a large brain in an ape-like body was used to deny the real truth, revealed by Dart's "Taung baby".

It was also an impressive fraud. I had the chance to examine the original material in 1993, and prepared by foreknowledge of the clumsy tooth-filing, which I had also seen in casts, I could see that the pattern was entirely wrong, but I am willing to say that I, too, would have been fooled by it, because scientists trust each other, at least until something fails to add up.

Scientific fraud involves making up the evidence to support your pet theory, but if your theory is wrong people will eventually notice that your data are impossible and you will be exposed. But if the theory turns out to be right, nobody will ever know what you did, not unless you were careless, that is . . .

Cyril Burt manufactured IQ data and correlations. In the end, somebody noticed that Burt's correlation coefficients never wavered from their original three-decimal-place value, even when more data were added. Everybody expects that calculated correlations should show some random variation so people looked more closely at his data, and the whole of Burt's work on IQ and inheritance was discredited.

Mendel massaged his results, Dulong and Petit concocted their results when they generated their law relating specific heat to atomic weight. Given the fraudulent data that I can demonstrate in their results, they probably faked more than half of the measurements, and fudged the rest like a second-rate physics student. But who cares? Their spurious law was more or less correct, and it allowed chemists to determine atomic weights accurately by electrolysis, ducking around problems caused by valency.

Ptolemy fudges his data and so did Isaac Newton and probably many others who proved accurate in their guesstimation of where the truth lies. We can only detect their naughtiness by careful statistical analysis. They guessed the end result more or less correctly and they did not hold up the progress of science, unlike the Piltdown fraud, whoever he was, and Cyril Burt.

People can engage in scientific fraud with the highest of motives, hoping to save lives or time by sounding an alarm that a drug is safe - or harmful. True believers can mistakenly attempt to pass off a mathematical fraud and still be entirely innocent. To that extent, they fall into Type 3. On the other hand, the technological frauds, the purveyors of snake oil and perpetual motion are a different breed. I will deal with them when I get to pseudo-science.

Back to the start of this section

Type 4 science: fiddling science

Most technology began as Type 4 science, as trial and error variation to find the best temperature to heat the steel of a samurai sword, to find the best way of putting an edge on a stone spear-tip, trying out new fuel additives or investigating a new cake recipe. The alchemists believed the Philosopher's Stone would transmute metals. In modern terms they thought if they added the right stuff to lead, the lead would turn into gold. To us, this seems like nonsense, but given their understanding of chemistry, the alchemists' ideas really made quite a lot of sense.

They were empirical scientists, people who explored whatever worked and twiddled with it. They knew that the base metal copper could be changed by adding black or brown cassiterite, or tinstone, to the mix. Today we would say they made bronze - an alloy of tin and copper - but they thought the metal had changed. They said the black or brown stone had "transmuted" the metal.

In the same way a white "stone", the mineral smithsonite, could be added to copper with charcoal in a sealed crucible to make brass. Both bronze and brass look more like gold than the original copper so it is no wonder they believed it was just a matter of finding the right "stone", in order to make gold from base metals. Today we realise the theory they adopted is all wrong, but it must all have seemed like a good idea at the time. The trouble was they could not control the effects. They could not predict what would happen in a given case. There was no real science to it all. Yet today, no real science happens without a lot of fiddling.

One common type of "fiddling" is statistical analysis of the sort which reveals the sins of Dulong and Petit, and the peccadilloes of Mendel. Traditionally, this sort of analysis is sneered at by those who will not collect stamps. "If your experiment needs statistics, you ought to have done a better experiment" Lord Rutherford is supposed to have said. Yet statistical analysis can reveal the underlying truths in complex situations - the sort of messes that true physicists shy away from. (In fairness to Rutherford - he did take classes in statistics in his later years.)

Fiddling is almost acceptable to the textbook writers. They call it something fancy like "pattern detection", and they look around for nice examples like Balmer's analysis of the "hydrogen lines" or Maria Goeppert Mayer's "magic numbers", which led to the shell model of the atomic nucleus.

Usually though, the textbook authors will avoid any discussion of "Bode's Law", which was neither by Bode, nor a law, but a fascinating mathematical model which predicts the locations of the inner planets of the solar system. (In fact, it was suggested to Bode by J. D. Titus, and so is sometimes called the 'Bode-Titus Law'.)

Like Balmer and Mayer, Bode used data-snooping to find a pattern, and having found a pattern, believed in it. Balmer used his formula to predict an extra line which turned out to be there, while the Bode pattern predicted a planet where Uranus was found, and another planet where the asteroid belt was found, (which is why even today, people speculate that there may be some deeper meaning behind the pattern which gave us "Bode's Law"). We don't know what the true cause of the pattern is but sensible scientists keep an open mind about such things.

Back to the start of this section

Type 5 science: speculation

Speculation is creative science, identifying, designing and planning what is to be observed and described, and then looking around for the unexpected as well. Speculation is usually left out of courses on the scientific method. This is a shame for at different times we need science from at least Types 1, 2, 4 and 5 for fertile science. Speculation is far and away the most creative form of science, and the most enjoyable. Speculation generates hypotheses, and it would be pointless if these hypotheses were not tested. Speculation often fails to produce results, but the occasional successes more than make up for the failures. And to a good scientist, even the failures can be extremely informative.

Speculation is probably at the cutting edge of science, because it is only when scientists decide to reject the "standard model", for whatever reason, that they can look for the evidence which may one day lead us all to reject the current paradigm. Speculation always stays in a scientific framework. We reject or set aside one key assumption which looks a bit shaky, and then explore the consequences. We look for evidence which says in effect, "Yes, you probably can safely reject that assumption, at least for the moment . . .". Always, the speculator is limited to reasonable hypotheses, varying one or maybe two assumptions here and there.

Take the "out-of-Africa" hypothesis. Mitochondrial DNA says we all descended from a single woman in Africa, but the fossil evidence says that the human races were in place and developing long before "mitochondrial Eve" was born. About the only thing the two sides agree on in this debate is that one of them must be wrong. I disagreed when I heard them asserting this at a public debate in Sydney. I have sometimes wondered where the eyebrow ridges went as humans became fully modern. One of the more bizarre pieces of speculation I heard about was that eyebrow ridges helped to keep long hair out of the eyes, but I have been looking at the standard view of female beauty, which requires a "baby face". We humans are, after all, paedomorphic apes - apes with juvenile characteristics.

Just suppose, I mused, a female with a severely baby face arose, maybe in order to accommodate a larger brain but with a face which was by comparison with the old model, stunningly beautiful. Strong ape men would come from miles around to seek out such a beauty, and if her genes were passed on to her offspring, some of her daughters would probably be abducted, and possibly taken over large distances, establishing new outposts of paedomorphism. The people who believe humans arose where they did are faced with one big problem: most humans do not spread over large distances, but now I have an hypothesis, which I will call the "piggyback gene hypothesis" to explain why the eyebrow ridges disappeared in modern humans and why we all seem to be descended from a single African female. And best of all, my hypothesis allows me to explain the observations of the two contending groups who argue about mitochondrial Eve.

So I have a nice hypothesis, which might allow me to write a best-seller, involving chapters of exalted salacity, steaming brimfully with sex, rape, and abduction, a book which will appeal to feminists because it implies that it was they, not men, who made humans smart. Sadly, the more recent evidence from sequencing the Y chromosome makes it all seem a bit unlikely, suggesting that the male line also originated in Africa at about the same time.

But that is just a minor problem when I am facing fame and fortune. Damn the torpedoes, publish and perish! There is one famous human skull, which I will deliberately not name here, often alleged to have a .22 calibre bullet hole in one temple, and a large exit wound on the other side. The date of this "murder" is a bit uncertain, but definitely well over a hundred thousand years.

Dr Chris Stringer was good enough to show me this skull on the same day that I viewed the Piltdown pieces at the Natural History Museum and he explained that the "bullet hole", while of about the right size, shows signs of healing in the bone. Whatever caused the hole, no weapon, no bullet, arrow or spear could have caused such a wound and left the victim alive long enough to heal the lesion to that extent. And yet, says Stringer, the claims of "spacemen with rifles" still keep coming. So speculation is needed, but it needs to be speculation within some sort of framework.

Without speculation, science will not advance, but without fervent advocacy the speculations that count may never gain due consideration. Equally, without fervent counter-advocacy, we may drown among the psycho-ceramics, charlatans, frauds, and their outputs.

Back to the start of this section

Type 6 science: Polemic

There is a lot of it going on, all the time, and most of it is comparatively harmless. My piggyback gene, like Desmond Morris' "Naked Ape", would carry people through a fair amount of scientific evidence but probably leave them fairly free to revise their views later. Spacemen with rifles, planets in collision, and other such catchphrases may do more harm than good.

This is because full-blooded polemic is generally all about saying "I'd be happy if this were so, therefore it is so". Just before World War I, the English were very keen to have their own fossil human, to match the French and German finds. God was an Englishman, so the first human must also be one of us, they thought! They wanted "Piltdown". They had predicted it. So the English scientists accepted it when it was found. They kept on accepting it for nearly forty years, even though it contradicted the truth.

Polemic often involves saying "Blow the facts, I want it this way, and I'll biff you if you don't agree". There are some fairly strong personalities around the scientific world and polemic is more common than many people realise or admit. Polemic frequently pays little regard to the facts and so it is only marginally scientific, but the people who engage in this sort of argument still call themselves scientists, and at least they argue about scientific facts.

The argument about cholesterol and heart disease is debated with all the fervour of a religious schism. Others may claim to have evidence that vitamin D deficiency can cause colonic cancer, or photocopiers and fluorescent lights cause skin cancer in office workers. They may just turn out to be correct. Most of the time they will not be right, but their cases are still argued furiously.

You see, scientists will not argue a case if they do not believe it whole-heartedly, and they will not bother to argue a case if everybody agrees with it. If you consider that you have been convinced by logical scientific analysis, it would be only human for you to go in, boots and all, to state your case.

At least the polemicists point the way to an observed (or alleged) pattern, something which can be investigated further. More importantly, there are times when the polemicists are right, so they do have a place in science. The rise and eventual triumph of Dart's Australopithecus over the Piltdown bullies, the slow acceptance of the Germ Theory of disease, even of the adoption of the theory of continental drift, now resurrected as plate tectonics, have shown the value of polemic.

As an undergraduate at ANZAAS in 1962, I watched as Professor Sam Carey, a Tasmanian geologist, seized upon almost every presentation to wring out a scrap of evidence for his own, partially correct, view of continental movements. His ideas were mostly wrong, but there were shreds of evidence there that would not go away, and in the end his polemic and that of others, wore everybody down - once there was some truly decent evidence to look at. Carey was at least partly right, His opponents turned out to be plainly wrong.

Back to the start of this section

Type 7 science: Pseudo-science

The proponents of pseudo-science usually take delight in pointing out that their system "transcends" the rules of science, obeys some "higher laws". They do not speculate on this, they assert it. They abhor hypotheses even harder than Nature used to abhor vacuums.

Interestingly, Type 7 people accept most parts of orthodox science, but add generally one "special", one hitherto unknown universal force, or one glaringly obvious, intuitively attractive, but completely unrecognised law. Everything else that they say is consistent with science, given that one supposition, so they fall just within the bounds of science. Watch out for their attitudes to people like Einstein or Newton: they usually attack one or the other of these two worthies, explaining blandly that they can prove mathematically that Einstein or Newton "got it wrong".

The only "plus" for pseudo-science is that like any other attempt at science, a pseudo-science can give us testable propositions. It is because the pseudo-sciences still allow this form of testing that they qualify (barely!) for listing among the "sciences". Earlier, I said technological fraud was not scientific fraud, because there is no possible way somebody can sustain a claim that a technological fraud works and be truly innocent. Some deviousness, some deceit is required as well, even if it is only taken far enough for the fraud to deceive him or herself.

Technological fraud should be impossible to bring off because people want to see the "new inventions" working. Luckily for the con-merchants, people are gullible, willing to believe bizarre things, to accept strange claims. There was even an Australian state Premier who praised a new car engine which used water as a fuel, and encouraged a cancer quack by offering him government support to create a clinic.

But don't you worry about that right now. The point is, there will always be people who are willing to suspend their disbelief and that is why technological fraud sometimes happens. Fortunately, it never lasts. Most children of average intelligence have dreamed of a motor which turns a generator which makes the power which turns the motor, and so on. There are variations like the water wheel which operates a pump to raise the water which powers the water wheel, weird and wonderful arrangements of magnets, rolling weights on wheels, even generators powered by burning hydrogen, which use the electricity to split water by electrolysis, releasing hydrogen to drive the generator.

These are all perpetual motion devices, and even though some of the demonstrations can be most convincing, none of them can truly work. There is absolutely no way of getting energy from nowhere. Still, you can find plenty of people who will claim they are on the verge of a breakthrough. All they need is slightly more powerful magnets, or more ductile widgets, and all will be well. The trouble is the final product never seems to arrive. There has never yet been a "genuine perpetual motion machine" that really was!

But why does pseudoscience sell so well? I think the answer is that, to many people, science is a total mystery. The teacher went too quickly or glossed over an important point and somebody lost the thread. As a teacher, I have to say that "science education" is partly to blame although I do not think it fair just to blame "the teachers" for this.

Whatever the cause, the science education system fails to develop the big ideas of science, fails to ensure that our audiences have a framework in which to understand science, and fails to deal adequately with those parts of science which are so horribly counter-intuitive - like kinetic energy, osmosis, geochronology, evolution, and the inverse square law, to name just a few. It is so much easier to teach, and test, Type 1 things like the meanings of terms or the operations of Ohm's Law.

Meanwhile the people who lost the thread of science are forced to make do with the tatters of understanding which remain with them. Depending on their creative abilities, they are either likely to invent pseudoscience or fall prey to the sharp operators who peddle pseudoscience. To show what I mean, I have invented a new pseudoscience, and I expect to find marketing spin-offs Real Soon Now.

Lying on the beach late one warm afternoon, my science teacher wife and I came to the question of why the beach was so pleasant. Clearly it was due to lying on, and being immersed in, warm sand grains which were spherical. In some way, the grains were absorbing the perverse energies of the universe. Then, because there were no sharp points where the perverse energies could congregate (as electric charge does), the energies were trapped, probably until the tide came up, when the ionic salts in the water would discharge the energies harmlessly into a passing shark (why else are sharks so savage?), or into a passing Angry Penguin.

Our teenage son wandered up at this stage, and suggested to us that there were spherical objects everywhere. In fact, with considerable scientific insight, he suggested that our theory was a load of spherical objects as well (which we took to mean that he thought it pleasing). Still, whatever he meant, the practical lad suggested there was money in our idea and we all began to explore it further. After a quick dip in the surf, (filled with spherical bubbles), our minds went up like balloons. We had a ball, you might say, as we generated the Ball Power Manifesto and Prospectus (application form for membership on the back, all credit cards accepted).

People overloaded at the supra-aural level with perverse energies are often said colloquially to have lost their marbles, so at a subconscious level, everyone has known about Ball Power all along, which imbues our soma like an over-arching but non-hubristic ethos. Pumpkin is generally disliked, and it is a collapsed sphere. Grapes, on the other hand, are much more spherical, and generally enjoyed more, even when they have been crushed, revealing that the inherent sphericity of the grape ethos imbues all portions of the fruit, probably even under-archingly, mutatis mutandis. And who would choose a turnip over an orange?

(Note the deft mix of gobbledegook and simple practicalities in the last paragraph which allows the reader to glide over the bits that hurt the brain, noting how right the simple bits are, and unconsciously extending that rightness to the rest of my arrant claptrap. But I digress: back to our proof.)

Then just look at sporting events, where Rugby League, AFL and ice hockey are nothing more than travelling biff fests. Soccer is less so, and brawls in real hockey, netball, tennis and cricket are rare indeed. Baseball presented something of a problem, but we realised that the ball, when struck by the bat, is seriously deformed from a spherical shape, and there is probably interference from the pseudocrystal shape of the baseball diamond. In any case, the baseball diamonds are green, and you have to avoid green crystals, because they aren't ripe yet, or so our son tells us.

The world has us worried. Most people would prefer a trip around the world to a trip across town, but the world is an oblate spheroid, a flattened sphere. We appeal to all world travellers to Polar regions to carry pebbles (rounded, naturally) with them, and deposit them evenly over the surface, in order to slowly bring our world back to its proper shape. We have a suitable supply of stones available at a remarkably low price for those who truly believe.

Readers are advised that we have copyrighted, patented, registered and staked claims to Ball Power. If you find a way of making money from this boondoggle, we expect a cut!

Back to the start of this section

Topic Ideas

Back to the contents

There are now some 1850 science fair and science project ideas here. I am still working on help files to give people planning projects extra ideas and pointers and hints. In the end though, your project is your project, so make your own decisions. My aim has been not to suggest complete tasks, but to provide sparks that will give people some ideas, either directly, or indirectly, by making them think of something of their own that they could do. Projects are about doing science, right?

Serious help will be available later this year in the form of other commercial ebooks written as companions to this one. Unlike this sample frolicking, most of them will be low-cost commercial products, for which you will be expected to pay. The ideas are roughly grouped by the type of activity, and by subject matter, but you would be well-advised to spend some time browsing around, or searching for key words to see what may suit you. The ideas tend to get harder as you work down the list, but you can make a simple task from the hardest idea, or a complex study from a simple idea.

No idea has gone in the list unless I thought I could probably do it myself, given the right materials and some research. There are a number of entries where all you have is a name (like Foucault's pendulum) where I have left you to find out what it was, and why it was important. In most cases, a Web search for the exact phrase will get you close to an answer.

About half of the ideas are mine, the rest are collected from a variety of sources. The ideas on their own are not copyright and so any of the ideas may be freely used and modified, but this collection as a whole represents a great deal of hard work and is copyright © Peter Macinnis, 2002. That said, this page may be copied for NON-Web use, so long as it retains this statement of claim, designed to make life hard for rogues. It may not be sold, reproduced in any other format or included in any product which is sold, or which has an ISSN or ISBN. No, you may NOT place a copy of it anywhere on the Web, no you may not clip part of it and present it as your own work: just point to this URL, where the latest version will be found.

If that seems blunt, please understand that one parasite in a Chicago school actually tried that -- and got caught.

Now here is some navigation help, but don't forget that there is usually a Find function that will let you find key words in the text:

Poster and display projects, Models and demonstrations, Collections, Experiments, Survey reports, Research projects

Poster/display projects

Animals

Human biology

Plants