The Scientific Method(s)

Note: The previous article by this name was both too long and too narrowly focused on a specific, idealized scientific method. To partially correct this, as of July 12, 2009, I am dividing it into two articles, another one entitled "What Is Science?" and this one entitled "The Scientific Method." The former defines and describes science, and the latter describes the idealized "scientific method" that scientists strive for.

While the phrase, "The Scientific Method," is commonly recognized as something one has heard and maybe even studied in school, it would be more accurate to speak of "scientific methods." Plural. There is no one-size-fits-all method that can be used in all situations. For example, historical sciences like archeology, geology, evolutionary biology, cosmology, and paleontology all require methods different from those of physics and chemistry.

Nevertheless, there are certain principles that every branch of science follows to the maximum extent possible. These principles can be stated to formulate an "ideal scientific method," to which a scientist should conform as closely as his or her circumstances will permit.

Parts of this "scientific method" have been used since even before humans appeared on earth. When your cat chases a mouse, it observes what the mouse is doing and probably forms a hypothesis about the little animal's intentions. Then it predicts what it's going to do next. Or tries to figure out where it went. (I'm not claiming the cat is actually consciously thinking of all this, but the process is going on somewhere in its brain. Otherwise, it would never catch a mouse.) These simple and obvious processes are the rudiments of the scientific method.

Experiments have been done for millennia, though such formalities as double blind testing and refereed publishing are of much more recent origin.

The exact methods used in science will vary from one field to the next and from one set of circumstances to the next. It's not always possible or feasible to follow the exact structure I've outlined below, but these guidelines are generally followed as closely as circumstances permit. Some of them should NEVER be omitted.

1. Observe.

ALL science is (or should be) based on observations. For example, somebody had to notice that liquids flow before they could wonder why and come up with a guess that fluids are composed of tiny particles we now call molecules.

2. Form one or more hypotheses about what you have observed.

A hypothesis is nothing but an educated guess that meets certain criteria. It should explain something about your observations by making a prediction or generalization. Tentatively, of course. Never forget it's just a guess. It must be both testable and falsifiable, because most hypotheses will eventually be proved wrong.

Does this sound too negative? Maybe, but it's true; so we have to live with it. The most important purpose of the "scientific method" is to weed out false ideas and keep us from "knowing" things that just "ain't true." ("Knowledge is a good thing. The only problem is that so many folks know so much that ain’t true." – Author Unknown, but probably Mark Twain)

Testable

A reasonable guess -- or hypothesis -- is not much use unless you can think of a way to test it. This is done by using your hypothesis to predict or generalize something and then verifying it. When possible, this should be done in a carefully planned and controlled experiment to avoid unknowns. If your prediction comes true, it indicates that your hypothesis MAY be correct. This is not "proof," however.

There may often be many different and mutually exclusive hypotheses -- any of which could be partly or completely right. To the extent feasible, they must all be tested by experiments to see which -- if any -- is most likely to be correct.

Falsifiable

Can you think of anything that might be able to convince you -- even potentially -- that your hypothesis is wrong? This is a different meaning of the word than the one in general use. In general usage, to falsify means to tell an untruth. In scientific jargon, it means to think of anything that could even potentially prove the hypothesis false. (Probably every group of people on earth with a common interest develop their own jargon. This is an unfortunate, but necessary, fact of life. Necessary because the jargon actually improves communication within the group, among people who understand it.)

If nothing could possibly convince you that your idea is wrong, even in principle, then you are obviously biased and your hypothesis is useless. If there's no way to prove it wrong, then there's no way to prove it right. You'll never know whether it's true, or whether you just believe it's true because of your bias. It can never be more than just a guess.

Many superstitious, religious, and paranormal ideas fall into this category; because most people with such beliefs are going to hold onto their "faith" with all their being; and NOTHING can convince most of them to consider that they might be mistaken. Many will even admit this. I was once told correctly by a religious reader that "the Bible tells us to be close minded." I agree that it does, but a close minded person has essentially decided to stop learning, and his or her work cannot be considered science. Science is the diametric opposite of a closed mind.

3. Design and perform one or more experiments to test each hypothesis. An experiment disproves a hypothesis far more often than it proves one. To the extent possible, a good experiment should:

Be replicable.

Any scientist anywhere who does the same experiment in the same way should get the same results. If he or she does not, then the hypothesis is either wrong, or at best only partially correct.

Great care must be taken to ascertain the experiments are done the same way, or the results might be different even if the hypothesis is correct within the limits of the original experiment. On the other hand, similar but different experiments can be used to extend the generalizations of the hypothesis being tested.

Provide safeguards against both deliberate fraud and unconscious experimenter bias.

Whenever possible, this should include a double blind setup. In medical experiments, for example, a large number of patients may be divided into two groups. Members of both groups are given a pill every day. But the members of one group get the experimental drug, while members of the other group get an inert substance that only appears to be the drug.

The patients don't know whether they are getting the drug or the phony. This makes it a "blind test." But their doctors don't know either. This makes it "double blind."

After being grouped at random, patients are identified by number and given pills from numbered containers. All the pills look and taste exactly alike. Only after the experiment is over and the doctors have already decided who got better and who got worse (or whatever), then the groups are identified so the experimenters can determine how well the drug worked. This avoids experimenter bias and allows the placebo effect to be accounted for.

Sometimes a double blind is not feasible, but it is very important when the circumstances of the experiment permit. Without it, the experimental results will be far less reliable.

4. Both the individual scientist and others repeat steps 1 through 3 as many times as needed. This may be very many times.

5. Publish the results, preferably in one or more refereed scientific journals.

6. Other scientists, and maybe the same scientist again, repeat steps 1 through 5 as many times as needed. Again, this may be very many times.

7. New hypotheses may arise at any time and from anywhere, and should also be tested.

A new hypothesis may come to mind because of new thinking about old observations or experiments, or because of new evidence derived from the experiments, or from other areas. Essentially, this is a repetition of steps 1 and 2. In this event, steps 3 through 5 must also be repeated again.

8. All conclusions are tentative.

Hypotheses may be adjusted slightly, or even changed completely, at any time, on the basis of new observations or experiments. This is a never-ending process. It is ALWAYS possible that the next observation or experiment may demonstrate something different from what we thought we knew. However, some conclusions are very much more tentative than others.

Hypotheses and theories that have withstood extensive testing by many scientists around the world for decades or longer are never likely to change very much. Even in these cases, though, details may continue to change.

9. While we should never assume we understand anything perfectly or completely, well established principles based on extensive observational and experimental evidence are not likely to change dramatically.

More often, new knowledge will merely improve slightly on what we already know; and it is extremely unlikely (but not impossible) that such principles will ever be overthrown completely. Many examples of this could be cited: Newton's Laws of Motion, Einstein's Theories of Relativity, and Darwin's Theory of Evolution through Natural Selection all fall into this category.

a. Einstein's improvement on Newton's laws of motion.

Newton's laws are so extremely accurate under "ordinary" conditions that we still use them to explore the Solar System. Einstein improved on them by showing they are incorrect under certain extreme conditions. Predictions based on Newton's laws can be very wrong when applied to conditions like travel near the speed of light or in a very strong gravitational field, but they are extremely accurate under ordinary conditions.

b. Darwin's description of natural selection was essentially correct.

This is absolutely amazing considering the fact that he probably had never even heard of genes. We know now that natural selection is just one of many forces that direct evolution. Such principles as punctuated equilibrium, genetic drift, lateral gene flow, and others have refined it considerably, thereby improving on Darwin's theory, but not replacing it.

Notes:

1. Skepticism - A scientist must be skeptical of new ideas, because most of them are simply wrong. Any good scientist will want a lot of evidence before endorsing one.

Because of their necessary skepticism, scientists are sometimes considered close-minded. Well, some probably are. After all, they're only human. On the other hand, most scientists appear willing to accept new ideas when enough evidence finally accumulates through the principles of the scientific method. This is exactly as it should be.

An example of this is our expanding universe. Scientists had believed in an unchanging "steady state universe" until Edwin Hubble made observations in 1929 that indicated the known universe was expanding. Virtually all scientists eventually accepted the idea of a universe expanding so fast that the outer reaches of it are moving away from us at a very high fraction of light-speed. Maybe even faster.

For two-thirds of a century afterward, we knew our universe was expanding (or at least the part if it we could observe); but we also "knew" the expansion must be slowing down because of the force of gravity trying to pull it all back together. This seemed to be a "known fact," totally accepted by all or nearly all astronomers, physicists, and cosmologists.

Beginning in 1998, though, various lines of evidence were shown to indicate the expansion of the known universe is actually accelerating! That is, the known universe is not only expanding; it's expanding faster and faster!

We don't know yet how this is possible, and various hypotheses -- most involving something mysteriously called "dark energy" -- have been proposed to explain it. But my point here is that nearly all scientists reversed their view on this within a period of just six years, because of overwhelming new evidence.

Similar examples could be given, like the Big Bang, plate tectonics, rocks falling from the sky, Darwinian evolution, Special and General Relativity, Quantum Mechanics, and many others. Scientists at the time were appropriately skeptical, but there is now almost 100% agreement with these ideas because of the abundance of evidence that has been produced by the scientific method.

This agreement in principle, of course, does not mean anybody thinks we know everything there is to learn about any of these things; and new observations, hypotheses, and experiments are being performed continually to make our knowledge more accurate and more nearly complete.

We work with what we have until new, convincing evidence is presented; or sometimes until a new fact or thought leads to reinterpretation of the old evidence. Then we go on with what we hope is a better and more accurate -- but still tentative -- picture of reality.

We should all get used to thinking this way. We should observe things around us, think about what happens, how and why it happens, and what effects it is likely to have on us and others.

No, we are not all scientists; so we may not do a lot of experimenting or publishing. But we can read about many of the experiments performed and published by others and accept the tentative conclusions that seem to be best supported by that evidence, while skeptically rejecting any ideas that are not supported.

2. Tentative Conclusions - Everything in science is always considered tentative and subject to further observations, experiments, and even new interpretation. However, it must be said again and again that some conclusions are very much firmer than others. For example, as mentioned above, we know by millennia of observations that the sun always appears to "come up" every morning. Our natural hypothesis is that the sun will "come up" again tomorrow morning. This appears to be so certain that we seldom give it any skeptical thought. It's just the way the world is.

The same is true of many other things. If I drop a ball, I assume it's going to fall. If I trip and can't catch myself, I assume I am going to fall, as I've done many times in my seven decades. These are firm conclusions based on previous observations, and they seem to be pretty certain. Conclusions like these are not likely to change significantly, which is precisely why I walk with a cane.

Nevertheless, it was only by this method that we eventually learned the sun doesn't really come up after all. It just appears to because the earth turns. Who would have ever thought it?

As science teacher Alom Shaha puts it, "Separating truth from fraudulent mumbo jumbo is just one reason why science is important."

Psychologist and Freelance Writer Dr. Valerie Tarico adds "Confirmation bias is so built into human thinking that the whole scientific endeavor is structured essentially to get around it. The scientific method has been called, 'What we know about how not to fool ourselves.'"

Physicist Bob Park sums it up, "Science is the only way of knowing."

This page was last updated 08/17/09 02:20 AM.

Home | Bible Contradictions | Dictionary | Einstein | Evolution | FAQ | From the Net | From Readers | Giggles | Home Page Archive | How Old Is Earth? | Is Science Just Another Religion? | Links | Questions for God | Religion: What Others Say  | Saved | Skepticism or Cynicism? | Terms | Scientific Method | Things I'm Skeptical Of | To Save a Mockingbird | What Can I Believe? | Update Archive | What Does a Believer Lose? | Who is Bill Dearmore?