The Scientific Method

The beauty of the scientific method is that it allows the consensus to be overthrown when necessary.

This table of comparisons between Science and Pseudoscience was adapted from text by Kendrick Frazier
in the introduction to his book, Paranormal Borderlands of Science (Prometheus Books, 1981)

SCIENCE	PSEUDOSCIENCE
Always undergoing revision. Has a no-holds-barred, let-the-facts-fall-where-they-may attitude.	Clings emotionally to pet ideas long after they've been shown to be wrong.
Limits claims to matters that can be supported reasonably well with good evidence.	Makes sensational and exaggerated claims that go far beyond the evidence.
*Actively seeks out comments and criticism from well-informed colleagues before* publishing. Shares research data.**	Avoids informed criticism before publication. Often keeps the details of their work obscured and secretive.
Claims are first published in professional journals that use peer review to ensure the work meets minimal standards of competence and accuracy.	Goes straight to the public, claims are presented in commercial books, magazines, and other venues whose publishers make no independent effort to verify accuracy.
Frames claims in such a way that if wrong, they can be proven wrong.	Frames claims in such a way that they cannot be proven wrong. Constantly shifts the grounds for substantiation, makes vague and ambiguous statements that cannot be tested.
Understands that the burden of proof is on the investigator making the claim. A hypothesis is not considered valid until it has stood up to many tests.	*Places the burden of disproof on the critics. Holds that the claim is true unless others "disprove" it.*
The more a claim contradicts previously demonstrated evidence, the greater the new evidence must be before it will be accepted.	Thrives on sensationalism, the more outrageous a claim, the more publicity it will receive, and thus more public following.
Realizes all information is imperfect. Attempts to assess the amount of error attached to all measurements and the degree of reliability associated with all claims.	*Often presents claims as infallibly true. Does not distinguish between the varying quality of evidence (if any) used to support claims. Claims are often personality-driven, not evidence-oriented.*
When shown to be wrong, science acknowledges the fact and modifies the work accordingly.	Sees any criticism as the sign of closed minds and ignorance of scientists. Quick to don the role of martyr and appeal to public sympathies.
Scientists build on other scientific work. They familiarize themselves with previous relevant work before attempting to extend or modify it.	Pseudoscientists often ignore previous studies altogether, especially work that conflicts with their pet theories.
Science is an error-correcting activity.	Pseudoscience is an error-promulgating activity.

Mr. Frazier acknowledges that scientists are also humans and therefore not perfect. At times they can behave just as emotionally and irrationally as their pseudo-science counterparts.
So, while science advances toward being an objective body of error-free knowledge, scientists themselves are not always as objective and error-free as they should be.

The usual tendency of amateur paranormal researchers is to ignore steps 2 and 3 while jumping directly from Observation to Conclusion. That does save a lot of time, thought, and effort, but it isn't science and it does nothing to advance the credibility of the work as far as the legitimate scientific community is concerned. And, while amateur investigators like to invoke the appearance of doing scientific work with their use of technology, they fail to apply the essential elements of methodology. So, despite all their EMF meters, infrared thermometers, audio recorders, specialized cameras, computers, and the like, as long as they ignore the Hypothesis and Experiment steps of the scientific method, investigators might as well be playing with plastic toys and cardboard boxes

One school of pseudo-scientific thinkers tries to argue that paranormal matters are, by their very nature, excluded from being subject to the scientific method. Such arguments are misguided and usually contrived by people who simply don't understand the essential steps for conducting a valid research project. Of course, some who claim thatcertain topics are immune from scientific analysis are the ones most personally invested in that topic -- and it's one that doesn't hold up to scientific scrutiny so they will try anything to keep their own ship afloat.

If paranormal researchers can ever hope to be taken seriously in the scientific arena they're going to have to start playing by the rules.
No shortcuts, no loopholes, no way around. So buckle up and let's get down to basics. Once again:

OK, you've probably got the Observation part covered, at least the preliminary, or "discovery" stages of observation. Whatever your particular paranormal subject of interest, chances are you've read books, watched some TV shows, surfed some websites, maybe you've even had some personal paranormal experiences of your own. Now it's time to think about all the things you've observed, all the information you've absorbed, and try and form a hypothesis regarding some aspect of your observations.

"A hypothesis is a question that has been reworded into a form that can be tested by an experiment."

Right off the bat we're in sticky territory because not all questions can be subjected to direct experiment -- but that's not the same as saying that nothing about the paranormal can be subjected to science. The trick is to focus on a narrow and well-defined aspect of your chosen topic, and to frame your question relating to that one aspect. In other words, don't set out thinking you're going to resolve the question "Do Ghosts Exist?" Instead, try to isolate one very specific point in your observations concerning ghosts, and then ask a question about that point in a way that can be determined through objective tests. (Note: It's OK if you think you already know the answer, the test will serve to either confirm or refute your preconceived notion, your hypothesis.)

For example, let's say you've read a number of ghost sighting reports and in the process you notice that more sightings appear to have occurred at night; thus, you might pose the question, "Do Most Reported Ghost Sightings Occur at Night?" Your next step might be to collect as many ghost reports from as many different sources as possible and to systematically record data concerning the time of day each sighting occurred. The data could be arranged in a table or a graph, statistical analyses performed, and conclusions drawn. Your conclusion would then have a respectable degree of scientific validity.

In this example, we've hedged a little with step number 3 from our statement of the Scientific Method since working with an existing body of data is "analytical" rather than "experimental" in the strictest sense. This analytical approach is more characteristic of the "social sciences" than the "physical sciences," but it also yields scientifically valid results. Of course, you will not have proven anything about the actual existence of ghosts, or anything about the validity of the sightings, but you will have at least helped to advance our knowledge about the nature of ghost reports themselves so that we don't have to rely merely on conjecture and popular opinion.

There might be any number of similar and relevant questions that you could address by working from the same data set. Eventually, you might build up quite a ponderous body of data concerning ghost reports: the most common times, locations, gender and age of people who file such reports, etc. Let's go out on a limb and say that at some point you notice a pattern within your collection of ghost data that leads you to think that a new ghost sighting will occur at a certain time and place. At that point, you might formulate a hypothesis, one that is "testable" by your being at a certain place and time armed with cameras and other appropriate gear to record the event and bingo, you've just conducted a predictive experiment concerning ghost activity.

OK, the overly-simplistic example above is something I came up with off the top of my head and I sincerely hope you can do better., but the point I'm trying to make here is that paranormal topics can be rendered subject to the experimental process which is a key component of the scientific method. There's usually a lot of tedious preparation involved in conducting a meaningful research project - a lot of thought, attention to every detail, and meticulous record keeping. And then there's the matter of sharing your resultys with others so that they can learn from and attempt to replicate your results, thereby supporting (or repudiating) your conclusions. It's an ongoing process of refinementand validation.

Here's a good article on The Scientific Method from Wikipedia: Scientific Method
Of course you can find many more references on the net, in libraries, and bookstores.

Following are selected excerpts from "EXPERIMENTAL SCIENCE PROJECTS: An Intermediate Level Guide"
as found at MiniScience.com See the full article here: Experimental Science Projects

What is common among all sciences, is the making of hypotheses to explain observations, the gathering of data, and based on this data, the drawing of conclusions that confirm or deny the original hypothesis.

A hypothesis is a question that has been reworded into a form that can be tested by an experiment.

Not all questions can be dealt with by the experimental scientific method. You must choose a question or problem that can be formulated in terms of a hypothesis that can be tested. Tests done to check hypotheses are called experiments. To design a suitable experiment you must make an educated guess about the things that affect the system you want to investigate. These are called variables.

As you do experiments, you will record data that measures the effect of variables. Using this data you can calculate results. Results are presented in the form of tables or graphs. These results will show you trends related to how the variables affect the system you are working with. Based on these trends, you can draw conclusions about the hypothesis you originally made.

For an experiment to give answers you can trust, it must have a "control." A control is an additional experimental trial or run. It is a separate experiment, done exactly like the others. The only difference is that no experimental variables are changed. A control is a neutral "reference point" for comparison that allows you to see what changing a variable does by comparing it to not changing anything.

Experiments are often done many times to guarantee that what you observe is reproducible, or to obtain an average result. Reproducibility is a crucial requirement. Without it you cannot trust your results. Reproducible experiments reduce the chance that you have made an experimental error, or observed a random effect during one particular experimental run.

Try to Answer Related Questions

What you have learned may allow you to answer other questions. Many questions are related. Several new questions may have occurred to you while doing experiments. You may now be able to understand or verify things that you discovered when gathering information for the project. Questions lead to more questions, which lead to additional hypothesis that need to be tested.

The material on this website (except where otherwise indicated) has been prepared by J. Hale.
All original content is copyright protected and all publication rights are reserved. Effective September 1, 2006, and beyond.