An organized approach utilized to systematically study and solve problems of the natural world.
1. Make a set of observations
Example: Annual precipitation values across and along a mountain range.
2. Define a problem evident from an examination of the observations
Example: More precipitation occurs on the windward slopes of the mountain range. Why?
3. Formulate a hypothesis to explain the problem using proven principles
Example of Proven Principle: Rising air cools.
Example of Hypothesis: Air is cooled to the dewpoint in being forced over the mountains. Hence, precipitation might be augmented on western side of range if the average wind direction is westerly.
| It is the responsibility of the researcher to ensure that both the problem and the subsequent hypothesis to explain the problem are grounded in the scientific research which has already been completed on the topic. Such a procedure is termed a "literature review." Scientists are expected to be familiar with all aspects of the research which has gone before on their given topic and that the research contributes to the body of literature on it, rather than reproducing what has already been done or abandoned. |
4. Test the hypothesis by experimentation
Example: At least 10 years of rainfall and wind data are examined for the mountain range to determine if there is a systematic STATISTICAL correspondence (called "correlation") between westerly winds and precipitaiton amounts.
5. a. Derive a theory based upon successful experimentation
Example: Moist air lifted by mountains is associated with enhanced condensation of clouds and formation of precipitation.
5. b. Modify hypothesis or redefine problem if experimentation does NOT bear out hypothesis.
Example: Other factors are responsible for precipitation distribution AND/OR set of observations was incorrect AND/OR wind directions were unusual during the 10 year experiment AND/OR there is no relation between precipitation amounts and wind directions.
Once 5a is achieved, return to (4) and reproduce results one or more times. If the hypothesis is correct, then the experiment should produce the same results no matter which years are chosen nor what geographic region is selected.
In order to ensure that the theory obtained in 5a is correct and that all steps in the procedure have been unpolluted by errors or biases, independent experts (called "referees") must examine the whole procedure in an anonymous way. This procedure is termed "peer review".
Those individuals who short-circuit the process by stepping to the media before peer review risk having undiscovered errors or faulty reasoning undermine their results. This has a twofold destructive effect: (a) the body of knowledge on the given topic is not advanced and the researcher does not contribute to the science; and, (b) the researcher does a disservice to the public since neither the average person nor the media is in position to judge the accuracy and integrity of the research procedures.
"Charlatan" scientists often choose to forgo or step around peer review, and often go so far as to publish their own results ("self publication" or "using the vanity press") while portraying such publication as peer reviewed. Again, the public is not in position to judge whether such publication is truly peer reviewed.
Such fringe scientists often resort to arguments that the peer-reviewed literature is designed to suppress their ideas or is too involved and time consuming. Recent examples of this include the "cold fusion" controversy of several years ago, and almost all of the papers "published" in the area of creationism.