The scientific method is the ideal process by which scientists say scientific research should be conducted.  It is a standardized process which involves asking questions, searching for answers, guessing a possible answers called a hypothesis, and then evaluating that hypothesis by experiments in a specific, rigid fashion. Most articles in scientific journals are formatted according to its steps.
Science is an active process where systematic descriptions of reality continue to accumulate as scientists ask questions and use improved experiments and analysis to answer those questions. The goal of the scientific method is to test the validity of a hypothesis - a proposed description of reality. It is not a set of directions for making original discoveries and it does not set out the means that scientists must use in order for their research to succeed. The whole point is to compare the hypothesis with the facts.
Hans Storch wrote:
- Data must be accessible to adversaries; joint efforts are needed to agree on test procedures to validate, once again, already broadly accepted insights.
Steps of the Scientific Method
Although by no means conclusive, the following steps are used by a majority of scientists in their work::
- Observation of phenomena
- Formulation of an hypothesis
- Predictions, using this hypothesis
A science educator lists 11 steps and 3 supporting principles. 
The scientist observes something interesting, and he wants to know how it happened. He lays down the basic questions as to what is responsible for the phenomena he observed, and from there begins to form his hypothesis.
In asking these questions scientists also look for research that has already been done on their topic to determine if they are duplicating a past experiment, doing something new, or building on a previous experiment. Such research, although tedious and time-consuming, simply builds on the knowledge yet to be gained by the scientist’s questions.
A hypothesis is a statement of what the researcher thinks will happen in the experiment. This is usually an educated guess using current theory and has to be testable and observable. It is a common mistake to let one's own scientific bias stop the process at this point, on the assumption that the idea is so logical that it needs no testing.
When designing the experiment, the researcher carefully controls as many variables as possible. In most experiments there is a control group and a treatment group. The two groups are as similar as possible, but the treatment group is the one that experiences the variable the researcher is studying.
After the data are analyzed and written down, the scientist checks the results against the hypothesis; if the results have proven the hypothesis to be wrong, then it must be discarded. Even if the hypothesis is not correct, conclusions can still be made and significant knowledge gained. If the hypothesis is indicated to be correct, then the results are published and sent to other scientists within the field in question.
Scientists must be able to take such published data and repeat the experiment. This not only confirms the validity of the original hypothesis, but advances it to the level of a “theory”, which in science means an interpretation or explanation that is well-supported by evidence which is testable and tested. A theory can also be falsified by evidence as well.
- A scientific theory summarizes a hypothesis or group of hypotheses that have been supported with repeated testing. A theory is valid as long as there is no evidence to dispute it. Therefore, theories can be disproven. Basically, if evidence accumulates to support a hypothesis, then the hypothesis can become accepted as a good explanation of a phenomenon. One definition of a theory is to say it's an accepted hypothesis. 
A classic example of the Scientific Method being used stemmed from a simple bet. In 1872 a railroad baron named Leland Stanford made a wager that a horse’s hooves do not touch the ground at some point in a gallop. To test the hypothesis, photographer Eadweard Muybridge  was hired; he installed a series of trip wires which were rigged from a long wall about two inches from the ground, each one tied to a camera’s shutter facing the wall; the experiment called for the horse to run past the wall, tripping the wires and getting a photo at each point. The results were factual and conclusive: a horse at a running gallop does have all four hooves off the ground at some point.
The agreement of an observation or experiment with a hypothesis does not on its own prove the hypothesis correct. It merely makes its correctness more likely. The hypothesis must agree with other aspects of the scientific framework of knowledge, and survive the test of repeated experiments by other people working independently. Over time, the accumulation of data will tend to confirm or refute a hypothesis.
Scientists may be influenced by their world-views to look for certain results that fit a preconception. The test of objectivity and rigor imposed on their work by the need for other scientists to replicate it tends to make the truth-seeking facility of the scientific method prevail in the long run, although this is difficult where the world-view is widespread.
- Department of Physics and Astronomy, University of California, Riverside
- The Scientific Method, from Clermont College
- Introduction to the Scientific Method
- Science Buddies.org
- Biology for Kids
- Example of the Scientific Method used, from NASA
- Campbell, Reece, Taylor, Simon, et al. Biology: Concepts and Connections 5th edition; Pearson Education, Upper Saddle River, NJ (2005)