Commentary Science and the Scientific Method

Science is the human enterprise that is based on knowledge, skill, and practice in combination with a discipline that allows either prediction or predictable outcomes. Research is the method that is used to gain knowledge in an organized way, using highly organized methods. There is research science and applied science. Research science focuses on the use of the experiment to examine and observe. Applied science is the use of research to meet specific goals or to fulfill specific needs.

Science is the formalized process where human knowledge is improved, increased and expanded through controlled methods and with specific disciplines that are applied.  Scientific understanding in the general population is improved through teaching, reporting, and illustrating through public expositions of scientific advancements.

There are natural and social phenomena that are observed, and evidence to support the observations is collected. Where possible, measurable data is collected. This requires quantification, written observations, photographic and other records that can be used to analyze the information. The information is then used to arrive at hypotheses, predictions, or theory to explain how things work or how they are likely to work.

Experiment is the process where hypotheses or theories are tested. The gold standard of experimentation is achieved when absolute control over every factor that could possibly affect the outcome of a test or process is guaranteed. The goal is for a completely different scientists to take all of the information related to the matter and to repeat the experiment, producing the same results.  It is fair to say that this is impossible in most cases of social, medical, and other scientific experimentation. As a result, there are many forms of experiment, from pure experiment, to quasi experiment, to non experiment.

There is experimentation cause and effect, where the gold standard is to eliminate any factor that is not pure cause or pure effect. The most likely form of experiment is in the quasi experiment where a null or alternative hypothesis is created and tested along with the main hypothesis in order to determine if there is something other than the expected relationship going on. In other words, there is not another explanation for the cause and effect.

There is experimentation with programs that are so complex, containing many variables that affect many other variables, with unknown and known intervening variables, that researchers do the best that they can to control what they can. Drug testing on human subjects is an example where the subjects might be engaging in any type of unknown activity, have unknown physical factors, or might be subjected to external influences that could affect the outcome. Social programs with multiple sub components and diverse populations represented in the control and test groups require quasi experimentation.

A non experimental process is used when there is no way to quantify data, there is missing data, or the phenomena is so new that there is nothing to compare it with in order to classify it’s properties.