BIO 111/111L

The Scientific Method

Science is a way of knowing. Originating from the Latin word scio meaning to know, science stands together with other disciplines such as art, music, and religion as a way man interprets and understands his universe. Science differs from other methods, however, in that the knowledge gained is tested for its validity.

The scientific method can be used for all knowledge of the natural world and depends on the following assumptions:

1) Causality: All events in nature are due to natural causes, e.g. the laws of

nature. Confidence in this assumption rests on a steady procession of causes we have discovered for events formerly believed to be supernatural.

2) Uniformity in Space and Time: Natural laws do not change with time or

distance. Geological, astronomical, physical and biological laws are the same today as they were millions or even billions of years ago. This means that data from geologic strata, light emitted from stars, and fossils of earlier evolutionary forms permit us to interpret the past, present, and future of nature.

3) Common Perception: All human beings perceive nature and events in fundamentally the same way. We can interpret data uniformly. Value systems like those in art, music, and religion do not assume common perception. Since these value systems are subjective,

not objective, science cannot be used to draw conclusions about them.

We all obtain most of our knowledge from various forms of observation: we see, read, hear, and from these observations which we may accept or reject, we often draw conclusions. Trial and error helps us to improve our interpretations, but we still come up with knowledge which may or may not be correct. Sometimes a body of such 'iffy' knowledge is labeled as so-called common sense. But the scientific method is used to determine the validity of interpretations and to arrive at correct conclusions about what we observe.

Science begins with observations which become variables. These are like cause and effect. The observed result or effect is the dependent variable. The possible causes are the independent variables. For example, suppose we observe a large number of dead waterfowl near a lake. The cause of this mortality is unknown, but we can see or imagine a number of possibilities. There could have been a disease, or severe weather conditions, or pesticide dumped in the lake, or waste from an abandoned mine site upstream. These are all independent variables, while the death of the ducks is the dependent variable. We select one independent variable for which to write an hypothesis. The independent variable is the one manipulated in an experiment. Only one independent variable can be tested in a single experiment.

The next step is to formulate the hypothesis, a logical explanation for what we've observed which relates the independent variable selected to the dependent variable. Some people call the hypothesis a guess, but if so it's a very educated guess. For instance, let the hypothesis be that "mine waste killed the ducks". From this hypothesis we write "If...Then" statements which logically follow (are predicted by) the hypothesis and which can be tested by experiment. "If mine waste killed the ducks, then... we should find a toxin in the waste which is known to cause death, we should find the toxin or its residue in the water, we should find it in the tissues of the ducks, etc. All of these can be tested and if found to be true, lead to increasing validity of the hypothesis. Tests can also show the absence of evidence for other variables as causes. But they do not prove that the toxin actually killed the ducks.

To obtain the greatest validity for your hypothesis you need to control all the other variables. A controlled experiment is one in which all variables are controlled (the same in the experimental and control groups) except the one being tested. The best controlled experiments are performed in the laboratory. So a control group would be compared with an experimental group in which all variables are the same, (e.g. temperature, water, food) and all other possible causes are absent, but the experimental group gets the toxin and the control group doesn't. This would lead to data which should, if they are unequivocal, determine the validity of your hypothesis. Statistics are used to establish if the correlation seen is greater than would be expected from chance. For instance, some of the experimental ducks might be somewhat resistant to the toxin and not die. Some of the control group ducks might die anyway, just from being handled. So the data should show that the death was significantly greater in the ducks fed the toxin that you would expect from chance. Statistical analysis will lead to a degree of probability for your hypothesis, but it will not prove it.

Results of experiments may validate, refute, or lead to revision of your hypothesis. Sometimes, for statistical purposes, the null hypothesis is used. The null hypothesis states just the opposite of what you believe to be correct. The null hypothesis for this example would be that mortality is unrelated to the toxin found in the waste from the mine. Since it is much easier to invalidate or disprove something than it is to prove it, any statistical correlation at all will disprove the null hypothesis and therefore validate its opposite. This method has often been used when a relationship is described as "correlated with" or associated with" rather than "caused by".

Objectivity is important in all phases of an experiment. For that reason many experiments use double, or triple blind methods, especially those involving medical testing. Neither the subject,

administrator, nor the evaluator should know whether a drug is a placebo or an experimental

substance. Otherwise expectations may play a large role in the response. The data also should be as objective as possible, preferably numerical data which can not be subjectively interpreted.

Repeatability is very important in science. other scientists should be able, using the same conditions, to obtain the same results. if not, then the science is not valid. This is why scientific publishing is so important in communicating with other researchers. Not only do these repeat experiments serve to validate the results, but they serve as a starting point for new and different experiments, using new ideas, new techniques, new tools etc. So the hypothesis is refined and honed as new data and new results are obtained. Related hypotheses developed by science are therefore tested and retested and eventually in this manner may be formulated into a theory. A theory consists of hypotheses which have been supported by a large number of observations and experiments and are considered valid by an overwhelming majority of scientists. The terms principle or law are reserved for those concepts that have existed for a long time and are accepted as proven fact. Even so, principles or laws are always subject to new testing and revision, as is the case now with many longstanding physical principles dealing with gravity, electromagnetism, and particle physics.

Biological Systems are much more complex than those studied in physics and chemistry. There are many more variables to consider. And many of these variables are difficult or impossible to control, especially in the field. If you ignore them and just focus on certain ones, your data may be clearer and easier to interpret. But it may lead to interpretations which are less valid and applicable to real life. Even when an experiment can be done in a laboratory and the variables controlled, the complexity that makes the system work in nature may be missing. Biologists speak of emergent properties which represent a synergy that exists when elements of a system come together, which is absent when they are separate. The new discipline of chaos or, more properly, complexity theory attempts to address the sometimes seemingly unpredictable results of these emergent properties.

Selected Scientific Disciplines and Their Areas of Study

Biological Sciences:

biology - the study of all living organisms

cytology - the study of cells

histology - the study of tissues

microbiology - the study of microorganisms

botony - the study of plant life

zoology - the study of animal life

ecology - the study of the relationships of organisms to their environment

entomology - the study of insects

herpetology - the study of reptiles

paleontology - the study of fossils

Physical Sciences:

physics - the study of physical laws

chemistry -the study of atoms and molecules

geology - the study of the earth

astronomy - the study of the universe

Behavioral Sciences:

sociology - the study of society

psychology - the study of the mind

Medical Sciences:

pathology - the study of disease

anatomy - the study of structures of organisms

physiology - f the study of unctions in organisms

endocrinology -the study of hormones

oncology - the study of cancer

Crossover Disciplines

anthropology - the study of man

archeology - the study of man's works