The Scientific Method There are many ways to obtain knowledge. Modern sci- entists tend to obtain knowledge about the world by making systematic observations. This principle is called empiricism and is the basis of the scientific method. The scientific method is a set of rules for asking and answer- ing questions about science. Most scientists use the scientific method loosely and often unconsciously. However, the key concepts of the scientific method are the groundwork for scientific study, and we will review those concepts in this section. The scientific method involves: ■ asking a specific question about a process or phe- nomenon that can be answered by performing experiments ■ formulating a testable hypothesis based on obser- vations and previous results ■ designing an experiment, with a control, to test the hypothesis ■ collecting and analyzing the results of the experiment ■ developing a model or theory that explains the phenomenon and is consistent with experimental results ■ making predictions based on the model or theory in order to test it and designing experiments that could disprove the proposed theory THE QUESTION In order to understand something, a scientist must first focus on a specific question or aspect of a problem. In order to do that, the scientist has to clearly formulate the question. The answer to such a question has to exist and the possibility of obtaining it through experiment must exist. For example, the question “Does the presence of the moon shorten the life span of ducks on Earth?” is not valid because it can not be answered through experi- ment. There is no way to measure the life span of ducks on Earth in the absence of the moon, since we have no way of removing the moon from its orbit. Similarly, asking a general question, such as “How do animals obtain food?” is not very useful for gaining knowledge. This question is too general and broad for one person to answer. Better questions are more specific—for example, “Does each member of a wolf pack have a set responsi- bility or job when hunting for food?” A question that is too general and not very useful is “Why do some people have better memories than others?” A better, more spe- cific question, along the same lines, is “What parts of the brain and which brain chemicals are involved in recol- lection of childhood memories?” A good science question is very specific and can be answered by performing experiments. THE HYPOTHESIS After formulating a question, a scientist gathers the information on the topic that is already available or pub- lished, and then comes up with an educated guess or a tentative explanation about the answer to the question. Such an educated guess about a natural process or phe- nomenon is called a hypothesis. A hypothesis doesn’t have to be correct, but it should be testable. In other words, a testable hypothesis can be disproved through experiment, in a reasonable amount of time, with the resources available. For example, the statement, “Everyone has a soul mate somewhere in the world,”is not a valid hypothesis. First, the term soul mate is not well defined, so formulating an experiment to determine whether two people are soul mates would be difficult. More importantly, even if we were to agree on what soul mate means and how to experimentally deter- mine whether two people are soul mates, this hypothe- sis could never be proved wrong. Any experiment conceived would require testing every possible pair of human beings around the world, which, considering the population and the population growth per second, is just not feasible. A hypothesis doesn’t need to be correct. It only has to be testable. Disproving a hypothesis is not a failure. It casts away illusions about what was previously thought to be true, and can cause a great advance, a thought in another direction that can bring about new ideas. Most likely, in the process of showing that one hypothesis is wrong, a – SCIENCE AS INQUIRY– 220 scientist may gain an understanding of a better hypoth- esis. Disproving a hypothesis serves a purpose. Science and our understanding of nature often advance through tiny incremental pieces of information. Eliminating a potential hypothesis narrows down the choices, and eliminating the wrong answers sometimes leads to find- ing the correct one. THE EXPERIMENT In an experiment, researchers manipulate one or more variables and examine their effect on another variable or variables. An experiment is carefully designed to test the hypothesis. The number of variables in an experiment should be manageable and carefully controlled. All vari- ables and procedures are carefully defined and described, as is the method of observation and measurement. Results of a valid experiment are reproducible, meaning that another researcher who follows the same procedure should be able to obtain the same result. A good experiment also includes one or more con- trols. Experimental controls are designed to get an understanding of the observed variables in the absence of the manipulated variables. For example, in pharmaceu- tical studies, three groups of patients are examined. One is given the drug, one is given a placebo (a pill contain- ing no active ingredient), and one is not given anything. This is a good way to test whether the improvement in patient condition (observed variable) is due to the active ingredient in the pill (manipulated variable). If the patients in the group that was given the placebo recover sooner or at the same time as those who were given the drug, the effect of pill taking can be attributed patient belief that a pill makes one feel better, or to other ingre- dients in the pill. If the group that was not given any pill recovers faster or just as fast as the group that was given the drug, the improvement in patient condition could be a result of the natural healing processes. An experimental control is a version of the experiment in which all conditions and variables are the same as in other versions of the exper- iment, but the variable being tested is elimi- nated or changed. A good experiment should include carefully designed controls. T HE ANALYSIS Analysis of experimental results involves looking for trends in the data and correlation among variables. It also involves making generalizations about the results, quantifying experimental error, and correlating the variable being manipulated to the variable being tested. A scientist who analyzes results unifies them, interprets them, and gives them meaning. The goal is to find a pat- tern or sense of order in the observations and to under- stand the reason for this order. MODELS AND THEORIES After collecting a sufficient amount of consistently reproducible results under a range of conditions or in dif- ferent kinds of samples, scientist often seek to formulate a theory or a model. A model is a hypothesis that is suffi- ciently general and is continually effective in predicting facts yet to be observed. A theory is an explanation of the general principles of certain observations with extensive experimental evidence or facts to support it. Scientific models and theories, like hypotheses, should be testable using available resources. Scientists make pre- dictions based on their models and theories. A good the- ory or model should be able to accurately predict an event or behavior. Many scientists go a step beyond and try to test their theories by designing experiments that could prove them wrong. The theories that fail to make accurate predictions are revised or discarded, and those that survive the test of a series of experiments aimed to prove them wrong become more convincing. Theories and models therefore lead to new experiments; if they don’t adequately predict behavior, they are revised through development of new hypotheses and experi- ments. The cycle of experiment-theory-experiment con- tinues until a satisfactory understanding that is consistent with observations and predictions is obtained. – SCIENCE AS INQUIRY– 221 . is consistent with observations and predictions is obtained. – SCIENCE AS INQUIRY– 221