Scientific Method

Many methods are employed to generate and synthesize knowledge. As such, no single series of steps could contain all the strategies that can be used to discover and explain the universe. Thus, science is better viewed as a process than as a recipe or formula that can be used to create knowledge. For example, considerable knowledge is generated and passed along via careful observations and subsequent descriptions. Basic observations and descriptions contribute to the discovery and documentation of patterns, and each of these steps is fundamental to increasing our understanding. However, the process known as the scientific method involves steps beyond observation and description. Specifically, these steps include development and testing of explanations for patterns derived from observations.

Discovery and documentation of patterns rooted in careful observations and descriptions certainly represents science. However, the hypothetico-deductive method is not employed when science terminates with observation and description. As such, discovery and documentation, rooted in observations and descriptions, do not comprise the process commonly termed the scientific method. The “scientific method” depends on the generation and testing of scientific hypotheses, which are defined as candidate explanations.

Scientific hypotheses typically are derived from patterns because demonstration of a pattern often generates the question: “What process causes that pattern?” The response to the question is a scientific hypothesis, or a candidate explanation. Providing a definitive answer to the question requires formulation and subsequent testing of potential explanations for the observed patterns; that is, it involves hypothesis testing.

A hypothesis generated by a pattern should not be confused with a prediction; a prediction is a statement that is likely to be factual. Most predictions can be evaluated via observation. Although the instruments used for evaluating predictions are variable and may be quite sophisticated, they should not be confused with hypotheses. The scientific method, as typically described, requires generation and testing of hypotheses. In contrast, generation and evaluation of predictions, which are fundamental to the generation of knowledge, are not part of the classically defined “scientific method.” Indeed, if generation and evaluation of predictions represent application of the scientific method, then the scientific method is not unique to science. Rather, it is used for everyday activities such as mowing the lawn (the grass is too tall in some spots, so I must have missed those spots), shopping for groceries (I need carrots, so I will look in the produce section rather than searching the entire store), and commuting to work (length of route and traffic patterns dictate my path). In other words, science has little to offer beyond everyday activities if observation and description are the only means by which we acquire knowledge.

Consider a relatively simple example of a prediction: Retention of information about human behavior is greater with inquiry-based methods than with lectures. This prediction can be (and has been) evaluated with observations of students in classrooms. It also can be (and has been) evaluated with well-designed and rigorously applied experiments (involving, for example, random assignment of students to groups that are taught material in different ways). Statistical hypotheses likely will be generated and evaluated, but we should take care to distinguish between these statistical hypotheses and scientific hypotheses. Whether researchers employ observations or experiments, however, evaluation of the prediction does not involve candidate explanations. Generation of such explanations—that is, scientific hypotheses—is a logical next step in the conduct of research because it addresses the question of “why.”

...