“Science is not the mysterious, distant, smoking-test-tube sort of a priesthood that many imagine it to be. Rather, it is simply an organized, formal method of finding out.” James Randi
The scientific approach to knowledge is based on systematic empiricism (Stanovich, 2007). Observation itself is necessary in acquiring scientific knowledge, but unstructured observation of the natural world does not lead to an increased understanding of the world. “Write down every observation you make from the time you get up in the morning to the time you go to bed on a given day. When you finish, you will have a great number of facts, but you will not have a greater understanding of the world” (Stanovich & Stanovich, 2003, p. 12).
Systematic Empiricism is systematic because it is structured in a way that allows us to learn more precisely about the world. After careful systematic observations, such as those in controlled experiments, some causal relationships are supported while others are rejected. Extending these observations, scientists propose general explanations that will explain the observations. “We could observe end-less pieces of data, adding to the content of science, but our observations would be of limited use without general principles to structure them” (Myers & Hansen, 2002, p. 10).
The empirical approach (as used in everyday observation) allows us to learn things about the world. However, everyday observations are often made carelessly and unsystematically. Thus, using everyday observations in an attempt to describe, predict and explain phenomena is problematic.
When observing phenomena a scientist likes to exert a specific level of control. When utilizing control, scientists investigate the effects of various factors one by one. A key goal for the scientist is to gain a clearer picture of those factors that actually produce a phenomenon. It has been suggested that systematic control is the key feature of science. Non-scientific approaches to knowledge are often made unsystematically and with little care. The non-scientific approach does not attempt to control very many factors that could affect the events they are observing (don’t hold conditions constant). This lack of control makes it difficult to determine cause-and-effect relationships (too many confounds, unintended independent variable).
The factors that the researcher manipulates, in experimental research, to determine their effects on behavior are called the independent variables. In its simplest form the independent variable has two levels. A variable is manipulated when participants / subjects are assigned to receive different levels of the variable. These two levels (or conditions) include the experimental condition; the condition in which the treatment is present and the control condition; the condition in which the treatment is absent. Only with experimental research can we determine cause and effect (or probability of causal relationship).
The measures that are used to assess the effect of the independent variables are called dependent variables (Shaughnessy & Zechmeister, 1990). Proper control techniques must be used if changes in the dependent variable are to be interpreted as a result of the effects of the independent variable. Scientists often divide control technique into three types: manipulation, holding conditions constant, and balancing. We have already discussed manipulation when we looked at the two levels of the independent variable. Holding conditions constant other than the independent variables is a key factor associated with control. This helps eliminate the possibility of confounds influencing the measured outcome.
Balancing is used to control factors that cannot be manipulated or held constant (e.g. subjects characteristics). The most common method of balancing is to assign subjects randomly to the different groups being tested. An example of random assignment would be putting names on a slip of paper and drawing them from a hat (flipping coin or number generator may also be used for random assignment). This does not mean there will be no differences in the subject’s characteristics, but the differences will probably be minor, and generally have minimal effect on the results.Reporting
How can two people witness the same event but see different things? This often occurs due to personal biases and subjective impressions. These characteristics are common traits among non-scientists. Their reports often go beyond what has just been observed and involve speculation. In the book Research Methods in Psychology (Shaughnessy & Zechmeister, 1990) an excellent example is given demonstrating the difference between scientific and non-scientific reporting. An illustration is provided showing two people running along the street with one person running in front of the other. The scientist would report it in the way it was just described. The non-scientist may take it a step further and report one person is chasing the other or they are racing. The non-scientist has a tendency to speculate more than the scientist. This type of reporting lacks objectivity.
Scientific reporting attempts to be objective and unbiased. One way to lessen the chance of biased reporting is checking to see if other independent observers report the same findings. Even when using this checkpoint the possibility of bias is still present. Following strict guidelines to prevent biased reporting decreases the chances of it occurring. Totally unbiased reporting, rarely, if ever occurs. Scientists are humans, and humans are susceptible to a wide range of conscious and unconscious biases.
In part 2 additional characteristics of the systematic approach to knowledge will be discussed.
To learn more about science, rationality and statistics read In Evidence We Trust: The Need for Science, Rationality andStatistics