Well, not really. Science isn’t really a verb, because if it were, I’d be able to find the phrase “to science” in the dictionary, which I cannot. However, if science is traditionally defined as “the study of nature by means of the scientific process . . . ,” obviously the process by which nature is studied is at least as important as the fact that the object of that study is nature. In fact, it may well be more important, because, as our tools improve for the analysis of human thoughts and emotions, the traditional definition of science is gradually being expanded to include various human endeavors as well as nature.
So, what is the scientific process? And how is this process related to the more familiar “Scientific Method,” of which so many of us have fond memories from our secondary education? The answer is that the phrase “scientific process” can be used interchangeably with the phrase “scientific method,” provided that everyone understands the following: (1) scientists, or those who actually utilize the scientific process, often find the phrase “scientific method” irritating because of its completely inaccurate implication that there is one, and only one, method for studying nature; and (2) the “Scientific Method” taught in schools is far more simplistic than the true scientific process/method.
If you’ll recall, the “Scientific Method” encountered in school consists of the following steps that students are required to satisfy when performing an “experiment” (I’ve enclosed the word, experiment, in quotes because such assigned activities are rarely experimental but rather laboratory exercises with known outcomes):
Students are first required to write down the question they are attempting to answer or the problem they are attempting to solve. Such questions might include “What is this object?” or “When/where does this event occur?” or “How does this object work?” or “Why does this event occur?”
Students are then asked to propose a hypothesis, often defined as an “educated guess,” that answers their question, and then to write it in such a form that it will predict the outcome of an experiment. Such a prediction is usually written in an “if, then” format that expresses how two factors are related to each other (“If A occurs, then B will occur.”)
Once a hypothesis has been proposed, students are required to perform an experiment, or a “test” of the hypothesis. They are advised that the design of an experiment must include three types of variables – the independent variable, the dependent variable, and the control variables. The independent variable is the variable that the student intentionally changes during the experiment, and there can only be one of these. The dependent variable changes in response to the changes in the independent variable, and changes in both the independent and dependent variables must be accurately measured. The control variables are those that remain the same throughout the experiment. In some instances, a “control experiment” is required in order to establish that no other variable besides the independent one is changing the dependent variable. While conducting the experiment, care must be taken that all materials, observations, and other data are recorded accurately and precisely.
(4) Data Analysis and Conclusion
Finally, the students are asked to carefully examine their recorded data, using statistical analysis when appropriate to determine the uncertainty or error of the data (data may be held invalid if such parameters are too great). They are encouraged to use tables and charts to organize and summarize data, with bar charts and line graphs being effective tools for comparing changes in the dependent variable with changes in the independent variable. Conclusions are in the form of whether the data support or falsify the hypothesis. If they falsify the hypothesis, an explanation must be proposed as to why this occurred, often in the form of another hypothesis that can be tested. (The definition of “to falsify” in this context is “to contradict” or “to refute” – not “to alter or to add to with the intent to deceive.” Falsifying a hypothesis is perfectly acceptable . . . falsifying data is not.)
At this point, you might be asking exactly what is wrong with this method. It seems simple and straightforward and . . . well, that’s pretty much what is wrong with it. As stated on the “Understanding Science” website, the “Scientific Method” “can easily be misinterpreted as linear and ‘cookbook’: pull a problem off the shelf, throw in an observation, mix in a few questions, sprinkle on a hypothesis, put the whole mixture into a 350° experiment – and voilà, 50 minutes later you’ll be pulling a conclusion out of the oven!” In other words, although this method does represent the primary logic of science – testing idea-generated expectations by obtaining physical evidence – it is much simpler than the process used by most scientists, and many scientists don’t even agree that this method is the primary way in which scientific knowledge is acquired.
Take, for example, the question/problem. This is almost a no-brainer for students, because the teacher has generally outlined the point of the exercise they are about to perform; hence, the student need only phrase it as a question or problem to fulfill the first requirement of the “Scientific Method.” Scientists, on the other hand, must come up with their own questions/problems, and the only way they can question is if, at a minimum, they have: (1) directly observed some sort of interesting or surprising phenomenon; (2) discovered an interesting or surprising phenomenon in their reading or discussions with others; (3) recognized that a need exists for an as-yet-uninvented object/technique; or (4) already tried to solve one problem, which has revealed yet another question/problem.
With respect to the hypothesis/prediction, a hypothesis is indeed a tentative explanation that can be tested by further investigation, but it is not necessarily an “educated guess.” In the case of classroom investigations, most students simply don’t have enough background in the appropriate subject to offer anything more than a guess – minus the education. Certainly my usual reaction as a student to this step was: “How am I supposed to know the answer to this question? Why can’t we just do the experiment and then figure out what the answer is?” Of course, one solution to this problem would be a requirement that students spend a day or two researching a question in the library or on the internet before formulating a hypothesis . . . but this doesn’t always happen. In the case of scientists, most will usually undertake literature searches, or brainstorm with fellow scientists, as to what and how similar questions have already been approached and tested. Hence, there is usually some sort of “education” behind their hypotheses, if only to avoid investing time and money into investigations that have already been performed. The resulting hypotheses, however, can range from logical extensions of previous investigations (in which case they can hardly be called “guesses”), to highly creative but still plausible explanations that spring from looking at problems in new and different ways (which is when science really gets interesting), to my personal favorites, WAGs (wild-ass guesses).
In addition, more than one hypothesis may be proposed (“maybe A is the answer . . . or B . . . or C”), or a scientist may have no hypothesis at all, particularly when he/she is the first to think of the question (“I don’t know what the answer is, so let’s just see what happens if I change this variable!”). Yes, just as I wished when I was a student, sometimes a hypothesis isn’t proposed until after an experiment rather than before. In fact, there are those in the scientific arena who argue that a hypothesis shouldn’t be proposed until after the first experiment, and that more than one hypothesis should always be considered and tested. Finally, unlike in the classroom, even if a hypothesis is proposed before the first experiment, it is rarely written down until it has been thoroughly tested, serving instead as the mental framework for a scientist’s investigations. It is arguable, of course, that all hypotheses should be written down in order to preserve the history of an investigation, but doing so may also bias a scientist too much in favor of his/her favorite hypothesis; hence, preserving a hypothesis in writing is rarely done unless a certain amount of evidence already exists to support it.
The process of experimentation is also more complex than the “Scientific Method” might indicate. Under ideal conditions, an experiment is an attempt to simulate a phenomenon under controlled conditions . . . however, it is not always easy to keep control variables constant; and it is not uncommon to have more than one independent or dependent variable, particularly when experiments are performed out in the field. In addition, the results of just one experiment can never be taken as absolute, i.e., an experiment is usually performed two or more times by the same scientist before he/she draws any conclusion with respect to a hypothesis. Finally, experimental results must be replicated by different scientists in order to expose any random errors (such as inaccurate perceptions or measuring instruments) or systematic errors (non-random errors that bias the results of an experiment in one direction), which is why it’s so important that scientists record in some detail how their experiments are conducted.
While our food scientist proceeds to make her chocolates,
It is also interesting to note that, as important as experimentation is in modern-day science, it wasn’t always the cornerstone of the scientific process. The natural philosophers proposed explanations based on observation alone, and attempts to imitate or apply natural principles was considered the territory of lower class artisans (perhaps because “doing” was not as exalted as “thinking”). The first known philosopher to champion experimentation and its reproducibility was the Arab scholar Ibn al-Haytham (965-1039), who made significant contributions to mathematics, physics, anatomy, ophthalmology, psychology, astronomy, and engineering. Experimentation was then gradually adopted by more and more investigators in the Muslim world, but it wasn’t until the 13th century that experimentation was discussed and implemented in Europe. Among others, English philosopher Roger Bacon (1214-1294) strongly argued for a process similar to the modern scientific method, involving observation, hypothesis, experimentation, and independent verification; and philosopher Francis Bacon (1561-1626) not only promoted the importance of experimentation over contemplation, but was also one of the first in Europe to maintain that science should be used to improve the existence of humankind. In the 17th century, physicists Galileo Galilei (1564–1642) and Isaac Newton (1642–1727) began to describe and analyze nature mathematically, moving the study of natural phenomena from the sole province of language, which can often be confusing and misinterpreted, into the province of numbers, which is usually less ambiguous. During the following centuries, those who studied nature increasingly utilized measurement and experimentation, but it wasn’t until the beginning of the 20th century that the scientific process/method was fully accepted as an essential part of “science.”
Our food scientist’s sister is also experimenting . . .
Similar to the use of hypotheses, however, even today experimentation isn’t always a part of the scientific process. Under many circumstances, such testing may be impractical/impossible, cost-prohibitive, unethical, or even insufficient to confirm or falsify a hypothesis . . . in which case mathematical calculations, often used in physics and astronomy, and observational studies, which typically involve description, measurement, inference, categorization, and/or pattern recognition, assume much more importance. Fortunately, observational studies today are much more accurate and reliable than they were in the past, due primarily to the invention of numerous instruments that enhance our senses, such as microscopes, telescopes, x-ray crystallography, mass spectroscopy, radar, thermometers, radiation sensors, and others. These tools can detect many more phenomena, and make much more precise measurements, than we can with our senses alone. Thus, observations can not only provide much important information on their own, but they can also allow the refinement of general questions to more specific ones, result in previously unthought-of hypotheses and experiments, and even confirm or falsify hypotheses without the need for experiments.
Our food scientist’s friend laughs . . .
With respect to the analysis of data and conclusions, many more issues than are normally addressed in the classroom must be considered by scientists. For example, there may be more than one way to interpret data that is extremely complex. In addition, the data may not always confirm or falsify a hypothesis – sometimes it is inconclusive or even completely unexpected, which means that not only might the hypothesis need revision, but also the assumptions underlying the hypothesis or the question itself. Occasionally, data will support more than one hypothesis, in which case the hypothesis with the greatest testability, application, parsimony (simplicity), and consistency with hypotheses in similar disciplines will most likely be favored and retested first. Finally, the replication of experiments or observational studies by other scientists, or simply the critical examination of data by other scientists (peer review), may also reveal that the data supports a different conclusion than that reached by the original investigator. Yes, even though scientists try to evaluate their data as objectively as possible, they are human like the rest of us, and may either have a strong belief that a particular hypothesis is correct, or be under pressure to produce data supporting that hypothesis. In such cases, a scientist might, consciously or unconsciously, ignore data that falsifies a hypothesis – a failure that will be caught by a more skeptical analysis.
And our food scientist’s colleagues debate . . .
Whether we call it the scientific process or the scientific method, it’s obvious that this process/method doesn’t always follow the strict series of steps described in the classroom. Indeed, as much as scientists attempt to approach the study of nature in a logical, orderly way, many of their investigations are no more methodical than trial-and-error. Additionally, as mentioned in my previous post, the role of serendipity (fortuitous accidents) in the scientific process also cannot be underestimated . . . although, as microbiologist Louis Pasteur stated in the mid 19th century, “Chance favors only the prepared mind.” Finally, science today is a far more social endeavor than it has been in the past, primarily because a team of scientists is much more efficient at obtaining scientific evidence than a sole investigator: for example, the more imaginative or creative members of a team may think of questions and hypotheses; the more mechanically precise may run the experiments or observational studies; and the more mathematical or detail-oriented may analyze the data. Such teams work on puzzles together, collaborating and competing, until all the pieces fit and those that do not can be safely thrown out. The popular view of the scientific process is that it’s an emotionally cold, purely logical, sterile endeavor, but to those who actually “do” science, it’s pretty much like any other profession – often tedious, sometimes frustrating, frequently satisfying, occasionally surprising – with the exception that no other job results in the discovery of previously unknown information . . . which, for a certain type of person, is the biggest thrill there is!