There are two components of the Nature of Science that I hit hard at the beginning of each year at all levels (9-12) that I teach and I continue to focus on them throughout each course: how science works and how to distinguish hypotheses from predictions. These concepts have often been oversimplified for students and can disrupt their ability to do science inquiry well.
The “Scientific Method.” Many teachers present to students the scientific approach to understanding the world as “the scientific method” and then have students practice a series of steps over and over and over again. This approach leaves students with the impression that all scientific knowledge of the natural world is generated by the exact same linear recipe. However, the process of science (and engineering) is far more dynamic than a simple method and the approach one takes to answer a question or solve a problem is unique to each situation.
HHMI BioInteractive, in partnership with “Understanding Science” at the University of California Museum of Paleontology, has produced a fantastic tool called How Science Works that students can use to illustrate their journeys through scientific inquiry. An example output is pictured below in Figure 1. Teachers can also have students learn about historical and new scientific discoveries and use the interactive to map the discovery process. With this tool, students can easily visualize how science works and can begin to understand that scientific inquiry and the path to knowledge can work in an almost limitless number of ways.
The Hypothesis. Another oversimplification in science education that has led to misunderstanding by students is how we have tried to make hypothesis writing for students easy by giving them a formula to follow: the classic “If…, then…” This formula almost forces unsuspecting students to write predictions at the expense of actual hypotheses: “If I do X, then Y will happen.”
The concept of the hypothesis in science has its origin in the 16th Century and is defined by Oxford as “a supposition or proposed explanation made on the basis of limited evidence as a starting point for further investigation.”
My suspicion is that the “If…, then…” formula emerged from the research hypothesis: “If X is a valid explanation or pattern for these observations I’ve made, and I perform Y controlled methods to collect data, then Z are the specific data I will likely collect.” In other words, “If hypothesis, and planned experiment, then likely prediction.” In scaffolding and simplifying the hypothesis for students, textbook writers and thus teachers have dropped the hypothesis component of the research hypothesis and left the statement as simply a method followed by a prediction. Again, “If methods/experiment, then prediction.” For example, the statement, “if I add salt to water, then the freezing point of the solution will be lower,” is a classic “textbook example” literally found in many middle school science textbooks. However, the statement is simply a planned test followed by a prediction. There is no obvious underlying explanation or pattern being tested. Students miss learning anything about why a solution is different than pure water in the context of freezing points.
Science Education specialist, Anton Lawson has written extensively on the nature of the hypothesis and has been recognized numerous times for his contributions to improving science education. However, most of Lawson’s papers are buried in journals that tend to be out of sight and reach of classroom teachers, yet many science teachers would benefit from reading Lawson. Indeed, there is so much confusion among teachers and students that Google searches for the hypothesis peak each year in September in the U.S. when school has started in most districts (Figure 2).
While I’ve never met Anton Lawson, my favorite breakdown of his on the research hypothesis come from two of his papers. In one paper, “The Nature and Development of Scientific Reasoning: A Synthetic View,” Lawson describes the scientific reasoning involved in a 1960s study on salmon migration. Among several hypotheses tested in the study on how salmon find their way back from the ocean to their home stream, Lawson shows how the research hypothesis might be structured in the context of the salmon migration question (Figure 3).
In another paper, “Sound and faulty arguments generated by preservice biology teachers when testing hypotheses involving unobservable entities,” Lawson reveals the scientific reasoning used by Gregor Mendel during his heredity experiments on pea plants (Figure 4).
In the same paper cited above, Lawson also shows that, in addition to explanations, nature’s patterns can be morphed into a research hypothesis. I find the example below (Figure 5) especially helpful because most testing of hypotheses students do in the science education classroom is the testing of nature’s patterns or the confirmation of a physical constant/law.
From the three above examples, you might notice that there are two kinds of hypotheses in science: the explanatory hypothesis and the generalizing hypothesis. The explanatory hypothesis is an explanation for an observation. Examples of explanatory hypotheses are shown in Figures 3 and 4. The generalizing hypothesis is a description of a suspected pattern in nature. The pendulum example in Figure 5 is a generalizing hypothesis. Experiments inspired by generalizing hypotheses test to see if suspected patterns are real and universal under defined conditions. Explanatory hypotheses test suspected mechanisms, often the mechanisms that are driving the patterns.
Teaching the Hypothesis. In teaching the hypothesis in the classroom I’ve found it is most helpful if students do not write comprehensive research hypotheses when planning their investigations, but instead delineate the hypothesis from the planned experiment and prediction. Below is a screen shot from a student’s paper. The student’s reasoning is clear and she shows where her hypothesis ends and her prediction begins. I also like the added predicted graph. But this year I will be moving my students away from column graphs and toward dot and box plots. But that’s a future post.
Last Thoughts. 1) Science inquiry cannot be reduced to a single method and thus should not be taught that way. Instead, students should come to understand science as a process that is unique to each question, problem, and moment of curiosity. 2) Hypotheses should not be required for every student investigation. Many inquiry endeavors are discovery science or engineering attempts and do not involve generating hypotheses at the beginning, if at all.