Developing Questions and Hypotheses (Snowpack)

By the end of this unit your students will have a research question, and a well-reasoned scientific claim that can be tested (supported or not supported) with data.
By the end of this unit your students will have a research question, and a well-reasoned scientific claim that can be tested (supported or not supported) with data.

Unit Overview

Now that students have learned a bit about stream ecology and culverts, how stream macroinvertebrates have certain habitat needs, and how culverts might affect stream systems, they can begin to think about what questions they might ask and then begin to answer with their fieldwork.

The questions we are heading toward are those that fall under one overarching question:

  • How does the nature of snowpack and timing of snowmelt differ in the different climate divisions of the State of Maine?

For example:

  • Do
  • Is
  • How do
  • How does forest type
  • How does

In this Unit your students will refine their own questions and propose a possible answer, in other words, state their hypothesis or scientific claim.

When the students have completed the research process they will be able to share their claim (or hypothesis), the evidence they have for their claim, and their argument for whether or not their claim was supported. They will be presenting the “Claim-Evidence-Argument” necessary for a scientific story. However, for the building of a claim the students need to apply what they know about the system to a question and make a case for their claim. While the end result of the research project may take the form “Claim-Evidence-Argument”, the claim needs to take the form of “Evidence-Claim-Argument”, or “If, Then… Because”.

Goals & Challenges

The goal is to guide the students toward a specific research question and claim that they can investigate by setting up an experiment, collecting data, and analyzing the results. You will need to guide this process so that the students’ questions meet a few important criteria:

  • Owned by the Students

Students should develop their own questions because they will learn from doing so and will be more invested in the research outcomes.

  • On the Right Topic

On the other hand, the students’ questions need to be consistent with the learning objectives for the course. For example, if the course is focused on chemistry, you might not want to let a student pursue a question about vegetation type, unless the student connects that question to chemistry.

  • Pedagogically Useful

You will also want to be sure that the questions that students are pursuing are ones that they can hope to investigate given the time and equipment that is available and given the students’ level of expertise. You are trying to direct the students toward a positive outcome, and you want to steer them away from paths that are likely to end in frustration or in a trivial, uninteresting result.

The degree to which you want to interfere and “correct” a model is another one of those judgment calls.

Balancing these objectives requires judgment, familiarity with the scientific principles involved, and practice. You are acting as a guide for the students on an intellectual adventure. Some journeys inevitably turn out to be more interesting and exciting than others. The more times that you lead these trips, the more you will know about what works. This is one of the key reasons that we are asking you to share your experiences, both good and bad, with other teachers. By pooling experience we all can learn from each other.


A hypothesis should grow out of some kind of a model of the system or mechanism under investigation. A student ought to have some idea as to why the outcome predicted by the hypothesis would happen.

For example, suppose a student offers a hypothesis that there will be more caddisflies below the culvert than there will be above the culvert. Then suppose you ask the student to explain why he or she expects this outcome. If the student simply says, “I don’t know, it’s just a guess,” then there is no model in place, and when the student actually gets the data back from the lab all that he or she will be able to say is either “Good, I was right,” or “Huh, I was wrong.”

But if the student has a model of how things work that is informing the hypothesis—saying, for example, that since the water will be more slow moving above the culvert than below the culvert, and caddisflies like faster moving water, then there will be more caddisflies below the culvert than above the culvert; then there is the possibility of having some learning happen. If the data support the hypothesis, then the student might recommend that future researchers see if there are any of the same individual species of caddisflies above and below the culvert. If the data do not support the hypothesis (often a more interesting outcome), the student might re-examine the model that they were using and suggest alternate lines of investigation. The model behind the hypothesis gives the student something to build on.


Students will understand that:

  • A hypothesis (or scientific claim) grows out of some type of model
  • Models are generated through relating a series of interactions in the system
  • Hypotheses are neither right nor wrong; they are either supported or not supported
  • Evidence to support or not support a hypothesis can be collected by setting up an experiment, collecting data and analyzing results
  • Investigations are strengthened when researchers participate in peer review

Instructional Strategies

We have found that:

  • Students have difficulty coming up with hypotheses that they can work with. An approach has been to have the students review their learning so far and then ask “I wonder…” and then “I think…”
  • Developing testable questions and hypothesis is an iterative process.
  • Stating hypotheses as “If, Then…Because” statements helps students to clearly articulate their hypothesis.
  • A hypothetical graph helps students clarify their thinking. It is a good idea to have students make a graph or chart of what they think the results of their data gathering might be. This will give them an indication of the viability of their hypotheses. This should a “back of the envelope” type of graph, drawn by hand. The students will need to decide what information goes on what axis and then plot what they think the data will show. As one teacher says “No graph? No question.”

For example: If a student hypothesized that since the water will be more slow moving above the culvert than below the culvert, and caddisflies like faster moving water, then there will be more caddisflies below the culvert than above the culvert then the hypothetical graph might look like:

Culverts Drawing

Words to avoid

Fact, prove, true, right, wrong, good, bad

We want to avoid using words like those above because they reinforce the notion that science and the interpretation of data are absolute, or black and white. Really, we are gathering evidence that supports or does not support a hypothesis. We can never know everything about every part of a system, so there’s always a chance we’ll find out something new down the road that will change how we think the system works. For example: Newton thought that light was made up of particles – for good, logical reasons. Later, when scientists observed diffraction, they started to wonder if it wasn’t a wave instead. We now think it travels as both particles and waves. Newton wasn’t wrong. He was basing his hypothesis on sound evidence, but as new evidence came to light our understanding shifted somewhat. It may shift again as we record new observations and phenomena regarding the properties of light.