Wednesday, January 30, 2013

Reasoning in Context

First off, the best writer on inquiry might be an electronics teacher at a technical school in the far reaches of Canada. If you haven't checked her blog yet (despite my multiple recommendations), do so now and thank me later.

I wanted to get down some things I've been thinking about with argument and specifically with Reasoning in the Claim Evidence Reasoning format that I use. It's pretty fuzzy still.

In an earlier post I cited Berland and Reiser's three reasons for making a scientific explanation:

  1. Sensemaking
  2. Articulating
  3. Persuading
This year I've been reflecting on the changes I need to make to how my class approaches a written or verbal explanation depending on which of these three purposes is our focus.

Reasoning is hard. There's no getting around that. I think I've been making it harder on my students though because I've been using "reasoning" as an umbrella term for "the part that requires a lot of thinking." Brian, as always, was ahead of me here when he commented on the distinction between argument and explanation.

What I've been learning is that I need to be more explicit (ongoing theme alert!) and identify for my students what reasoning means in context. 


When we're engaged in a launch activity and students are using evidence to construct a claim, they're primarily focused on sensemaking. In this case, when I ask them for their reasoning, I'm really asking them to explicitly connect their evidence and claim in a consistent manner. I would expect my students to explain the patterns in their data and what those patterns might mean.

In a written prompt or lab practicum, I'm asking students to articulate their content knowledge. This is a test of what you've learned so far. In this case, when I say reasoning what I really expect is for my students to link scientific principals from class to whatever is on the page in front of them.

In whole class discussion and whiteboarding roundtables, we're engaged in persuading. Depending on the context, I would expect some combination of the above. But I would also expect you to directly address the opposing viewpoint and explain why your claim is better in some way.

It's also important for middle schoolers, and probably all ages, to understand that these different types of reasoning wouldn't just differ in content, but also in language. Google around and you can find lists of words that would appear in persuasive writing and expository writing. I use a lot of sentence frames and starters, but it's also a good self-check for students to go back and look for certain key words. I've got a couple of examples I'll put up in future posts.


  1. I read the first few words, "The best inquiry writer" and immediately knew. Can't agree more.

    I too am coming to a similar place with my students and writing–the need to be more explicit and to draw distinctions between different purposes. You also seem to be saying that those purposes arise in different but sometimes overlapping contexts of doing science, AND that the specific writing tools to be successful at each are different.

    Makes me think you need all three ingredients –establishing meaningful contexts, explicating a sense of purpose, and developing specific writing tools.

    This semester, I am carving out four different kinds of writing I want my students to

    * Articulating (Initial) Thinking
    * Elaborating upon Mechanism
    * Speaking to Claims
    * Constructing Definitions

    Maybe I'll write a post about it. Hope all is well.

  2. Aw, thanks guys. :) It's an honour to be in the club.

    Boy, do I ever know what you mean about the need to be more explicit (and more clear in my own thinking!) about the different kinds of reasoning. I laughed when I read your note that reasoning is "the part that requires a lot of thinking," and I've stopped using the word "reasoning."

    So far what I've asked students to do is "explain what could possibly be causing this," and "give some evidence from the model that supports this." I encourage them to come up with as many causes as possible (it sounds like this might overlap with Jason's category "sensemaking"). It's not the smoothest part of my day, and I find myself saying "Make sure you're talking about electrons," or else I get answers like "An LED's behaviour is caused by its design and construction."

    I wonder if my students would do better with the word "mechanism," as Brian suggests above?

    The idea of "persuading" overlaps with what I ask students to do when they find support from the model. In my class, we fall down when two mutually exclusive possibilities are both supported by the model. The students don't seem to know how to reconcile or resolve these differences, or how to determine if one of the arguments is stronger... and I'm not sure exactly what I'm asking them to do. Find a third possibility that allows for both sets of evidence? Notice a flaw in the reasoning of one position that makes it untenable?

    As for "articulating," I don't distinguish it from persuading. I've really emphasized the tentative nature of knowledge, and idea that there are no "facts" -- just ideas with more or less support. Nothing is "correct" unless it's persuasive. I wonder sometimes if this is a mistake -- making everything seem so uncertain.

    I'm also not sure how to help students use formulas well. Formulas can be considered to support an explanation, but they can not cause a phenomenon. This distinction is often hard for students to make.

    I'd love to be asking students to write definitions, but I'm not. It would probably help their causal thinking if I did. I'm missing opportunities to do this... maybe I'll start looking for them.

    1. re: mutually exclusive possibilities. Haven't you just summed up the current state of quantum mechanics?

      From the post, "In the end, all that matters is that your method of calculating predictions aligns with what you’ve observed." I'd probably say that 1) It's good that they understand that as long as the predictions work out, both models are equally valid. Leaving it unresolved is exactly what we have in science. And 2) It's your responsibility as a teacher to break one (or both) of the models unless the two competing ideas are actually where we are currently in the field. But isn't that what you're already doing when you're refining as you go? I guess I assume that an incorrect model would only yield correct predictions in a very narrow space.

      That being said, I leave my students with an incorrect (Bohr) model of an atom and nothing more than a warning and a brief throwaway about "smearing" and "probabilities" that none of them understand anyway but I'm doing for my own conscience.