One of the things I don’t like about most of the modeling curriculum I have experienced is the tendency to jump into taking data. In the summer workshops I attended at ASU and the several mini-workshops I have attended elsewhere, I was pretty much always asked to go take some measurements of things.
(1) Go measure some distance and time data
(2) Go measure some force and distance data
(3) Go measure some current.
(4) Go measure some frequency and length data.
And then we graph the data and look for relationships among variables. Next, we’ll often try to make a best fit curve to model the situation and interpret the meaning of the parameters in the equation.
The problem is that I always felt like I was just measuring things and looking for relationships without purpose. There were no puzzles we had identified as worthy of my epistemic curiosity. We had identified no perplexing questions that made me wonder, “what kind of data should I take to help answer that question?” What relationships am I expecting and why? How will I know if the measurements I’ve taken are good enough to either support or refute one idea or another? None of that was going on–not for me at least. And that was the thing that was puzzling to me.
Now don’t get me wrong–empirical data is important in science. But it alone is not science, at least not to me. I have been more apprenticed into starting science with something perplexing and letting that perplexing situation be the source of ideas, arguments, and explanations that need to be sorted out. In my mind, there’s now a reason to take data–that data will be EVIDENCE to support for claims. To me, the the subtleties that entangle and distinguish data and evidence are crucial for understanding the nature of science–both to do science and to be scientifically literate.
Leslie Atkins is fond of this quote that supports this view:
Observation and experiment are not the bedrock on which science is built, but rather they are the handmaidens to the rational activity of generating arguments in support of knowledge claims. (Driver, Newton, & Osborne, 2000, p.297)
Now, I’m certainly not saying that scientists never just muck around with data, or discover interesting things by looking for at relationships, or plotting data, or trying to make sense of mathematical equations. So what am I saying? I’m not exactly sure. Maybe, I’m asking someone to explain to me what I’m not understanding about the modeling curriculum.
Teach. Brian. Teach. 
Brian,
I totally hear this, and there are certainly times when my students go out to the lab solely to find the relationship between acceleration and total force, or position and time for a buggy. I’ve thought it might be possible to motivate this with a deeper questions, such as when would this buggy reach the end of the building, etc, but I’m not sure that would quite get my students on track for really exploring the relationship—I worry they might settle for finding an answer to the particular question and not generalize to find the larger relationship, which is at the heart of what I want them to do in the lab. I guess I need to find better ways to push them toward developing the general relationship from the specific example they are studying.
Yeah, in the big scheme of things I really like modeling, both for its use of mathematics in order to make sense of the world and for its focus on a few core ideas (models). At the same time, I think it possibly lends itself to a “science fair” view of science, where science is about deciding what variables to look at and then ‘testing’ for relationships. This, unfortunately, is the bread-and-butter of elementary school science fairs.
I have no experience using either paradigm, but one might be able to mash case/problem-based learning together with modeling to come up with something that sparks curiosity and plays off the strengths of modeling.