Paradigms, Puzzles, and Citizen Science in Schools

A month or so ago I finally got around to reading Thomas Kuhn's classic, The Structure of Scientific Revolutions. I say "finally" because the book was first published in 1962, while I was still in junior high school. The book was--and still is-- widely read and referenced. As an indicator of the book's impact, consider that the term "paradigm shift" came into common usage through this book.

The Structure of Scientific Revolutions (I am going to call it TSSR) is engaging--the best book I have read in quite a while--though not an easy read. Kuhn refers to it as an "essay." The term is apt, because TSSR is a closely reasoned argument for what was, in 1962, a new view of science.

It's worth spending a moment to summarize the old view and the new one that Kuhn sets forth in TSSR because, despite TSSR, the old view is still alive and well in science education. The contrast of views also raises interesting questions about what we should be doing in science education and about how citizen science might fit into that program.

The Old View

Kuhn was a science historian, which is why a key question in TSSR is how we should understand scientific progress. Kuhn describes the traditional, pre-TSSR view of scientific progress as follows: (All quotations are from the paperback, third edition of TSSR. I also note that Kuhn only refers to all scientists as men ... sigh ... that was then.)

If science is the constellation of facts, theories, and methods collected in current texts, then scientists are the men who, successfully or not, have striven to contribute one or another element to that particular constellation. Scientific development becomes the piecemeal process by which these items have been added, singly and in combinations, to the ever growing stockpile that constitutes scientific technique and knowledge. (p. 1)

The key image here is the constantly growing stockpile of knowledge. In this traditional view, the story of science is one of incremental accumulation in which, as each decade and century passes, we are a little closer to complete understanding.

The work of science education, given this view, would be to introduce students to the knowledge in that pile. Kuhn's use of the word "constellation" acknowledges that the stockpile is structured. The word "edifice" is sometimes used to describe this view of science: it is a large, complex building constructed brick by brick since the time of Aristotle. Science education should then be about introducing students to the structure and content at the foundation level.

This view of science matches pretty closely with what students tell us when we survey them about the nature of science. It is a pile of facts about the world acquired through a process called the scientific method.

Kuhn's View: Paradigms

TSSR argues that this picture of science is not supported by historical fact. Kuhn makes this argument by looking at examples. The transition from the Ptolemaic, earth centered view of the universe to the Copernican view did not add another brick to the edifice of science; instead it completely rearranged the bricks. Similarly, the movement from a phlogiston theory of combustion to Lavoisier's oxygen theory was a complete turn around in thinking about what happens when something burns. Or, to take another example, the quantum view of light as electromagnetic radiation completely displaced wave theory and a belief that there must be something called "ether" to support the propagation of light waves.

Kuhn recognizes that these upheavals, as important as they are, are the exception rather than the rule. He distinguishes between "normal science" and the much more unusual condition of science in a time of revolution. Normal science is where most of the day-to-day work of science gets done. It is where the steady, cumulative work happens.

So, you have normal science working away for hundreds of years, piling up observations, most of which fit with the theories in place at that time, but including a few that don't fit. When the things that don't fit are numerous or important enough, the normal scientific work enters a time of crisis, followed by a revolution. Then normal scientific work returns. The revolution in thinking about geology in the 1950s and 1960s, from a geosyncline theory of movement of the earth's crust to plate tectonics, is a recent example of such a revolution followed by a return to productive, normal science.

Kuhn asks whether the scientists working before such a revolution--say, on the phlogiston theory of combustion--were somehow any less "scientific" in their work. Take, for example, Joseph Priestley, one of the people credited with discovering oxygen. He went to his grave defending phlogiston theory, despite his discovery. Does that make him less a scientist? Was his application of "the scientific method" somehow less complete than Lavoisier's?

Of course not. The point is that there is more than "scientific method" that binds normal scientific work together in any discipline at any point in time. Kuhn calls the collection of values, beliefs, methods, shared examples, and shared problems a "paradigm." What separated Priestley and Lavoisier was immersion in two different paradigms.

The word "paradigm" has come into common use thanks to Kuhn--and now means different things to different people. In understanding Kuhn's use of "paradigm" it is helpful to start with his view that scientific work is primarily about "puzzle solving." Contrast this with "discovery." Think of the image of Sir Issac Newton sitting under a tree. An apple falls and Newton suddenly has an "Aha!" moment in which the law of universal gravitation pops into his head. That's discovery.

What actually happened is that Newton went to work on the problem of explaining what was already known from Kepler's observations about the motion of the planets. The law of gravitation was the resulting solution. That's puzzle solving.

"Discovery" implies surprise: you find something that you didn't know was there. Puzzle solving, on the other hand, implies that you know in advance what the solution will look like--the puzzle is in how to get from what you know to the desired solution. Puzzles also typically come with constraints. In a crossword puzzle, all the words need to share letters. In su doku every each number can only occur once in a row, column or square. In Rubik's cube you can't take the cube apart.

A paradigm gives scientists a picture of the solution--of what they need to figure out--and the constraints that must be satisfied in reaching that solution. Scientists working within a paradigm generally know where they are headed and know the accepted practices they will use to get there.

Thinking of science as puzzle solving is very different than thinking of it as discovery. It is an important difference when we think about science education.

Paradigms and Learning Progressions

Kuhn tells us that paradigms are essential to scientific progress. They provide a "map" for scientists to follow; normal scientific research is the exploration and elucidation of the detailed landscape charted by that map. Moreover, "paradigms provide scientists not only with a map but also with some of the directions essential for map-making." (p. 109)

When scientists try to work without a map (which is how science MUST work before a paradigm is established and broadly accepted) their writings are like those of the blind men describing an elephant. They talk past each other, working from different assumptions about what is real and what is important. We see some of this when we watch students present their research. There is a lot of "It could be this, or it could be that" without a way to connect to each other's work).

The other thing that we see when students engage in research without a good map of the landscape is confusion and frustration. (See the parable "Joe and the Hundreds of Holes.")

Thinking about all of this in terms of science education, it seems that TSSR is giving us important insights into the learning progression required to support student inquiry: We must somehow introduce students to the paradigm that encompasses the research they will undertake. How do we do that? If it were really a physical map, we could just draw it. But a paradigm is a map in a broader, metaphorical sense. How do we help students see the problem space covered by the paradigm? What are they key elements? How do they connect?

Kuhn suggests that the answer lies in the direction of having students solve a common core of problems. He notes that "Scientists solve puzzles by modeling them on previous puzzle-solutions." These "puzzle-solutions" become what he calls "exemplary problems." The value of having students work problems, either with pencil and paper or in the context of structured (contrived?) research settings is not just that they get better at solving problems: Working the problems introduces students to the accepted paradigm. They learn what to pay attention to, what to ignore, and how to see the world:

After he has completed a certain number [of exemplary problems], which may vary widely from one individual to the next, he views the situations that confront him as a scientist in the same gestalt as other members of his specialists' group. For him they are no longer the same situations he had encountered when his training began. He has meanwhile assimilated a time-tested and group-licensed way of seeing. (p. 189)

Citizen Science

My organization works with teachers to engage students in citizen science rooted in research in the National Parks. In the Northeastern United States, mercury pollution is impacting ecosystems. The students in the schools that we work with collect samples that we analyze for mercury content. Gathering information about mercury levels in target species in different ecological settings across the Northeast helps us construct a better picture of how mercury moves and where it settles. The idea is that as the students help scientists out by collecting samples, they learn earth science, biology, and chemistry. They also learn about their National Parks and--just as important--learn what scientists do and how they do it.

This kind of work--outdoors, working with issues that matter outside the classroom, engaged in research where no one (including the teacher) knows what the result will be--is a new experience for students and teachers. We see evidence of increased student engagement, but also see evidence that students need more preparation and background knowledge to get the most from the experience. (Again, see the parable "Joe and the Hundreds of Holes.")

Kuhn's focus on paradigms and on exemplary problems as a way to assimilate them--as a "time-tested and group-licensed way of seeing"--raises two important kinds of questions about the intersection of citizen science and science education:

1. Citizen Science as Exemplary Problem: If we think of citizen science as a means to science education, we will ask, "Is the citizen science work a good exemplary problem? What is it an example of? Where does it lead? How do students build on the experience with us to get better at working these problems?"
   
   
2. Exemplary Problems to Support the Citizen Science: Thinking instead of the citizen science as the goal--something that we want to build scaffolding in earlier grades to support--we will ask, "What kinds of exemplary problems should students work on in preparation for the citizen science experience? When should they do this, and how long will it take for them to acquire the "group-licensed way of seeing" that they need to engage productively in the citizen science?"
   

These two kinds of questions could be asked with regard to any activity in science education. Restated, they are "What future learning does this activity support?" and "What support needs to be in place before I do this activity?" But the questions come into particularly sharp focus for citizen science because students are engaged in real, useful research--not just schoolwork--and are doing it in collaboration with working scientists. Our program's value as an example of scientific work is why teachers sign on.

I don't have answers to these questions. Some of them have to do with what science educators and schools want to achieve. Others have to do with what works, what's possible, and what's necessary. All are questions that we need to explore over the next couple of years.

Hearing from others--both in response to these questions and in the form of additional questions--would be a good way to start on this work.

Joe and the Hundreds of Holes: Ruminations on Inquiry Learning

Twenty years or so ago, when a was in charge of design and engineering in a small software firm in Boulder, Colorado, I had an employee named Joe who was from New York City. He was so completely from New York that he had never learned to drive, even though he was in his early twenties. Boulder was not and still is not the kind of place where you can hail a cab. So, Joe had a bicycle.

Bikes have been popular in Boulder for a long time. The town has bike lanes everywhere. So, especially since software engineers tend to be young, Joe was one of a good number of people who rode a bike to work. They were pretty nice bikes, and people brought them inside where they would be safe. We had a lot of space in cubicles and aisles filled up with bikes.

To get the bikes out of the cubes and free up space I got together with my senior engineer--who was actually a semi-pro bike racer--and fabricated racks that would allow us to hang a good number of bikes on a wall back in an odd shaped corner of the office that didn't get much use. (The office was converted from the garage and service area of an Alfa Romeo dealership. High ceilings, industrial skylights, and lots of oddly shaped spaces.)

Joe was one of our most junior engineers and so was assigned the task of mounting the racks on the wall. I brought in an electric drill, bits, and screws and set Joe to work. I told him that the racks would hold a fair amount of weight, which meant that they needed to be screwed directly into the studs behind the wallboard, rather than just into the wallboard itself. I suggested that he use a small drill bit--1/8" or so in diameter--to drill test holes to make sure that he was drilling into a stud. Then I went back to my office.

Some time later--I am guessing that it was the better part of an hour--I looked out over my cubicle to the corner where Joe was working. I expected to see at least some of the racks up on the walls.

Nothing.

So, I went back to have a look. What I saw was Joe still drilling away on test holes to locate the studs. Joe was a software engineer, and so this was very neat and well-organized work. He had drawn a horizontal line about 3 feet off the floor and was methodically drilling 1/8" diameter holes along that line, spaced 3/4' apart, and marking the places where he had found a stud. The wall looked like a "fold and tear here" perforation on a statement from the power company. Without question, Joe was finding every stud in that wall.

Expectations and Focus

Joe was--and surely still is--a bright guy. Despite this, I had a nearly overwhelming impulse to say something including the word "stupid" when I saw him kneeling there drilling away. On the other hand--this whole investigation was so carefully laid out: the penciled line, the evenly spaced holes. It was beyond stupid.

What Joe was doing was a little bit like what some students fall into as they design their own experiments in our inquiry learning program. It is a citizen science program in which the students explore mercury levels in plants and animals in forests, ponds, and marshes around their schools. Their work can be carefully thought out, but also somehow misdirected -- heading off in surprising directions that are not likely to be fruitful.

I think that their problem, like Joe's, is that they are missing information about the bigger picture of the system that they are working with.

Joe was missing information in two important areas. The first was that he needed a working model of the structure of the system hidden by the wallboard--an expectation about the likely arrangement of the studs. He apparently did not know that studs are, for the most part, evenly spaced. If studs could be just anywhere it would make sense to drill lots of holes. But if you knew that the spacing is usually regular, you could predict where the next stud would likely be, given the location of the first two studs.

It would also have been helpful for Joe to know that there is usually more than a foot between studs. Even better, he could have come to the job with the information that, in many walls, studs are placed on 16 inch centers. (Or I could have told him that.) This knowledge about typical wall systems would have enabled him to formulate a hypotheses about the location of the next stud after finding just one. Testing the hypotheses would be a simple as measuring out 16 inches from the first successful hole to drill the next test hole. This is what I had expected Joe to do, since it was what I would have done. My mistake was in not telling Joe what I knew about walls that allowed me to formulate such a hypothesis.

The second general kind of information that Joe needed had to do with the larger objective of his inquiry. He needed to stay focused on his goal. Even without knowing anything about the way that walls are built, he did know that we were intending to hang a bike rack on the wall and that the two uprights would not be side-by-side, but would instead be spaced apart so that they could support bikes. Our objective was to hang up a bike rack, not to locate every stud. So, even without knowing anything about the usual spacing of studs, Joe could have avoided drilling holes in the wall right next to the first stud simply by focusing on the goal of the research.

Student Research

Both of these observations apply to some of the student work we have seen this year. Sometimes students are working without a good, rough model of what is probably "behind the wall." Other times they run into trouble because they lose sight of their research objective and get lost in the details, effectively drilling hundreds of little holes to no end.

Take, for example, the efforts of a number of students who looked at mercury in soil. In some of the projects, which were carefully executed and presented, the students focused on the idea that mercury comes down in the form of rain and so hypothesized that there would be less mercury in soil under trees than in the open. Sometimes their reasoning was that the trees sheltered the area around them and sometimes it had to do with the idea that the trees would remove mercury from the ground as they took up water from the soil. What these students were leaving out--evidently because it was not in their own models of the system--was the important effect of dry deposition, where trees act as big filters, picking up mercury attached to particulate pollution which is later washed off the leaves onto the ground below.

As another example, several student groups looked at the amount of mercury in individual fish, tabulating mercury and fish length, fish weight, and fish age without any strong reference to fish species, beyond recording it. In listening to these students talk about their work, it seemed that in some cases they did not really have the idea of food webs and bioaccumulation included in their mental models of what might be going on. Like Joe and the wall studs, it seemed that the students thought that mercury could end up anywhere, and their job was to find it in all these different places. In other cases their diffuse focus seemed to be more a matter of losing sight of the objective--or perhaps not having ever formulated a firm hypotheses to guide their investigation.

Two Very Different Approaches

For me, these examples raise the question of what we want these students to get out of this fieldwork that we are having them do. That question then leads to a second one, having to do with what preparation: What intellectual "scaffolding" do we need to put in place for them to reach the learning goals that we have in mind?

Going back to Joe's example, if I had been wanting to engage Joe in self-directed inquiry, maybe I was actually off to a good start. Seeing the line of hundreds of carefully spaced holes, I might have used this as a clue to his conception--his "mental model"--of how the studs were arranged behind the wallboard. I might then have guided his inquiry a bit at that point, suggesting that it would be interesting to see if there was any pattern in the arrangement of the studs. Given this prod--a "nudge' in direction of discovery--perhaps Joe would have then come up with the idea of measuring the distance between successful test holes, looking for regularities. Looking at the empirical data that he was so carefully gathering, he might have reached the insight that studs are often on 16" centers. Knowing Joe, he might have come up a very un-carpenter-like 40.6 cm centers, but that would have been OK too.

Or, I could have taken a very different approach--an option that leads to my question about scaffolding. I could have started out by giving Joe the key information that studs are often on 16" centers--perhaps drawing him a picture of what typical wall framing looks like. Before he started work, we could have talked through the sequence of steps. "So, you will begin by thumping gently on the wall with your fist or with a hammer, being careful not to mar the surface, listening for whether is sounds hollow. When you have a place that sounds solid, drill a test hole. If the hole does not hit a stud, then what do you do?"

"I drill another one."

"Where?"

"Ummm, on either side of the first one? Maybe 3/4 of an inch away?"

"Drilling on either side of the first hole is a good idea. Do you know how wide a wall stud is?"

"No."

"Wall studs are an inch and a half wide. Do you want to revise your estimate of how far the next test hole should be from the first one, assuming the first one is a miss?"

"Ummm, an inch and a half?

"Sounds good. Maybe just a little less. The idea is to make sure that you don't miss a stud by drilling on either side of it. Okay, now suppose that you find a stud. Where do you drill the next hole?"

"You said they were usually 16 inches from center to center."

"Right."

"So 16 inches."

"Good. Now, keep in mind that we are hanging up this bike rack, and we actually want more than 16 inches between the uprights. So, you might skip one test hole and measure out 32 inches and see if you find a stud."

Which Way Forward?

Which approach would have been better? We work with some teachers who take a "pure" view of inquiry and voice concerns about providing too much information through direct presentations--teacher to student--telling students what to do rather than letting students discover facts and relationships. These teachers would be uncomfortable with a conversation like the imaginary one that I just described, in which I walked Joe through the work plan. They would, instead, be inclined to let Joe figure the pattern out for himself. Their argument would be that by drawing a picture of wall framing and telling him about the typical arrangement of studs, I would be robbing Joe of the opportunity to see a pattern in his measurements and learn to make inferences from such a pattern.

Other teachers prefer the second approach, giving Joe an overview of the way the system works before sending him "into the field."

I suppose that making this choice might depend on the learning objectives that you are aiming at. If you are most interested in having students learn to look for patterns, then having Joe drill hundreds of holes might be the way to go. On the other hand, if you were wanting Joe to come away with an understanding of wall framing systems, it seems to me that the second approach would be the better choice. The reason for this is that even if a student drilled all the holes and saw the pattern (Hooray!), it would not necessarily follow that the student saw how that pattern is part of a more complicated system.

I should also add that if your goal is to hang up a bike rack, the second approach is way better.

The story of Joe and the hundreds of holes is kind of simple and goofy (real life is sometimes like that). But it also has a good bit of relevance to what we are seeing in our work with teachers and students. Here are a few thoughts and questions that emerge from the story for me:

• Key Paradigms: Even though I am beginning to have an allergic reaction to the term "scientific method," I will argue that, whatever a scientific method is, it is NOT (at least not very often) about drilling hundreds of holes and seeing whether there is a pattern. As Thomas Kuhn argued--nearly 50 years ago now--in The Structure of Scientific Revolutions, normal scientific work builds on a paradigm that is already accepted by a community of scientists. The paradigm embodies theories about the makeup and working of the system under investigation. To the extent that science is work that builds on these paradigms, wouldn't it follow that a large part of science education should be about introducing important, broadly accepted paradigms? In my interaction with Joe, that would have meant drawing a picture of the way that walls are framed.
  
  
• Actual Scientific Practice: I worry about giving students the wrong idea about science. Science is not doing stuff to see what happens, or just looking at stuff to see what you find. This is another aspect of science's dependence on paradigms. A scientific paradigm not only encompasses theory, but also prescribes the accepted questions and methods for exploring and extending that theory. As Kuhn puts it on page 109 of my edition of The Structure of Scientific Revolutions, "paradigms provide scientists not only with a map but also with some of the directions essential for map-making. In learning a paradigm the scientist acquires theory, methods, and standards together, usually in an inextricable mixture." I am concerned that if we allow students to follow a path similar to Joe's--drilling hundreds of holes without a foundation in theory, methods, and standards--we are giving them a misconception of what actual scientific practice is all about.
  
  
• Proliferation of misconception and bad information: As I listened to students present their work to audiences of other students this year I winced as I heard some of them say things that were misinformed and, in some cases, just plain wrong. For example some students concluded that fish diet was not a factor in determining mercury burden. Another gave a small lecture to a group of students saying that mercury was due to local waste streams, setting aside the effects of air pollution. Much of this misinformation was, I suspect, related to students' overgeneralizing from their data or, in some cases, setting aside data completely and venturing into the realm of opinion. This is perhaps yet another reason to invest more time in providing students with a firm grounding in what we already know before sending them off to investigate.
  
  
• Student frustration: When I went back to see what Joe was doing, it was obvious that he was not happy with what was going on. He was confused and didn't know what he was doing or why he was doing it. Learning rarely happens in the context of such confusion and frustration.
  
  Listening to some of the students present their work I had the sense that they were in the same place that Joe was. Inquiry learning aims at that "Aha!" moment when students see, on their own, that the data come together into a pattern, but it can sometimes happen that someone can drill hundreds of holes--or collect and analyze samples of many macroinvertebrates and fish--and not see a pattern. Referring back once again to Thomas Kuhn, much of the value of a paradigm is that it actually provides the pattern. Most work in science starts from a well-understood pattern and seeks to fill out details or extend the pattern to new situations. Scientists don't like working without a clear understanding of context and purpose anymore than students do.
  
  
• Citizen science is like hanging a bike rack: One of the strongest "take-aways" from this year's work with teachers and students is that much of the value of using citizen science as part of science instruction arises from the service dimension of the work: the data are actually useful. This value arising from usefulness is not just value perceived by scientists, but is also--and most importantly from the standpoint of our educational goals--value perceived by students. The students are engaged in meaningful work.
  
  Putting up a bike rack was also meaningful work, but for Joe the meaning got lost in the frustration of drilling hundreds of holes. Joe knew that, whatever it was that I had in mind when I set him to the task of finding the wall studs, what he was doing probably wasn't it. So he was frustrated, and the frustration overtook the meaning.
  
  For some students this year, their experience was at least a little like Joe's. Meaningful work feels purposeful, rather than random and confusing.

Supporting scientific learning through inquiry and work in citizen science is complicated. Teachers using this approach are faced with finding a balance between direct instruction--transmitting information and instructions required to do the work and to support learning--and guiding a process where students are engaged in figuring things out for themselves.

The image of Joe kneeling there while he perforated wallboard is a reminder of what happens when we get this balance wrong. The system that Joe was "researching" was a simple one compared to, say, as pond or stream. It is easy to see that the balance I created between direct instruction and Joe-directed inquiry was SO wrong, resulting in no work getting done and little learning. I offer the story as a parable.