geology and science fairs
I helped judge a science fair on Thursday. There were only ten entries in the Earth and Space Science category, from middle school and high school combined, so I ended up helping judge the 6th grade Environmental Science and Consumer Science, Mathematics, and Computer Science categories.
On the good side, many of the consumer experiments were repeated and included controls (washing an unmarked piece of cloth, for instance), and some of the students talked to me about possible sources of error, and about how much experimental error there was (is there really a significant difference between two brands of diapers)? As a result, I'm not quite as dismayed about the state of US science education as Tafel is in his/her post about science fairs and ostriches (linked from a post on _scientists_).
On the bad side, the environmental science projects weren't as good, overall. Yes, a couple of the projects were clearly more creative - they were based on things that the students observed in the natural world at home. But even the best projects didn't show any understanding of possible sources of error, or - possibly worse - much awareness of the ways in which the students were simulating things that happen in the natural world. Some of the worst projects made conclusions that were completely unrelated to the experiments that the students had done - the conclusions were based on web research, and in some cases were completely at odds with the students' own observations.
It made me wonder: what would make a good earth science/environmental science/geology science fair project? Observations of the natural world are tricky: the earth is big, and things happen slowly, and it is difficult to make significant repeated observations of anything in one small place over a short period of time. (Example: imagine a student trying to test whether global warming is real by measuring the temperature at his/her house. Even if the measurements were taken every day for a month, the students wouldn't be able to really test the hypothesis.) There was an interesting philosophy-of-geology article in Geology a few years back that argued that a historical science like geology necessarily uses different sorts of tests than a lab science would. And the best geology science fair projects that I've heard of do science the way geologists do - mapping features on Venus, or doing field work with undergraduate researchers. But those were special cases, and the students had unusual opportunities to work with college faculty. (In one case, the student was the daughter of a structural geologist, and had probably absorbed a lot of geologic background already.)
yami_mcmoots's blog post on the way the general public observes earthquakes gave me one idea. She has noticed that people tend to categorize earthquakes by what they feel - a "jolt" or a "shake" or a rolling or twisting motion. But people don't seem to realize that their observations are just one data point - that the experience of an earthquake varies from one place to another. That made me wonder - could a kid do a science fair experiment to test which is more important, the earthquake itself or the location of the observer? I wonder if a kid could use the internet to collect descriptions of earthquake experiences, and plot the locations of "jolts" vs "shakes" vs "rolls" vs "twists" on a map? (In this case, "internet" means some kind of social networking site -- myspace, or some other way of communicating with lots of kids in, say, the greater SF Bay area or the greater LA area.) The internet might be able to help a kid deal with variations in space. As for variations in time - or for earthquakes in different places - I don't really know how a kid could address those things. (And spatial statistics, to deal with the error in the data... heck, most professional geologists don't handle that very well.)
And I don't know how to get the general public to see their experiences as limited by space and time, either. That's one of the problems that geoscientists need to face when communicating what they know to the rest of the world, I think. But I don't have any good answers.
On the good side, many of the consumer experiments were repeated and included controls (washing an unmarked piece of cloth, for instance), and some of the students talked to me about possible sources of error, and about how much experimental error there was (is there really a significant difference between two brands of diapers)? As a result, I'm not quite as dismayed about the state of US science education as Tafel is in his/her post about science fairs and ostriches (linked from a post on _scientists_).
On the bad side, the environmental science projects weren't as good, overall. Yes, a couple of the projects were clearly more creative - they were based on things that the students observed in the natural world at home. But even the best projects didn't show any understanding of possible sources of error, or - possibly worse - much awareness of the ways in which the students were simulating things that happen in the natural world. Some of the worst projects made conclusions that were completely unrelated to the experiments that the students had done - the conclusions were based on web research, and in some cases were completely at odds with the students' own observations.
It made me wonder: what would make a good earth science/environmental science/geology science fair project? Observations of the natural world are tricky: the earth is big, and things happen slowly, and it is difficult to make significant repeated observations of anything in one small place over a short period of time. (Example: imagine a student trying to test whether global warming is real by measuring the temperature at his/her house. Even if the measurements were taken every day for a month, the students wouldn't be able to really test the hypothesis.) There was an interesting philosophy-of-geology article in Geology a few years back that argued that a historical science like geology necessarily uses different sorts of tests than a lab science would. And the best geology science fair projects that I've heard of do science the way geologists do - mapping features on Venus, or doing field work with undergraduate researchers. But those were special cases, and the students had unusual opportunities to work with college faculty. (In one case, the student was the daughter of a structural geologist, and had probably absorbed a lot of geologic background already.)
yami_mcmoots's blog post on the way the general public observes earthquakes gave me one idea. She has noticed that people tend to categorize earthquakes by what they feel - a "jolt" or a "shake" or a rolling or twisting motion. But people don't seem to realize that their observations are just one data point - that the experience of an earthquake varies from one place to another. That made me wonder - could a kid do a science fair experiment to test which is more important, the earthquake itself or the location of the observer? I wonder if a kid could use the internet to collect descriptions of earthquake experiences, and plot the locations of "jolts" vs "shakes" vs "rolls" vs "twists" on a map? (In this case, "internet" means some kind of social networking site -- myspace, or some other way of communicating with lots of kids in, say, the greater SF Bay area or the greater LA area.) The internet might be able to help a kid deal with variations in space. As for variations in time - or for earthquakes in different places - I don't really know how a kid could address those things. (And spatial statistics, to deal with the error in the data... heck, most professional geologists don't handle that very well.)
And I don't know how to get the general public to see their experiences as limited by space and time, either. That's one of the problems that geoscientists need to face when communicating what they know to the rest of the world, I think. But I don't have any good answers.