How Do We Remove Scaffolding?
I'm not sure we're doing the Kids These Days any favors
The title doesn’t refer to literal, physical scaffolding— there’s a sheet of ice on everything right at the moment, so it’s not like anybody is doing any external work on Chateau Steelypips these days. It’s a reference to a very common metaphor in educational contexts, and a stock piece of advice for how to improve student performance, particularly in intro courses.
Whenever there’s any discussion of students struggling in an intro course, the suggested solution is almost always to “add more scaffolding.” Meaning to break topics and assignments down into more sub-parts that provide a clear path onward and upward. Rather than specifying a launch velocity and angle for a projectile and just asking “determine where this hits the ground,” for example, you would break this up into parts asking first for the horizontal and vertical components of the launch velocity, then for the time the projectile spends in the air, and then for the position when it reaches the ground. This is the same process an expert would use to solve the problem in the un-scaffolded formulation, but it provide cues that guide the student forward. This works very well in that it dramatically reduces the number of papers that come in essentially blank, or go off in some utterly bizarre direction that can’t possibly lead to the correct solution.
This also extends to things like the structure of courses. Rather than simply stating a general goal, a course with more “scaffolding” will provide a detailed road map of what will be covered when, with numerous small assignments having very set deadlines. Again, the idea is to lead students through the course as a whole in a way that reduces confusion, thus freeing up mental energy from trying to parse out the syllabus and allowing it to go toward doing the actual assignments.
This is, by and large, a good practice when it comes to reducing student frustration and confusion, and tends to lead to somewhat higher grades (though this is a somewhat circular claim since the better grades are often coming from increased partial credit thanks to the highly “scaffolded” assignments…). I worry a bit, though, about the longer-term effects of this. In particular, to push the metaphor a bit, I am somewhat concerned that the helpful scaffolding we’re providing early on is turning into those sidewalk sheds that have become a signature part of the New York City pedestrian experience: once they go up, they just become part of the landscape forever.
This is prompted by the fact that I’m doing some prep for one of the courses I’m going to be teaching in the Spring term, our upper-level laboratory for junior and senior Physics majors. We’re going to be making major changes to the structure of the course this time out, in ways that will probably make the course a more pleasant experience, but that also make me wonder about the knock-on effects of all the scaffolding we’re doing earlier on in the curriculum.
This course has always been something of a problem because it’s kind of overloaded, but one of the nice things about it from a faculty standpoint has always been that it’s a little loose. These are supposed to be more advanced labs, and we have traditionally tried to emphasize independent work— we give students a brief introduction to the experiment and the apparatus to be used, card access to the lab, and a deadline two-ish weeks out. The actual data taking and analysis, they’re supposed to do on their own.
This used to work reasonably well, but the last couple of times out it’s been a dismal failure. Students in the course put absolutely everything off until the last possible moment and then discover that it’s physically impossible to do a 36-hour data collection run in one night, or they sit down to write the lab report the night before it’s due and find that while they have most of the measurements they need, they’re missing one key value and can’t actually do any of the analysis. Nearly every lab ends up being late and most are inadequate in one respect or another.
The solution to this, you will be unsurprised to learn, is going to be scaffolding. From student comments and conversations outside of the class, it’s become clear that part of the problem was the lack of sub-deadlines, especially compared to other classes. When our course had a two-week deadline to do a large task and some other course had tight deadlines for smaller assignments, many students were prioritizing the definite deadlines and putting off working on our labs. Even when the weights of the assignments didn’t remotely match— one lab would be worth around 20% of our course grade, but they’d put off working on it to do a homework worth 4% of the grade in a different course, just because that had a specific deadline.
So, we’re adding structure, and largely going away from the independent-work format by assigning specific tasks to specific class meetings, and checkpoint steps along the way for each of the labs. I’ll also be providing much more detailed lab handouts than I have previously— in past years I’ve given students copies of a research paper and asked them to duplicate the procedure from that, or copies of a textbook chapter and asked them to figure out how to apply that to the analysis they need to do. That’s just not working any more— too many of them find it overwhelming to not have a clearly laid out path.
I am reasonably confident that adding more structure will make this go more smoothly from a student-experience standpoint. I’m a little less sure from the faculty side, but I think it will still be a net win— the increased structure will require more direct supervision than the independent-work model, but should reduce the number of frustratingly late or wildly inadequate lab reports that I have to grade.
At the same time, though, I wonder whether we’re actually doing these students any favors with all this. Adding structure will help with the problems we’ve been seeing with regard to time management and prioritization of tasks, but it’s also sort of kicking the can down the road. At some point students need to figure out how to do these things on their own, without being walked through it, and if they’re not getting that by the time they’re in a junior-level course in their major field, I don’t know when it’s going to happen. Grad school? Their first job?
It feels a bit like we’ve correctly put up scaffolding in the early stages of a masonry repair project to prevent pedestrians from being brained by a falling trowel, and that’s good. But now the facade work has been done for five years, and people are still picking their way around metal posts in permanent shadow, wondering exactly what’s dripping from the pipes when it hasn’t rained in a week.
Again, it’s good and appropriate that we’ve added scaffolding early on in the curriculum; on the whole, it’s improved the experience for everyone. At the same time, though, we need more discussion of how to get rid of the scaffolding later on, phasing out the extra support structure and encouraging students to operate more independently. As it is, though, all of the energy in faculty and curriculum development seems to be going into extending the scaffolding further and further.
I totally understand it because all the incentives push in that direction— I mean, I’m doing it myself, as described above. But the whole time, I have this nagging feeling that this isn’t really helping students to grow and develop in the ways they need most.
That’s your dose of “Old Man Yells At Clouds” for the week, after climbing back out of a bit of a class-prep rabbithole this morning. If you like this kind of thing, here’s a button:
And if you have any great ideas or resources on how to phase out “scaffolding” in the course of a major without making everyone miserable, please, leave a pointer in the comments:
I worked in engineering for more than 40 years, and almost everything we did was driven by a schedule. The customer needs project X completed by a given date, so the first thing we did was break that X down into its components (define input values, do calculations A, B, and C; write reports x, y, and z; etc. etc.). The schedule laid these all out, with links (A has to be done before I can do report y...). Simple projects have simple schedules and complicated projects have complicated schedules. All of this was completely foreign to me as a new hire. But that's how we got things done on time, and *that's* how we got the customer to give us future work.
Maybe the first sub-assignment is to have the students prepare their own schedule. At least a calendar showing the due dates for each of their courses, as far as they can be known. Then try to fill in the intermediate tasks. Few freshmen would be able to do this, but by third year I'd think they could at least take a stab at it.
Sometimes the metaphors you use trap your thinking, and refitting a new metaphor can offer new solutions. Scaffolding are semi-permanent structures used in construction. What your students need instead are training wheels, which are only meant to be used until riders can balance on their own. For example, instead of having staff create intermediate deadlines and assigning them to students, why not consider giving them a project management template they can reuse? It's hard to ignore your project turning danger red because you have to do a 36-hour task and only have 48 hours left...
Hope that's helpful.
Cjf