How’s that for a fine contradiction! Like many teachers, I complain about students not caring enough, not understanding the value of basic things like preparing for class, not necessarily even understanding the importance of showing up for class. So why would I want to lower the stakes, make stuff seem less important? Precisely because high stakes are stressful, and lots of stress isn’t conducive to great teaching and learning.
This is the message I got from reading Tomorrow’s Professor post #1271, “Designing College More Like a Video Game – Motivating Change with High Standards and Low Stakes” by José Antonio Bowen. (The excerpt is from chapter four of Bowen’s 2012 book, Teaching Naked.) Citing evidence from Arum and Roksa (2011) and Bain (2004), Bowen writes:
The best teachers focus on challenging students in a supportive environment where failure is tolerated. The combination is essential; just having high standards is not enough to help students learn. Bain discovers repeatedly that the best teachers expect more of their students yet treat them with genuine caring and give them a sense of control. Students learn best when they believe that the professor wants them to succeed.
There’s more than one key detail here. First, Bowen argues that failure needs to be made part of a classroom’s set of expectations: when we fail we learn (as I’ve argued, anecdotally, before on this blog), but most students (me included!) reach high school or university age conditioned to fear failure rather than to recognize it as an essential component of their learning practice. So working with failure is one key tactic for raising the bar by lowering the stakes. Second, Bowen argues, via Bain, for demanding, challenging classrooms (“high standards”), in which teachers expect a lot from their students – including that trying to meet challenges means inevitable (and essential) failure somewhere along the way. Bowen then makes a bit of a cognitive leap, seemingly equating student failure with teacherly support to produce his last claim, that “students learn best when they believe that the professor wants them to succeed”. I think what he’s actually trying to do here is communicate an equation for the ideal high-bar, low-stakes classroom: challenging students + expecting and supporting students through failure = a better rate of student success, both in real-time learning and in learning how to learn for all time.
I really like Bowen’s argument (and it’s much more extensive than the part I’ve just summarized; check it out in full here). In theory it makes perfect sense: demand a lot, encourage experimentation and risk-taking, and remind students that risk will be rewarded, as long as it is genuinely interrogatory, curiosity- and research-driven, regardless of outcome. This means changing what students understand “success” and “failure” to mean – effectively, shaking these terms loose from their binary relationship. But how to do this in real life? Bowen argues that “Practically, our ability to lower the risk of failure while maintaining high standards means we have to rethink what and how we assess”: that might mean creating more assessments worth fewer marks each, creating do-over opportunities, or building a learning-from-failure component into the marking of an experiment, an essay, or a group performance.
I have favoured the do-over option in the past; for a while now, I’ve been working on different kinds of do-over essay assignments, and my current favourite goes like this: students are given the option of handing work in on a preliminary due date, receiving feedback and a grade from me, and then getting the chance to improve their work and hand in again on another, final deadline. Sometimes peer feedback is built into this loop, but I’m still working on best practices for peer feedback. (I’d love to hear how others work with peer-to-peer support on writing assignments – I find it genuinely tricky because so few students understand, even with some practice, what makes writing good, bad, or great).
The key to my do-over assignment is two-fold: first, students get concrete suggestions from me on how to improve their papers, and if they do those things and hand in again their marks will go up. (If they do what I suggest but the results aren’t that much better, the mark doesn’t rise that much, but it still rises. How to deal better with this relatively common scenario is something else on which I’m still working.) Second, students are guaranteed that if they hand in again their marks won’t go down. Thus, if they take the second chance, the stakes are pretty low, but the potential for success, defined as a good mark, is quite high.
Last semester I used the above do-over on my students in my Naturalism seminar, and the outcomes were overall very good – as was student feedback on the assignment. Bowen confirms my hunch as to why: “Assessments that promote learning combine low-stakes and high-quality feedback. Both foster change and are highly motivating; it is easier to try something new if the stakes are low and easier to change when you are being encouraged and when you know exactly what change is needed.” I’ve hit an ethical snag, though – one I hadn’t even considered until it was pointed out to me by a colleague whose teaching I greatly respect. This method of assessment may work well for me and my classroom, but it can also create a perception of unfairness when my students are part of a tightly-knit cohort working in a small, close department. Queen Mary Drama has a rule in place (principally to promote fair, equitable practice across the board, while also protecting instructors’ time) that prevents us from reading and commenting on students’ work in advance beyond a very small proportion (about 10% of a paper); this gives us a very clear line to offer students who try to impinge on our time by getting us to, essentially, do their editing and leg-work for them in office hours. Thus, even though my do-over assignment has a basis in sound pedagogical practice, and even though it was built clearly (and equitably, for all students in the class) into my Naturalism syllabus from day one, it skirted a bit close to our departmental red line, as it risked placing unwarranted and unfair expectations on my colleagues.
I understand the issues at stake here, but I remain on the fence about their consequences. Naturally I want to be as fair as possible to my colleagues, but my first responsibility, as a teacher, is always to my students. Every year I see more and more students struggle to write well, and I know that, even though I’m trained to teach theatre, not writing skills, it’s my job to educate the whole student, which includes the student writer – and the student writer needs more help than the vast majority of Arts and Humanities programs currently offer. How to solve this catch-22? I suspect one answer is a curriculum review (luckily, I think Queen Mary has one on the horizon, and next week is our first Teaching and Student Support Committee meeting of the year…). That just might make (sanctioned) space for a few do-overs.