Bombed is an overstatement, but, almost across the board, students' test scores brought down their overall grade. And, surprisingly, students tended to do worse on the basic recall questions than they did at the short response questions that drew on higher-order analytical thinking. At first glance, this seems to reinforce the narrative of traditional testing being a poor indication of student understanding, but I hesitate to throw the written exam by the wayside, if only for the mystery of why my students are better at the "harder" thinking than the "easy" test questions.
Again, in the spirit of UbD, I turned back to myself and my lesson planning in order to "[Accept] Responsibility for Learner Success" (and lack thereof) (Tomlinson & McTighe 2006). I noticed pretty quickly that the type of thinking I was testing my students on wasn't emphasized very much in my day-to-day teaching. Even if the thinking on the test should be "easier," building student skills in higher-order tasks doesn't automatically build recall skills (much less comfort and familiarity with test taking). And despite both my and my students' preference for lessons that exercise these higher-order, "more authentic" mind muscles, it's definitely a problem if my students can't reliably memorize a small handful of definitions of literary terms (that short multiple choice section was an absolute massacre--I think only one student out of 31 got all 4 definitions correct).
So, if I want to keep the traditional exam a part of my curriculum, I need to give students the opportunity to build the skills tested by these exams and build their comfort with test taking in general. This means giving an instructional emphasis to recall items (rather than simply telling students, "get this definition in your notebook," and trusting them to understand that they're accountable for it later), as well as explicitly teaching study and test taking strategies (which many of my students may not have explicitly learned in middle school). I've also begun incorporating some of Jim Burke's assessment strategies from The English Teacher's Companion (2013). These changes tend to be small--he advises "integrating assessment into the course" by making day to day classroom activities look and feel a bit more like an assessment, as well as incorporating choice into assessment, because answering 5 of 7 questions allows students to emphasize their strengths, as well as being much less stressful than answering 5 of 5. Small tweaks like this can do wonders to defuse test anxiety and make sure students are building both higher and lower-order thinking skills
Works Cited
Burke, Jim. (2013). The English Teacher's Companion. Portsmouth, NH: Heinemann.
Tomlinson, Carol Ann, and McTighe, Jay. (2006). Integrating Differentiated Instruction and Understanding by Design. Alexandria, VA: ASCD.