Assessment

Exams are coming up in just a little over a week.  Before the break, I gave a quiz in CS I over what we’d covered so far, and the range of grades was what you might expect.  A few failed, a few did super well, and a lot were in the middle.  Many of my students are super worried about the exam, based on how they did on the quiz.  What they may or may not know is that whatever they missed on the quiz, they probably understand now.  I went over the quiz in detail, round robin style, having each student  try to answer the question.  We also talked about strategies for eliminating questions and I gave them hints like loops usually reset variables, so if you see a loop without that, it’s probably wrong.

On the one hand, I’m not a fan of tests.  On the other, it is a good way to solidify your recall of certain concepts.  I’m not going to lie.  My test is hard.  It’s on par with the AP Test just in a different language.  I’d say I cover 1/2 to 2/3 of the concepts covered on the AP test and my questions are quite similar.  So a little worry is in order.  However, the bulk of my students’ grades come from the work they do in class: labs and projects.  What I find is that students who are doing well on those, which is most of them, aren’t affected by the exam, even if they do poorly.  Students who struggle on those tend to struggle on the exam and therefore do poorly on the exam.  Which makes sense.  You have to understand the concepts to complete the projects.

I do, however, want to change some of my assessment strategies.  I’ve been reading Specifications Grading, and while it’s geared toward Higher Education, there are some ideas in there that are worth considering and modifying.  My department works on a project-based level, but we do want to make sure we can clearly articulate the skills our students are acquiring, some of which are soft skills like figuring out problems independently and coming up with creative ideas.

I’m going to try a version of this in my Mobile Computing class that starts in a few weeks.  I’ll report here on the process and progress.  I’d love to hear other ideas for assessing longer-term projects and skill mastery.

3 Replies to “Assessment”

  1. I also much prefer projects to tests. When designing our intro course, I wanted it to be as project based as possible but it turned out that we HAD to put in a good number of tests because our kids are so conditioned to taking tests that we can’t just drop them cold turkey. Our kids come in as seasoned “test takers.” They expect tests and build there study patterns around classes based on tests — no tests and they don’t effectively study.

    We actually have to ween them off of this mentality so that we can go project based by the end of their second year of CS (APCS, junior year).

  2. Great question and insights, Lauren. I am currently exploring ways to get into K-12 computer/technology instruction, but I have some ideas based on the professional training I’ve conducted. Bottom line is to be being as transparent as possible about expectations and parameters of the assignment. I just wrote a more elaborate explanation on my nascent blog, http://www.teachingkidstocode.com.

  3. Great question and insights, Lauren. I am currently exploring ways to get into K-12 computer/technology instruction, but I have some ideas based on the professional training I’ve conducted. Bottom line is to be being as transparent as possible about expectations and parameters of the assignment. I just wrote a more elaborate explanation on my nascent blog, http://teachingkidstocode.com/developing-assessmentsgrading-for-project-based-assignments/.

Comments are closed.