Here are more thoughts on assessment in addition to yesterday’s post. They should be timely seeing as teachers are talking about end of quarter exams, and upcoming mid-terms.

So, individually assessing student speaking one-at-a-time in the hall? This is a big waste of time, and here’s why…

Any role that speaking might play in language acquisition aside, while this individual assessment is going on, the other students in the classroom are doing some kind of assignment (busy work exercise?) that might or might not contain any input. This is bad. Since the entire event probably lasts an extra day just to assess everyone, you might as well subtract those 2 classes from total contact hours for the year, but it doesn’t stop there. Since it’s pretty irresponsible to hit students with a single assessment grade destroyer, there should be at least a few of these per grading term. Why? Because multiple sets of data are needed to account for “those days” that kids inevitably have when averaging occurs. So, not only subtract 8 days of these assessments for a typical year on grading quarters, but multiply that by the number of assessments you need to fairly assign a grade, which brings it to at least 24. Now, consider what an extra month of input would do for your students each year…

And now, also consider how much follow up time it would take, even if just entering grades, then determining who needs to make it up due to which absence, etc., not to mention any kind of analysis or feedback you decide to give. Let’s say you have 100 students. If—if—you can do all of that in about 1min per student, you’ll spend a little over 1 1/2 hours on this house keeping alone, aside from any other legitimate planning during your prep time. Remember, you just did this to yourself!

In fact, any quiz you give that isn’t immediately self-graded by students will be added to your responsibilities. N.B. in-class self-grading could (should?) be entirely in the target language (i.e. an opportunity for more input!).

The alternative to hallway assessing and multiple assessments in many grading categories? Spend as little time as possible assessing and grading because the instructional adjustment is really no adjustment at all; it’s what you’re doing anyway—providing more input (unless, of course, you work with self-selecting intermediate to advanced speakers with more specific goals/needs, though at least one of those rules out public school teaching).

Self-Grading & Batch “Assessments”
I prefer to think about each real-time interaction (even non-verbal cues) as authentic assessments. That teacher in the hall might actually be authentically assessing as well, but the speaking rubric likely being used removes some authenticity, as well as any likely adjustment. For example, instead of making Spanish more comprehensible, the teacher likely circles the rubric box that reflects how the student can’t do X or sustain Y. This is unhelpful for the student, even if they receive that feedback the next day. In fact, I would argue that any feedback given asynchronously, whether numbers, letters, or words, is completely useless. Also, I won’t even mention how students must have known about the speaking assessment ahead of time, which means any speaking data could just be a practiced performance. Oh, and if students didn’t know, at least every kid after the first could be spending class time practicing what to say when called into the hall. No bueno.

The things we need for the purpose of school accountability usually fall into “quiz/test,” and not really “assessment,” because authentically assessing occurs with every interaction. A solution? Do the following to avoid losing precious time, but still get something for the gradebook:

• Instead of pulling those students 1-by-1, do something that reaches all students.
• Keep quizzes short.
• Self-grade short quizzes as a class immediately following the quiz.
• Get back to input.
• Use planning time for stuff that actually impacts acquisition.

This thinking can also be extended to those seemingly benign practices that have profound effects on one’s teaching. From a post in August, look at what can happen just from 1) assigning homework, then 2) deciding to grade it…

The teacher needs to determine a) what is even graded (e.g. completion? a rubric? other criteria?), b) late, or makeup policies (e.g. partial credit? diminishing scale per day late? accounting for excused absences?), c) grade weighting (e.g. 5%, 10%, 20% of the course grade), and d) then make the time to evaluate the homework based on a. Any experienced teacher knows that b is probably the most frequent, yet most headache-causing aspect of grading anything, especially homework. Kids will be out of class for a host of reasons, and there’s nothing that will ever change that. Why not establish policies that anticipate and avoid, rather than complicate and frustrate?! So, with this one deceptively simple decision to grade homework, the teacher has created a considerable amount of more work for him/herself, and has complicated the daily goings-on for the student. Nevertheless, many schools still require homework, and there need to be solutions for sanity.