Assessing Speaking

Output.

Let’s talk about it.  First, I’d like to state that based on the research I’ve done through my grad program at MSU, I’ve solidified my beliefs that speaking does not aid acquisition. It is a product of acquisition.

Students don’t need to speak to learn more. BUT being able to speak is the goal for most students. They want to be able to communicate. So I of course encourage students to speak, but I do not often assess speaking formally. I also think that showing off can be a great motivator for students, when they try to produce language and get positive feedback (the interlocutor confirmed that they understood the message) then they are more willing to try.

On a recent #langchat twitter chat we were talking about Task Based Language Teaching. There are two types of tasks, essentially. Input oriented tasks and output oriented tasks. I often times do input oriented tasks, but I make sure there are opportunities for students to speak when they are ready.

Swain’s (1985) Comprehensible output hypothesis is often misinterpreted, I think. In reality the goal of CO is twofold: to notice the gap (students see what they don’t know) and providing a prompt for an interlocutor to give them more input.

I’m torn on the noticing hypothesis… I think the idea of noticing is real. Students probably DO notice what they don’t know when they are trying to produce language. I’m not sure of the long term effects though. Do learners put puzzle pieces together? They notice the gap, and look for the matching piece in input to modify their own language? And even if they do, are they getting enough examples of that missing piece to form mental representation?

This is not where I thought this blog post was going… So I’m going to move on to assessing speaking now.

This year, I tried assessing speaking online. I recorded prompts (similar to how the AP test is set up), gave students a time limit to complete the prompts to ensure spontaneity, and also shuffled questions (I recorded 10 questions, each student only got 5). It… worked. However, it did not work well. So I switched my approach.

Whenever I do PQA, I arrange the chairs in a circle. Some students absolutely LOVED “circle time” others were indifferent. Oftentimes I would give students the PQA questions ahead of time, they could try to formulate some sort of answer and they liked being able to speak Spanish to participate in class. However, that is not really authentic, is it? We don’t often write down what we are going to say when we have a discussion. If I wanted to assess what they wrote I would have a writing assessment!

Once in a circle I would have my iPad open to our LMS and mark students on how well they participated interpersonally. That is, how well could they follow up with what they initially say. Their pre-written answers were really topics for ME to make comprehensible and facilitate discussion with other students in the class.

You could of course have your rubrics printed out and jot down scores as you go, maybe make a copy for you and a copy for students so they can receive feedback.

Throughout the last semester I (tried to take) took 3 speaking scores. Which gave me six weeks to to hear everyone speak at least once, and I would always put in their most recent score on the LMS, but put their highest score in the gradebook at the end of the grading period.

I did start off using Martina Bex’s rubrics, but since Indiana changed their standards to I-Can statements I am using those to be better aligned with the state.