Speaking Tests: My Favorite and Not So Favorite Options

by Jackie Bolen

There are a few different options for speaking tests for your conversation classes, all of which have their advantages and disadvantages in terms of teacher effort, ease of grading and how well it measures what you’re trying to measure. I'll briefly outline the four major options:

A Speech

Some teachers put a big emphasis upon this, but a speech is not really a true test of conversational ability so if you teach a conversation class, it’s not ideal. Communication is more of a 2-way thing and is far better judged by the following speaking test styles I’ll mention below.

That said, I have taught public speaking and presentation classes before and of course their final exam was a speech! Here's a blog entry about my experiences teaching that class for the first time and what the students reported at the end of it: Interesting Results from my Presentation Class.

A Dialogue

This is something that students prepare at home and perform in front of the teacher, or class. Many of my fellow teachers do this, but I’m not sure this is a great option either. Generally, the best student in the group will write it, and then the rest of the team will memorize it. And we all know that Korean students are generally excellent memorizers so it's often relatively easy for them to do this. However, memorization also isn’t really a true test of English communication or conversational ability and this dialogue style of test often ends up with the dreaded "robot" conversation, void of any emotion or spontaneity.

Question and Answers, with Another Student

This happens between 2 students, chosen at random on the day of the test. These days, it's by far my favorite way to conduct speaking tests, since the teacher can just listen and not act an examiner at the same time which makes it far less exhausting. It's also an excellent way to do testing, student-centered style.

I give my advanced level students a set of around 10 general topics and then I bring students into my office in groups of 4. I choose two of them to go together at random and hand each of them a slip of paper with a general topic, also chosen at random and the pair has to have a conversation about those two things for around 7-8 minutes. I listen and only interfere when it's time to begin, switch topic and finish.

With lower-level students, I'll reduce the number of topics, spend time in class teaching them how to start the conversation and also require only 3-5 minute conversations.

Lately I've been using this ESL Speaking Rubric to evaluate my students. 

Questions and Answers, with Me

This is where the student has a short conversation with the teacher, individually. This is a very accurate way to test students because it can be controlled quite easily but the huge negative is the exhaustion factor, especially with very low-level students. There is basically nothing more tedious than having 1-1 conversations with someone who is just waiting for the teacher to ask them question after question when they give only 1-2 word answers in return. It can also be quite difficult to act as interlocutor as well as examiner, especially at the end of a long day of testing.

Here's a blog post where I talk about my experience using this type of test: 4 Corners 2 Speaking Test.

About Jackie Bolen

Jackie has been in Korea for almost a decade and has taught every age group and level of student. She's currently working at a major university in Busan. She blogs at My Life! Teaching in a Korean University.