Assessing Speaking

In practice, we often use the words test and assessment interchangeably. But they do have slightly different meanings.

Students need to know what is expected from them, how they can best prepare for any assessment, what counts and what does not count, and how they’re going to be assessed. And this also gives them an opportunity to learn how they are making progress.

Text by: Anastasia Spyropoulou

An assessment is like a mirror; it allows you to keep checking what you look like, and you can make changes based on what you see.

A test, on the other hand, is like a photograph - it captures one moment.

Continuous assessment means that assessment is ongoing throughout the duration of a course. It is often informal and formative; it may also contribute to students’ grades.

Assessing Speaking

What is speaking?

The CEFR helps us to describe language levels. When you think about your own language ability, you can think of a huge range of examples of language use from your whole life. When thinking about your students’ levels, however, you only know what you’ve seen or heard them do in class.

So to know what level our students are at, we need to listen to them speaking. In order to think about how to sample students’ spoken language, we need to be clear about what we mean by speaking.

Most of us speak every day. It’s part of who we are. We speak to our friends, our colleagues at work, even to ourselves. When we speak, our brain is engaged in a whole range of activities, from finding words to putting them together in a meaningful way. Because we speak to different audiences for different reasons and under different conditions, we can look at speaking as being multidimensional. Since an assessment should try to replicate as closely as possible the real-life use of the language in terms of the brain activity and the social conditions in which the language is used, a test of speaking needs to be as close to the context we’re interested in as possible.

Obviously, we can’t observe students’ language use all the time, so in any assessment we need to take a sample of students’ language use. We would like these samples to be representative and reflect their real lives. For example, if we want to know about students’ ability to speak in everyday situations, we should include such situations in the test (e.g., giving directions, ordering food in a restaurant, etc.). This gives us a sample of students’ speaking in the context we are interested in.

There are three important factors that affect the way we speak in different situations:

Audience - who we’re speaking to

Reason - why we’re speaking

Social conditions - the social context we’re speaking in.

For example:

Audience: My husband

Reason: To decide which of us is going to cook dinner in the evening.

Social conditions: Informal conversation while eating breakfast with family members.

Formats, tasks and settings

In an assessment, there are lots of ways in which we can try to reproduce the conditions of real life speaking.


Common formats for speaking tests include:

Interview: This has the advantage of being interactive, but is not like a real conversation because one person usually asks all the questions and the other answers them.

Oral presentation: Presentations can be useful in business or academic situations, but they are not very interactive.

Interactive task: When two or more students work together, it encourages natural student interaction and genuine communication. However, one student’s performance might affect the other’s.

Group discussion: This can work well because it saves time. However, teachers need to be careful not to allow any students to dominate while others say very little.


Speaking tasks commonly used in assessment include:

Describing something: For example, describing a picture. This gives students something to talk about, and it doesn’t require students to read anything.

Comparing things: This is similar to describing something, but can be more demanding. It encourages students to compare and contrast two different things.

Telling a story: This is a very natural speaking activity. Stories can be based on pictures. Alternatively, students can tell their own stories, but some students might find it difficult to think of a story to tell.

Giving some personal information: This is also a natural speaking activity, and focuses on personal topics that students are familiar with. As well as basic information, it can include talking about personal experiences and opinions.


Live: This means the assessment is completed in real-time.

Recorded: This allows students’ speaking performances to be recorded; it also allows markers to listen more than once. This setting, however, means that markers need time to listen to the recordings after the test.

Remote (via phone or internet): This may feel different from face-to-face communication, but it can make tests easier to access, especially for students in remote locations.


After getting a realistic sample of students’ speaking in an assessment, we need to think about how to rate it or how we give it a score. When the examiner is asked to award a score for a speaking test performance, they usually use a set of descriptions of what to expect at different levels of ability. This set of descriptions is known as a rating scale. Some rating scales are simple in design and simple to use. Others are more complex.

As with most things, the complex ones tend to be the best, but are more difficult to create and use. Improvements in voice recognition technology, when combined with advances in computer power and speed, as well as our understanding of artificial intelligence, mean that we are getting closer to having an interactive conversation with a machine which would then rate our performance.

Practical considerations

One of the biggest practical problems for teachers is time. We need to listen to each student, but often there is limited time available. Peer assessment helps students learn how assessment works. They learn what they’re supposed to be doing, when they’re involved in a classroom assessment. And hopefully, they learn what they should be doing when they’re involved in a communication as well.

Feedback is really important after a test, because it tells people what they did well. It’s often tempting to concentrate on mistakes or errors, but people need to know what was good. So that should be included; they also need to know how they can get better. What’s the next level on your scale? What aspects of the students’ speaking would need to be improved to take them to that next level?

Rating scales

Rating scales can have positive washback in the classroom. Students’ speaking ability gradually develops over time. A rating scale helps us to describe a student’s performance in an assessment. It also describes the next level up. Looking at the differences between a student’s current level and the next level on the scale can help teachers to give focused feedback and set realistic goals.




ELT News

ELT News

Leave a Reply

Your email address will not be published. Required fields are marked *