Should placement tests be given without students’ doing any preparation so that we can see their natural level in English?
Ideally, placement tests should not need any special preparation. The format and types of questions on the test should be straightforward. This ensures that all students can understand what they need to do.
How should the feedback from progress tests be given? Should we give it individually or work with the whole class?
It’s great if you have time for individual feedback, but working with the whole class is much more efficient. Of course, good feedback does not usually just involve the teacher talking to the class and explaining things, but encouraging students to show how they think. Having students working together and teaching each other can often help them to understand concepts better.
Besides proficiency exams, are there any tools to compare the students’ level to the CEFR? How I can evaluate them according to the CEFR? For example, a B2 student should be able to do this and that.
One of the aims of the CEFR is to help teachers and students to understand their level without using tests. Students can use the CEFR to judge their own level, see what languages can be used for at different levels, and evaluate other peoples’ performance. The European Language Portfolio (http://www.coe.int/en/web/portfolio) is a great place to start looking for ideas on using the CEFR in the classroom.
Practice tests can be practised in class, where students are asked to practice with new points of language…right?
I think this kind of test would be what I called a progress test. Progress tests give students extra practise with skills or knowledge taught in class as well as checking that they have understood and can apply those skills.
Ideas for testing lesson progress?
Course books and their teachers’ guides have a lot of good suggestions and materials you can use for English language assessment. There are also some good resource books available with ideas for teachers. I would (of course) recommend my own book, Exploring Language Assessment and Testing (published by Routledge) and (a bit more theoretical) Focus On Assessment by Eunice Jang, published by Oxford University Press.
Why does level B1 always take a longer time to teach? I notice from the books we use…there is B1 and B1+.
The six CEFR levels A1 to C2 can be divided up into smaller steps. In the CEFR there are ‘plus’ levels at A2+, B1+ and B2+. In some projects I have worked on we have found smaller steps useful – such as A1.1, A1.2, A1.3. Generally, real improvements in your language ability take longer as you progress. Thinking just about vocabulary, the difference between someone who knows no words and someone who knows 100 words of a language is very big. The person who knows a few words can do many more things with the language than the person who knows none. But the difference between someone who knows 5,000 words and a person who knows 5,100 words is tiny.
Could you please tell us more about English language assessment?
I’d love to! At the moment I am working with some colleagues around Europe on a free online course for teachers. Our project is called TALE and you can follow us on Facebook: https://www.facebook.com/TALEonlinetrainingcourse/
What CEFR aligned placement test would you recommend?
The best placement test is the one that works most effectively for your students. I’m happy to recommend the Oxford Online Placement Test (OOPT). Whatever system you use, keep a record of how often teachers/students report that a student seems to be in the wrong class. If you find one placement system is not very useful, do try to find a better one.
How reasonable is it to place the keys to the tests in students books?
In the webinar, I said that different tests have different purposes. If the test is for students to check their own knowledge, it would be strange not to provide the answers. If the test results will be used to award grades or certificates, it would be crazy to give the students the answers!
Is cheating an issue with online placement tests?
Again, the answer is ‘it depends’. If cheating is likely to be a problem, security is needed. Online tests can be at least as secure as paper and pencil tests, but if it is a test that students can take at home, unsupervised, the opportunity to cheat obviously exists.
Could you please explain how adaptive comparative judgement tests work? Which students are to be compared?
Adaptive comparative judgement (ACJ) is a way of scoring performances on tests of writing and speaking. Traditionally, examiners use scales to judge the level of a piece of student work. For example, they read an essay, look at the scale and decide ‘I think this essay matches band 4’.
ACJ involves a group of judges just comparing work produced by learners. Rather than giving scores on a predetermined scale, each judge looks at a pair of essays (or letters, or presentations etc.) and uses their professional judgement to decide which essay is the better of the two.
Each essay is then paired, and compared, with a different essay from another student. The process continues until each essay has been compared with several others. ACJ provides the technology for the rating of Speaking and Writing responses via multiple judgements. The results are very reliable and examiners generally find it easier to do than rating scales. Take a look at the website nomoremarking.com to learn more.
Besides the CEFR, what we can use to evaluate students in a more precise way?
See my answer to the last question for one interesting suggestion. A more traditional suggestion is working with other teachers to agree on a scale to use with your students. Then have training sessions (where you compare the marks you each award to the same written texts or recordings of student work) to make sure you all understand and use the scale in the same way.
Can you suggest applications for correcting MCQ tests?
How can placement tests be applied in everyday classrooms where they are split-level classes and students with disabilities learning together with others? What about people with some sort of disability/impairment (eg. dyslexia)
Sometimes there are good reasons to mix up learners of different levels within a class. Tests are not always the most suitable means of deciding which students should be in which class. Where learners have special needs, decisions about placement may involve professional judgement, taking into consideration the nature of their needs and the kinds of support available. In most circumstances, placement should be seen as a provisional decision. If teachers and learners feel that one class is not suitable, moving to another class should be possible.
What about just giving a practice test before a major summative assessment at the end of a semester?
Yes, that seems a good idea. If students aren’t familiar with an English language assessment, they may perform poorly because they get confused by the instructions or misjudge the time available. Having a little practice is usually helpful.
If you missed Tony’s or any of our other PD webinars, why not explore our webinar library? We update our recordings regularly.
Professor Anthony Green is Director of the Centre for Research in English Language Learning and Assessment at the University of Bedfordshire. He is a Past President of the International Language Testing Association (ILTA). He has published on language assessment and in his most recent book Exploring Language Assessment and Testing (Routledge, 2014) provides teachers with an introduction to this field. Professor Green’s main research interests are teaching, learning, and assessment. Today, we share some of the questions and answers asked of Tony during his recent webinar, Assessment for the Language Classroom.