On Assessment – part 1
Yesterday evening the discussion on #edchat was about assessment. As usual, a stream of thought-provoking tweets and a lively discussion took place. I had been thinking about writing a post on assessment and my thoughts on the matter, so I feel now is a good time to do it.
As I see it, there are certain principles which should be taken into account when we discuss assessment, and I borrow these from Brown (Language Assessment – Principles and Classroom Practices) and some other readings.
To begin with, assessment is part of teaching, just as tests are part of assessment. This means that there is the group TEACHING, the subgroup assessment, and, inside assessment, the subgroup of testing. Teaching is more than assessing, just as assessing is more than simply testing. I guess the first problem lies there. Many teachers tend to merely equate assessing with testing. I’m going to start this series talking about testing.
Tests are usually standardised and tend to measure discrete points of what has been taught. They’re tools teachers have to gauge how much students have learned. Tests are summative instead of formative, i.e. they aim at measuring and summarise what has been taught through a period of time, and usually come at the end of a unit or a course. Tests, just as any kind of assessment, may be good or bad, they aren’t necessarily the bad guys of education. The way most teachers have been treating tests is the main problem, I guess.
If we think about the principles Brown mentions in his book (practicality, authenticity, reliability, validity, and washback), standardised tests are high on practicality by nature. This means they’re usually easy to administer and grade. There’s nothing wrong with this aspect of standardised tests per se. However, any kind of assessment should provide learners with meaningful and effective feedback. Learners should be able to use their tests results to find out how to improve and what they need to work on. Unfortunately, most teachers don’t do anything else but giving students a grade, be it a number or a letter. Students need guidance to find out what they need to study. Now, it’s not that most teachers don’t want to give students useful feedback, but, depending on the context, it’s simply impossible.
There are classrooms around the world with 50 students, and some teachers have to teach 16 or 18 groups. This means some teachers have more than 800 students. Not only do these teachers have to plan their lessons, but they also need to design and grade all these tests, and they usually are forced to have reports on students’ progress every other month. Now if teachers have 16 to 18 groups of 50 minutes each, they’re in the classroom around 30 hours a week. Add to that all the time it takes to assess students outside class, planning lessons, and being an educator in the core meaning of the word (worrying about each student and his or her learning, and empowering your learners), then you tell me how such an educator would be able to radically change his way of assessing students, going from summative to formative, using portfolios (for instance) instead of standardised tests, or tests made by the teacher him or herself. This means keeping track of 800+ students’ writing. I can’t blame teachers for not doing that. Besides, if a teacher has to assess that many students, there’s the serious risk of rater-reliability issues. But this is something for another post.