The article stresses the importance
of discourse as a part of communicative competence. There is a need for testers
to come up with different ways to assess discourse skills. I thought the quote
at the beginning by McNamara was very true. It spoke about how oral assessment
models focus on a candidate instead of on how they interact. Since the language
tester is seeing how the candidate can orally communicate, it is important that
the tester is studying the interaction part. Assessments need to be in context,
because if they aren't the tester cannot truly understand how the candidate
communicates. The use of direct tests proves this. When a candidate is orally
tested using a direct test, he/she uses more elaborate and indirect language
and involves more pragmatic and social devices.
Chapter 23
was about language assessment. It outlined different assessment types, such as performance
based. These portfolios, projects, and experiments show how students progress
without them having to memorize facts to copy down onto a pencil/paper test.
Even though traditional testing offers higher levels of practicality, those
types of tests are not individualized. This does not mean that they cannot be
used. More authentic assessment types take longer to create, implement, and
grade. I suggest placing more emphasis on alternative assessment methods
because they show more of what a student is capable of doing.
Brown’s
chapter 24 went into detail on classroom based assessment. It also gave
practical steps to creating a test, my favorite one being “form your
objectives, draw up test specifications” because it gave a sample outline of a
test. It encompassed many domains by including listening, multiple choice, and
writing production. When it comes to alternative assessment types, I think that
portfolios truly show the progress of a student from the beginning of the year
until the end. It allows the student, instructor, and parents to see the work
of the student in the beginning of the course and how/if they improved with
various assignments.
No comments:
Post a Comment