The
summary of Developing standardized test of language proficiency:
By:
I.G.A. Lokita Purnamika Utami and Rina Sari
Standardized test for language proficiency presuppose a
comprehensive definition of proficiency.Swain (1990) refers proficiency
assessment to three linguistic traits: grammar, discourse and sociolinguistics
that can be measure through oral, multiple choice and written responses.
Another definition of proficiency is offered by ACTFL which offer a more holistic
and unitary view: superior, advanced, intermediate and novice.
In the planning and
developing language proficiency test, as any assessment development principles,
the purpose of the test must be clearly specified in order to have a valid
interpretation in terms of scores attained from the test. An assessment may be
appropriate for one purpose may be inappropriate for the others. For example,
an assessment which is appropriate for academic purpose may not be useful for
screening job candidates in the workplace. For example the purpose of TOEFL is
to evaluate the English proficiency of people whose native language is not
English. More specifically, to help institution make valid decision in terms of
English proficiency based on their requirement.
It is also essential to
develop the language proficiency test based on a good construct. A good
construct of an assessment means an assessment have a strong theoretical
foundation which lead to a valid assessment. To make sure the test have a good
construct, assessment specification can
be used to define what specific language knowledge, skills or abilities are
intended to be measured. Before specs can be addresses, a comprehensive program
of research must identify a set of construct underlying the test itself. Test
specification commonly includes the purpose, the target users, and the
blueprint outlining the types of test task and how they will be scored.
What should be
considered too is the item selection and arrangement. The item for the test
task should be selected and arranged carefully. For example TOEFL test uses
content coding to make sure that the test takers receive test questions which
measure a variety of skills, a variety of contents which are universally
understood by the test takers from different academic fields.
Test developer should
consider making different evaluation for different kind of items. For example
multiple choice items should be evaluated through looking at item facility,
item discrimination and distractor analysis. These 3 indices are a must since
the test are standardized and will be marketed commercially and administered a
number of times. For other types of responses, namely production responses,
other form of evaluation is required. The principles of reliability and
practicality is important a long with the concept of facility. For practicality
issue, standardized test should involve, for example, the clarity of direction,
timing of the test, ease of administration and how much time is required to
score responses. For reliability issue, more than one scorer are employed to
maintain the standard. Facility is also a key success for a valid item for
example: clear topic, uncultured bias
item, clear direction, simple language etc.
The test developer
should also specify scoring procedure and reporting format. TOEFL has quite
straight forward scoring procedure. Scores are calculated and reported (a)
three section of TOEFL (essay writing are combined with the Structures and
Written expression score) and (b) a total score (range to 40-300 on
computer-based TOEFL and 310-677 on the paper and pencil TOEFL. A separated
essay score (range from 0 to 6) is also provided.
Any standardized test,
once developed, must have an ongoing construct validation studies. For example
the TOEFL program has an impressive program of research. This is important to
maintain or even improve its effectiveness.
Reference
Brown, H. D. 2003. Language
Assessment: Principles and Classroom Practices. White Plains, NY: Pearson Education.
Tidak ada komentar:
Posting Komentar