|
|
|
| 2018年度 異文化言語教育評価論 |
Chapter
9 Coming to Terms with Assessment Innovation: Conclusions and Recommendations
Part 1 (9.1～9.4: p.189～p.201)
Yukimi
Hayafune
Modern
Language and Cultures, 1st year student of doctor’s course
(P.189)
9.1 Introduction
This chapter overviews and concludes “interact”,
which was introduced to New Zealand’s second language speaking assessment in
2011.
Ultimately, interact can only be considered useful or fit for purpose to
the extent that it does promote the collection of useful and meaningful
evidence of students’ spoken communicative proficiency.
9.2 Theoretical Underpinnings of Interact
The introduction of interact was intended to move
the focus of spoken assessment away from the high-stakes nature of a ‘one-time
testing event’ and towards viewing the assessment as fundamentally integrated
into on-going classroom work.
The move was intended to be towards genuine, authentic and natural interactions
that would tap into, and provide evidence of, students’ proficiency benchmarked
against a broader spoken communicative proficiency construct.
A further driver for interact may be derived from a theoretical perspective
that peer-to-peer interaction is thought be beneficial to students’ language
acquisition.
The end-goal, whether viewed cognitively or socio-culturally, is automaticity
in use.
(p.192)
9.3 Summary of Findings
9.3.1 Overview
At first sight, there are two parallel but contrary
discourses at work among the teachers. These contrary discourses may serve to
shed some light on the range of views held by the students.
On the one hand, there is a discourse that highly favors interact as an authentic
and valid measure of FL students’ spoken communicative proficiency, leading
to positive impact and interaction for many students.
On the other hand, there is a contrary discourse that questions the usefulness
of interact because it is highly impractical, has negative impact on some
students, and cannot really live up to the expectations for ‘spontaneous
and unrehearsed’.
(P.193)
9.3.2 Positive Dimensions of Assessments Such as
Interact
In comparison with one-time summative teacher-led
interview tests such as converse, the on-going peer-to-peer interactions
anticipated by assessments such as interact have clear advantages.
Assessments such as interact are also able to measure a broader
construct of communicative proficiency than one-time interviews.
On the whole, students are more relaxed and less stressed.
Positive interaction is heightened by developing tasks, perhaps in consultation
with the students, that students perceive as relevant and interesting,
something they both want to talk about and have some knowledge to talk about,
thereby promoting “meaningful language communication” that will elicit the most
meaningful instances of proficiency.
Teachers’ feedbacks:
In the open-ended comments, two French teachers had their positive comments
as follows:
- Students enjoy working together, encourage each other to do well. It is in
their interests to try hard and to work co-operatively.
- I like the idea of spending more time conversing in authentic type
situations. The students’ skills did improve.
One Japanese teacher noted:
- students responded very well…. portfolio work where they could submit work
they were happiest with meant they were highly motivated to achieve. ..they
enjoyed interacting with each other. ..my lessons now far more focused on
communication skills … and thus far more enjoyable… the emphasis has totally
changed from the old standard which was rote learnt.
Students’ feedbacks:
“I thought that it was a really good opportunity to actually speak in
French, and I’m glad I did it otherwise I wouldn’t have known what I could
actually do.”
“it did give me a sense of accomplishment and increase my confidence in
speaking Japanese”
“I really enjoyed the interact standard. It was not at all stressful or
hard, and did not make me anxious at all. I found it was a good opportunity to
demonstrate my speaking ability.”
(p.195)
9.3.3 Negative Dimensions of Assessments such as
Interact
From the perspective of interlocutor variables (who
was paired with whom), it was not always possible for the students to
demonstrate their full proficiency. It also casts questions on the validity and
fairness of the assessment.
Some students reported that several tasks they were asked to complete did not
interest them or were perceived to be irrelevant.
Also, several students responded that when interact was operationalized
as a series of ‘assessment events’, there was in fact minimal washback into
classrooms, and other skills (such as writing) tended to dominate classroom
practice.
When it comes to test, interactions were disauthenticated and a sense of
stress was enhanced.
In that respect, interact could confer neither advantage nor
disadvantage over converse.
(p.196)
9.4 Static or dynamic: A Fundamental Problem
9.4.1 Is Interact a Test?
The largest issue facing interact going forward is
that there is a tendency for stakeholders to approach interact as if it
is a test.
When the perceptual (and perhaps actual) emphasis of interact is on a
test-like format, it is not surprising that several negative perceptions ensue.
Some of them are as follows:
- interact takes time and is more impractical than converse. it
requires of the collection of at least three times the amount of evidence, and
throughout the year
- there was no perceived difference between the two assessments in terms of
student stress
- some teachers viewed negative impact in unequivocal terms (‘ridiculous’ and ‘unrealistic’)
- teachers and students were continuing to resort to practices
Margaret (in Chap.8) illustrates a perspective that an approach that
embeds the assessment seamlessly would not be acceptable. She cannot accept the
seamless change to interact as the assessment does take time.
Peter (in Chap.6) brings out two negative corollaries for students of
continuing to focus on interact as an assessment event. When students
considered interact is also a test, the interaction no longer sounds
natural and is not spontaneous.
As the author noted in Chap.2:
The big dilemma is that the two assessment
paradigms are not mutually exclusive. We cannot say that either one is ‘right’ or ’wrong’,
‘better’ or ‘worse’. They are just different.
What is the goal of the assessment?
What do we want to know by virtue of the assessment data? The question of how best achieve the goal is
arguably unanswerable until we are clear about what we want to measure.
(P.199)
9.4.2 What Do We Want to Measure?
Ideally, we want to capture instances of
automaticity in operation.
The fundamental issue at stake, seen in the light of the intentions of
interact, the data generated from this project and the range of perspectives
presented regarding ‘spontaneous and unrehearsed’, is this: the need to capture
instances of interpersonal interaction in the FL that, regardless of the level
of preparation that has preceded them, and regardless of the language used
(simple or sophisticated), provide evidence of automaticity from which
conclusions regarding future real-world appropriate interactive language use
can be drawn. This is what we want to measure.
Discussion point:
1. In New Zealand assessment reform, they set “automaticity”
as a goal. Can you think of any goal of speaking assessment in a context in
your country? Think of an idea in a context such as to enter university or high
school, end-term test for high school or junior high school students and so on.