Views From the Chalkface: English Language School-Based Assessment in Hong Kong

Abstract

The Hong Kong Examinations and Assessment Authority (HKEAA) has recently moved from norm-referenced to standards-referenced assessment, including the incorporation of a substantial school-based summative oral assessment component into the compulsory English language subject in the Hong Kong Certificate of Education Examination (HKCEE). Starting in Form 4, teachers now assess their own students' oral English language competencies through a range of classroom-embedded activities over 2 years (SBA Consultancy Team, 2005). This high-profile assessment initiative marks a significant shift in policy as well as in practice for the HKEAA. Although school-based assessment (SBA) is in line with the Education and Manpower Bureau's general move to align assessment with curriculum reforms, in the early stage of implementation the reforms raised a number of concerns in the wider school community, including sociocultural, technical, and practical concerns. This article first describes the specific content and structure of the HKCEE English Language SBA component. It then reports on the result of the initial analysis of teachers' and students' responses to the initiative in the first stage of its implementation, including the perceived benefits for learning and teaching. The article concludes with a brief overview of how this initial analysis led to the development of a number of subsequent research studies aimed at monitoring and developing teacher knowledge and skills and evaluating more systematically the impact of the reform on teachers, students, and schools in Hong Kong.

Full text

Title
Views from the chalkface: English language school-
based assessment in Hong Kong
Author(s) Davison, C
Citation
Language Assessment Quarterly, 2007, v. 4 n. 1, p. 37-
68
Issue Date 2007
URL http://hdl.handle.net/10722/57428
Rights
Language Assessment Quarterly. Copyright © Lawrence
Erlbaum Associates, Inc.

LANGUAGE ASSESSMENT QUARTERLY, 4(1), 37–68
Copyright © 2007, Lawrence Erlbaum Associates, Inc.
HLAQ1543-43031543-4311language Assessment Quarterly, Vol. 4, No. 1, April 2007: pp. 1–47language Assessment Quarterly
Views From the Chalkface: English
Language School-Based Assessment
in Hong Kong
Views from the ChalkfaceDAVISON
Chris Davison
The University of Hong Kong
The Hong Kong Examinations and Assessment Authority (HKEAA) has recently
moved from norm-referenced to standards-referenced assessment, including the
incorporation of a substantial school-based summative oral assessment component
into the compulsory English language subject in the Hong Kong Certificate of Edu-
cation Examination (HKCEE). Starting in Form 4, teachers now assess their own
students’ oral English language competencies through a range of classroom-embed-
ded activities over 2 years (SBA Consultancy Team, 2005). This high-profile assess-
ment initiative marks a significant shift in policy as well as in practice for the
HKEAA. Although school-based assessment (SBA) is in line with the Education and
Manpower Bureau’s general move to align assessment with curriculum reforms, in
the early stage of implementation the reforms raised a number of concerns in the
wider school community, including sociocultural, technical, and practical concerns.
This article first describes the specific content and structure of the HKCEE English
Language SBA component. It then reports on the result of the initial analysis of
teachers’ and students’ responses to the initiative in the first stage of its implementa-
tion, including the perceived benefits for learning and teaching. The article con-
cludes with a brief overview of how this initial analysis led to the development of a
number of subsequent research studies aimed at monitoring and developing teacher
knowledge and skills and evaluating more systematically the impact of the reform on
teachers, students, and schools in Hong Kong.
The Hong Kong Examinations and Assessment Authority (HKEAA) has recently
moved from norm-referenced to standards-referenced assessment, including the
Correspondence should be addressed to Chris Davison, Faculty of Education, University of Hong
Kong, HOC 125C, Hui Oi Chow Building, Pokfulam, Hong Kong, SAR, China. E-mail: cdavison@
hkucc.hku.hk

38 DAVISON
incorporation of a substantial school-based summative oral assessment compo-
nent into the compulsory English language subject in the Hong Kong Certificate
of Education Examination (HKCEE), a high-stakes examination for all Form 4–5
(F4–5) students.
1
In school-based assessment (SBA), assessment for both formative and sum-
mative purposes is integrated into the teaching and learning process, with
teachers involved at all stages of the assessment cycle, from planning the
assessment programme, to identifying and/or developing appropriate assess-
ment tasks right through to making the final judgments (see SBA Consultancy
Team, 2005, for a detailed description of the activities). As assessments are
conducted by the students’ own teacher in their own classroom, students are
meant to play an active role in the assessment process, particularly through the
use of self- and/or peer assessment used in conjunction with formative teacher
feedback.
This high-profile assessment initiative, led by a team of researchers at the Fac-
ulty of Education, University of Hong Kong, in partnership with the HKEAA,
marks a significant shift in policy as well as in practice for the HKEAA.
2
The ini-
tiative aims to align assessment more closely with the current English language
teaching syllabus (Curriculum Development Council, 1999) as well as the new
outcomes-based Senior Secondary curriculum, to assess learners’ achievement in
areas that cannot be easily assessed by public examinations and at the same time
enhance student self-evaluation and lifelong learning. Although this is in line
with the Education and Manpower Bureau’s general move to align assessment
with curriculum reform (Curriculum Development Institute, 2002), in the initial
process of implementing the SBA initiative a number of challenges arose. This
article focuses on these challenges through an analysis of the perceptions of a
range of F4 teachers and students involved in the initial introduction of the
reform.
3
Studies of the impact of earlier changes in the Hong Kong external exami-
nation system in English language (e.g., Andrews, 1994; Andrews, Fullilove,
& Wong, 2002; Cheng, 1998, 2005) found that changes to summative assess-
ment did not automatically lead to improvement in learning, as the teacher and
school mediated the nature of the change. Studies of the implementation of the
1
For a full description of the 2007 HKCEE English Language syllabus, including the external
examinations, see http://web.hku.hk/∼sbapro/doc/Annex%206%20Revised%202007%20HKCEE%
20Eng%20Lang%20Syll.pdf
2
The research team was led by the author, Dr. Chris Davison, and Professor Liz Hamp-Lyons,
Faculty of Education, University of Hong Kong.
3
Subsequent studies are supporting and monitoring the implementation of SBA over its first
2 years, as well as researching the impact of the reform on a number of schools, see <http://
web.hku.hk/∼sbapro/projects.html> for a full list of these studies.

VIEWS FROM THE CHALKFACE 39
Target-Oriented Curriculum in Hong Kong primary schools (e.g., Adamson &
Davison, 2003; Carless, 2004; Cheung & Ng, 2000) also found assessment
innovation to be severely constrained by traditional school culture and by
teacher, parent, and student expectations. Studies of SBA in other subject
areas in Hong Kong, such as the Teacher Assessment Scheme (Yung, 2001),
also suggest that there may be wide variation in teachers’ interpretations of
student performance and of their role in the assessment process.
Although SBA as an integral component of the formal senior secondary
examination system is established practice in a number of educational systems
internationally, including Australia, New Zealand, and the United Kingdom
(Black, 2001; Black, Harrison, Lee, Marshall & Wiliam, 2003; Black &
Wiliam, 2001; Sadler, 1989; Wiliam, 2001), as well as in some developing
countries (Chisholm et al., 2000; Pryor & Akwesi, 1998; Pryor & Lubisi,
2002), there has been little specific research into the large-scale use of SBA in
English as a second or additional language. In Asia there are embryonic
attempts to develop SBA in Singapore and Malaysia as a complement to exter-
nal examinations at the senior secondary level but virtually no research into
the issue. In Australia, several studies of the use of large-scale criterion-refer-
enced English as a Second Language assessment frameworks in schools
(Breen et al., 1997; Davison & Williams, 2002) have revealed a great diversity
in teachers’ approaches to assessment, influenced by the teachers’ prior expe-
riences and professional development, by the assessment frameworks and
scales they used, and by the reporting requirements placed on them by schools
and systems. Concerns have been raised about mechanistic criterion-based
approaches to SBA, which are often implemented in such a way that they
undermine rather than support teachers’ classroom-embedded assessment pro-
cesses (Arkoudis & O’Loughlin, 2004; Black & Wiliam, 1998; Davison, 2004;
Leung, 2004b).
Research into SBA internationally is further complicated by the considerable
uncertainty and disagreement around the concept and by its intrinsically
teacher-mediated, coconstructed, and context-dependent nature (Black & Wil-
iam, 1998; Brookhart, 2003; McMillan, 2003; McNamara, 1997; Stiggins,
2001). Traditional conceptions of validity and reliability associated with the
still-dominant psychometric tradition of testing are themselves a potential threat
to the development of the necessarily highly contextualized and dialogic prac-
tices of SBA (Hamp-Lyons, 2006; Rea-Dickins, 2006). In a traditional exam-
dominated culture, formative and summative assessment are seen as distinctly
different in both form and function, and teacher and assessor roles are clearly
demarcated, but in the new SBA component of the HKCEE English Language,
summative assessments of the students’ speaking skills are meant to be used for-
matively to give constructive student feedback and improve learning. Hence, the
implementation of the HKEAA English SBA initiative has both theoretical

40 DAVISON
importance and significant practical implications at the local and international
level.
In this article I first describe the specific content and structure of the HKCEE
English Language SBA component, then report on the result of the initial analy-
sis of teachers’ and students’ responses to the initiative in the first stage of its
implementation, including the perceived benefits for learning and teaching. The
article concludes with a brief overview of how this initial analysis led to the
development of a number of subsequent research studies that aim to monitor and
develop teacher assessment knowledge and skills and more systematically evalu-
ate the impact of the reform on teachers, students, and schools in Hong Kong.
THE HKCEE ENGLISH LANGUAGE SBA COMPONENT: ITS
CONTENT, STRUCTURE, AND PROCESSES
The SBA component, worth 15% of the total HKCEE English mark, involves the
assessment of English oral language skills based on topics and texts drawn from
a programme of independent extensive reading/viewing (“texts” encompass
print, video/film, fiction, and nonfiction material). At the time the SBA was
introduced, students were required to choose at least four texts to read or view
over the course of 2 years; keep brief notes in a logbook; and undertake a number
of activities in and out of class to develop their independent reading, speaking,
and thinking skills. For assessment it was suggested they to participate in several
interactions with classmates on a particular aspect of the text they have read/
viewed, leading up to making a more formal group interaction or an individual
presentation on a specific text and responding to questions from their audience.
The assessment format and requirements as originally specified in the introduc-
tion to the SBA in September 2005 are summarized in Table 1.
4
In terms of assessment, an important distinction is made between the two
kinds of oral activities—presentation and interaction—which are characterized
by distinctly different organisational and communicative strategies. An individ-
ual presentation may be quite informal, depending on task and audience, but
requires comparatively long turns, hence a more explicit structure and an ability
to hold the attention of the audience. In contrast, an interaction, an exchange of
4
These initial requirements were modified slightly as a result of teacher concerns over workload,
systematically documented by the SBA developers. Adjustments included reducing the number of
texts from four to three and reducing the number of tasks, but the assessment focus, procedures, and
criteria remain unchanged. In this article I present the initial requirements, as that was the document
to which the teachers and students discussed in this study were responding (see http://web.hku.hk/
∼sbapro/doc/LET–2007%20CE%20SBA.pdf for the modifications that have been made since
September 2005).

VIEWS FROM THE CHALKFACE 41
short turns between two or more speakers, requires less explicit structuring but
more attention to turn-taking skills and planning how to initiate, maintain, and
control the interaction through suggestions, questions, and expansion of ideas.
Both activities, or text-types, also require the students to speak intelligibly with
suitable intonation, volume, and stress, using pauses and body language such as
eye contact appropriately and effectively, and to draw on a range of varied
vocabulary and language patterns.
A variety of assessment tasks can be used to elicit the required kinds of oral
language from students, including teacher-made tasks adapted from one of the
exemplars collected from F4 and F5 teachers as part of the trial of the assessment
initiative (see Appendix A for an example of one of these assessment tasks).
Assessment tasks can vary in length and complexity according to a number of
factors, including the communicative function, the number of people involved,
the position and status of the people interacting, and the nature of the response
required. This diversity of assessment tasks aims to ensure schools can provide
students with appropriate, multiple, and varied opportunities to demonstrate their
oral language abilities individually tailored to students’ language level and inter-
ests. For instance, in an individual presentation, the more orally proficient
TABLE 1
Initial Hong Kong Certificate of Education Examination English Language School-Based
Assessment Requirements
R
equirements F4 F5 Total
No. and type of texts
to be read
Minimum of two texts,
from two categories
Remaining two texts,
remaining categories
Four texts, one from
each category (print
fiction, print
nonfiction, nonprint
fiction, nonprint
nonfiction)
No. and timing of
assessment tasks to
be undertaken
Minimum of two
interactive tasks to
be undertaken
anytime during F4,
must be on different
texts
Minimum of one
interactive task, one
individual
presentation to be
undertaken anytime
during F5, must be on
different texts
Four tasks on four
texts, one from each
category
No., %, and timing of
marks to be
reported
One mark, best mark
out of the two tasks,
5% of total English
mark, reported at end
of F4
Best mark for the
interaction and best
mark for the
presentation, 10% of
total English mark,
reported at end of F5
15% of total English
mark
Note. Source: Hong Kong Certificate of Education Examination (2005, p. 5).

42 DAVISON
students can be challenged by being asked to persuade the whole class to read a
particular book, whereas the less orally proficient students can be asked to
describe the physical appearance of a particular character to a friend. In terms of
group interaction, where each student has read different texts, the more orally
proficient students can be challenged by being grouped into four and being asked
to agree on which book should be set as a class reader, and the less orally profi-
cient students can be placed in pairs and asked to find the three most important
differences between their texts. Students in the same school, even the same class,
may do different tasks or view different texts, so long as they all have the oppor-
tunity to produce the required type of oral language.
To ensure that the oral language produced is the student’s “best” own work, and not
the result of memorisation without understanding, there are several mandatory assess-
ment conditions (SBA Consultancy Team, 2005, pp. 7–8). First, students must be
assessed by their usual English teacher, in the presence of one or more classmate(s).
Second, students must be familiar with the type of task used for assessment and given
sufficient opportunity to produce enough oral language to be confidently assessed. To
facilitate this process, teachers are allowed to ask the students questions as appropriate
to prompt or extend the range of oral language produced and/or to verify the students’
understanding of what they are saying. Third, students are not permitted to refer to
extended notes nor take any notes during the assessment activity.
Students are assessed according to a set of assessment criteria, consisting of a
set of descriptors at each of six levels across four domains (see Appendix B,
Assessment criteria), which were developed and trialled by teachers and students
from a wide range of Hong Kong schools. The domains are briefly described next.
Domain 1: Pronunciation and Delivery
Pronunciation comprises phonology and intonation. Phonology includes the
articulation of individual sounds and sound clusters, whereas intonation refers to
the flow of words with appropriate stress and rise/fall across the sentence(s).
Delivery is made up of two important subaspects: voice projection and fluency.
Fluency refers to the naturalness and the intelligibility of a person’s speech.
Domain 2: Communication Strategies
Communicative strategies involve body language, timing, and asking and
answering appropriate kinds of questions. Body language includes gaze, facial
expressions, head movement, and body direction—the more students rely on
notes or memorized material, the weaker their body language is likely to be.
Timing is important; if student takes too long for an individual presentation the
audience may get bored; if the student is too brief, she or he will not be able to
give enough ideas or support.

VIEWS FROM THE CHALKFACE 43
Domain 3: Vocabulary and Language Patterns
The vocabulary and language patterns domain consists of three important areas:
vocabulary and language patterns (including the quantity, range, accuracy, and
appropriacy), and self-correction/reformulation.
Domain 4: Ideas and Organisation
The ideas and organisation domain consists of the expression of information and
ideas, the elaboration of appropriate aspects of the topic, organisation, and ques-
tioning and responding to questions. Organisation works differently in individual
presentations and in group interactions. In a group interaction students share the
responsibility for providing enough ideas and information to carry the dialogue
forward. They need to stay focused on the topic and say something at the right
time to move the conversation forward by elaborating on a point another group
member has made or by bringing up a new but relevant point. This kind of orga-
nizing is much harder to do in spoken than in written language, so in F4 and F5
group interactions it is not emphasised very much. However, in an individual
presentation the speaker has sole responsibility for planning what she or he will
say and how, and each student is expected to have thought how to organise what
he or she will say.
Within each domain each feature needs to be weighed against the others
holistically to reach an overall judgment. In the same way, the levels are con-
ceptualized not as discrete entities but rather as a continuum of development,
thus it is possible to talk of a “strong 5” or a “weak 3.” An assessment record
(see Appendix C) is used to provide a record of the key features of the assess-
ment activity and help standardize the assessment process. In addition, teachers
are encouraged to video- or audio-record a range of student assessments to
assist with standardization and feedback, involving the students as much as
possible (e.g., asking students to collect a portfolio of their oral language
assessments, both formative and summative, using an MP3 player or by video-
recording each other). During the class assessments, which might span a num-
ber of weeks, individual teachers at the same level (i.e., F4 or F5) are encour-
aged to meet informally to compare their assessments and make adjustments to
their own scores as necessary. Such informal interactions give teachers the
opportunity to share opinions on how to score performances and how to inter-
pret the assessment criteria.
Near the end of the school year, there is a formal meeting of all the English
teachers at each level, chaired by the SBA Coordinator in each school, to review
performance samples and standardise scores. Such meetings are critical for
developing agreement about what a standard means (i.e., validity, consistency in
and between teacher-assessors; reliability, public accountability, and professional

44 DAVISON
collaboration/support). The adjusted marks for each student are then listed on
a class record. At the end of each year there is a district-level meeting for pro-
fessional sharing and further standardisation. Each SBA Coordinator is
encouraged to take a range of typical and atypical individual assessment
records (and the video- or audio-recordings) and the class records for sharing.
Once any necessary changes are made, the performance samples are archived
and the scores are submitted to the HKEAA for review. Video and audio
records can be compiled on a CD-ROM for storage. Maintaining notes of all
standardisation meetings and any follow up action is also encouraged so
schools can show parents and the public that it has applied the SBA consis-
tently and fairly. The HKEAA then undertakes a process of statistical
moderation
5
to ensure the comparability of scores across the whole Hong
Kong school system.
THE RESEARCH STUDY: INITIAL PERCEPTIONS
OF ASSESSMENT REFORM
Brindley (1998), in a wide-ranging study of the issues arising in the implemen-
tation of outcomes-based assessment and reporting in language learning pro-
grammes in the 1990s, identified three common types of issues and problems:
what he called political issues, to do with the purposes and intended use of the
assessment; technical issues, primarily to do with validity and reliability; and
practical issues, to do with the means by which the assessment was put into
practice. As SBA is still a very new concept for Hong Kong schools, many con-
cerns and issues were systematically gathered from key stakeholder groups,
including teachers and students, during the process of the development and
initial implementation of the 2007 HKCE English Language SBA component
in 2005.
There were two main stages to the collection of teacher and students’
responses to and perceptions of the SBA initiative. In the first stage in January to
June 2005, prior to the actual implementation of SBA, while the assessment
activities, procedures, and criteria were still being developed and trialled, data
were gathered from the 66 teachers and 513 students in the 21 schools involved
via questionnaires, individual and focus interviews, classroom observation, and
5
The use of statistical moderation for the SBA is controversial but seen as essential to maintain
community confidence in the HKCEE. Each school’s SBA results are compared with the schoolwide
result of the external oral paper, and the scores of each school (but not the scores of individual stu-
dents) are adjusted if there is a marked discrepancy (see -http://www.hkeaa.edu.hk/doc/tas_ftp_doc/
CE-Eng-StatModerate0610.pdf <http://www.hkeaa.edu.hk/doc/tas_ftp_doc/CE-Eng-StatModerate–
0610.pdf> for a more detailed description of the moderation process).

VIEWS FROM THE CHALKFACE 45
stimulated recall. Data included information about teacher and student back-
ground, existing assessment practices, perceptions and beliefs about teaching
and learning, and attitudes toward the assessment reform. In a follow-up
questionnaire in November 2005, 3 months after the formal introduction of
SBA in schools, responses to the initiative were systematically collected
from more than 173 secondary schools in Hong Kong, including both English
and Chinese-medium schools with different student populations, banding lev-
els and geographic locations (N = 500, response rate = 34.60%). Qualitative
data from both rounds of data collection were coded and analyzed using
NVivo (QRS, 2002) and the key themes and patterns identified. Triangula-
tion, peer debriefing, and member checking were then used to test the robust-
ness of the categories.
The remainder of this article deals with the key issues and concerns high-
lighted as a result of this study (the quantitative data are reported elsewhere; see
Davison & Hamp-Lyons, in press). Adapting Brindley’s (1998) taxonomy of fac-
tors, the issues and concerns were broadly classified into three types: sociocul-
tural, technical, and practical. I briefly describe these in turn.
Sociocultural Issues
Brindley (1998) observed that
if the theoretical underpinnings of the (assessment) statements or the testing for-
mats used are seen to be at variance with the strongly held views of powerful inter-
est groups representing particular theoretical or pedagogical orientations, then their
validity may well be publicly challenged, thus greatly reducing the likelihood of
their adoption by practitioners. (p. 62)
Until very recently, assessment practices in Hong Kong were driven by the
need to provide data to select students for higher education or employment
(Biggs, 1995), hence external examination results have traditionally been the
dominant way students and schools (and teachers) have been evaluated and
held accountable, although this may gradually change with the recent introduc-
tion of value-added indices
6
to compare schools more equitably. As in many
6
The value-added approach involves comparing the performance observed with performance expected
of students in the school, using a regression module that examines the performance of all schools at two
time points, for example, at entry to using the F1 attainment scores and at the end of HKCE using the
HKCE grades. A school is classified as “improving” (i.e., it has a positive value-added score) if the
observed English performance of students exceeds the expected English performance by a sufficiently
large amount. Similarly, a school is classified as “declining” (i.e., it has a negative value-added score) if its
students fail to achieve their expected English performance by a sufficiently large amount.

46 DAVISON
other countries in the region (e.g., Cheah, 1998), the traditional role of assess-
ment in the classroom (as opposed to classroom-based assessment
7
) has been
exam preparation.
However, the official adoption of the UK Assessment Reform Group’s (1999)
distinction between assessment for learning, and assessment of learning by the
Education and Manpower Bureau has stimulated the beginnings of a major para-
digm shift from a culture of testing to a learning and assessment culture (see
Hamp-Lyons, 1999, 2006, for a more detailed discussion of this shift in the Hong
Kong context). This shift, in many ways in the opposite direction to most English
-speaking countries, is being accelerated by the introduction of SBA. However,
extensive experience with earlier educational reforms in Hong Kong has shown
that assessment theories from the international research literature cannot be
incorporated into public policy without resolving the fundamental opposition,
even competition, with local “cultures” and institutional discourses (Adamson &
Davison, 2003; Carless, 1999). In Hong Kong cultural assumptions about assess-
ment are deeply entrenched in the wider community, going well beyond Brind-
ley’s (1998) “political” issues to embrace very different preconceptions about
teaching, learning, and the purpose of education. This was strongly exemplified
in this study by a widespread concern among teachers and students (and their
parents) about the purposes of assessment, encapsulated by ongoing and public
debate over the fairness of SBA. Fairness is fundamentally a sociocultural, rather
than a technical, issue, “a justice that goes beyond acting in agreed upon ways
and seeks to look at justice of the arrangements leading up to and resulting from
those actions” (Stobart, 2005, p. 1). In a highly competitive examination-driven
school system such as Hong Kong’s, fairness has traditionally been seen as treat-
ing everyone equally, giving them the same task with the same input under the
same conditions for the same length of time. Not surprisingly, this is also the
standard interpretation in measurement theory. In contrast, one of the underlying
principles of the SBA initiative was that students should be assessed against
criteria, rather than against other students (a core element of a standards-based as
opposed to norm-referenced system). Teachers were also given the freedom to
tailor the assessment task—its focus, timing, and grouping arrangement—to suit
7
I am making a subtle but critical distinction here between the physical location of the assessment
and the assessors/candidates (i.e., in the classroom, rather than an examination hall) and the philo-
sophical orientation of the assessment. Simply changing the physical parameters of the assessment is
not in itself sufficient to change the nature of the assessment. Many teachers reported they cannot
“assess,” only “mark.” They feel unable to make a difference in teaching and learning, to respond to
individual needs, because of community expectations of convergence and commonality. Teachers
feel their assessment processes are expected to change, without the fundamental purposes being
explicitly challenged. Such role conflict results in increasing stress and a decline in perceived teacher
expertise.

VIEWS FROM THE CHALKFACE 47
their students’ very different language levels, interests, and needs. Congruent
with the principles of assessment for learning, it was assumed that “fairness”
meant every student should be given the opportunity to demonstrate his or her
best (similar to Swain’s 1984, notion of “bias for best”
8
). In this sense, fairness
sometimes required students to be treated unequally, for example, students with
very low levels of English might be asked to read a simple story, then assessed by
a simple recount with perhaps extensive scaffolding to ensure they gain a sense
of achievement and learn from the assessment process. It would be considered
“unfair” to give the same task or support to very proficient students, as their lan-
guage would not be extended, and they would invariably underperform. As
Hamp-Lyons (2001) notes, this view of fairness is actually closer to the more
traditional notion of “fairness” embodied in the classical examinations for the
Chinese civil service, that is, that conditions should be consciously created to
make opportunities open to all.
However, many teachers in the SBA trial, despite eagerly endorsing the rheto-
ric of assessment for learning, in practice found it extremely difficult to free
themselves and their students from their existing conceptions. A significant
minority of teachers were found to be treating SBA as if it were a separate, one-
off externally set and assessed exam, albeit located in the school. Some were
very concerned that “students will be using texts of different lengths and levels of
difficulty. The assessment tasks may also be different from school to school. It
doesn’t seem fair.”
Some teachers in the SBA study perceived the differences between the old and
the new assessment cultures as a major stumbling block to assessment reform,
exemplified by the following comment:
I feel it takes time because the culture, the education culture in Hong Kong
is different from other western countries and the students may not be used
to that kind of assessment. They like to do exam paper. They think (then)
they have something to learn.
However, another teacher newly returned from overseas, commented:
Actually I was so surprised that … how slow we Hong Kong people are in
terms of education … because I remembered when I was in Canada, we
never … you would never be graded on just one exam. It’s quite like what
8
Swain (1984) advocated that to elicit the best performance from students in a communicative
classroom, the following conditions must be met: more than adequate time to complete a task, oppor-
tunity given to review and change work, access given to reference materials, checking that students
are on task, clear instructions including what is being assessed, and useful suggestions about how to
do the task.

48 DAVISON
we are trying to do actually, I believe that (assessment for learning) has
been practised in those places for years and I was actually surprised
nobody did anything (here). … I am totally for assessment for learning.
Not surprisingly, the trial schools that were already doing extensive reading
and whose students engaged in oral group work and individual presentations on a
regular basis found it the easiest to integrate the assessment tasks into their exist-
ing practice, albeit with adaptations to meet the SBA guidelines. For example,
one teacher commented:
I just briefly tell the students about the task because it is in mid May, so
they were quite busy that moment. So I asked them just make use of they
been doing say they just, they can just took from ERS report and work on
it, say prepare a better review so that they can just have their presentation
based on the review and I told them the date and the time of presentation.
That’s what I did at the very beginning. Later on, I, I met them some days
later, and I asked them to show me the book review they had written and I
took a very look at it and I found that there weren’t any major problems in
it. So I just returned them the review and they started to prepare those tasks
… and later on, just right before they did the presentation, I helped them
with the vocabulary and the names because they didn’t know how to pro-
nounce them. So I just helped them pronounce them correctly… (but) I
gave them more guidance according to the SBA documents … because the
five questions listed there suggested some sort of high order thinking skills.
… So I try to scaffold them to think in that way.
However, the trial showed that for many teachers, implementing SBA
involved a steep learning curve and a significant change in their approach to
teaching and to their role in student assessment. Responses were quite varied
between schools and sometimes between teachers in the same school. Some
teachers seemed very enthusiastic right from the beginning and stretched their
creativity in thinking about how to build on the opportunities offered by SBA. A
few teachers took much longer to come to grips with the principles involved in
SBA, and their implications for teaching and learning as well as for assessment
practice, as can be seen from the following quote from a trial school:
For students of higher forms, the time (eight minutes) is quite limited.
They can’t have enough time to introduce their books and ask each other
questions.
In fact, the SBA guidelines asked teachers to set their own time limits accord-
ing to the needs of the students, but this was interpreted through the prism of

VIEWS FROM THE CHALKFACE 49
teachers’ existing experience—many schools used buzzers and stopwatches to
allocate an identical period of time to each students, with the result that in some
schools students’ stress levels were high and their “performance” very contrived
and/or rushed.
As an outcome-oriented standards-referenced system, SBA is a significant
cultural and attitudinal change, not only for teachers but for the whole school
community, including students and parents. Hence, it is not surprising that fair-
ness was a deep-seated sociocultural, not just political concern.
Technical Issues
At the technical level, as in Brindley (1998), concerns with SBA expressed by
teachers and students revolved around the understanding and interpretation of
traditional concepts such as reliability, validity, and authenticity. As indicated
earlier, SBA is by its very nature teacher mediated, co-constructed and dialogic,
context-dependent, multiple and varied, and dynamic and evolving (Black &
Wiliam, 1998; Brookhart, 2003; Stiggins, 2001). In many ways SBA is the oppo-
site to traditional testing in which context is regarded as an extraneous variable
that must be controlled and neutralized and the assessor seen as someone who
must remain objective and uninvolved throughout the whole assessment process
(Lynch, 2001). SBA, in contrast, derives its validity from its location in the
actual classroom where assessment activities are embedded in the regular curric-
ulum and assessed by a teacher who is familiar with the student’s work. Like
qualitative research, SBA builds into its actual design the capacity for triangula-
tion, the collection of multiple sources and types of evidence under naturalistic
conditions over a lengthy period of time. Hence, Rea-Dickins (2006) argues:
The traditional positivist position on language testing, with the tendency to map the
standard psychometric criteria of reliability and validity onto the classroom assess-
ment procedures, has been called into question, and the scope of validity has been
significantly broadened (e.g., Chapelle, 1999; Lynch, 2001, 2003; McNamara,
2001) and taken further by a number of researchers. (p. 512)
Teasdale and Leung (2000), attempting to clarify the epistemological bases of
different types of assessment in relation to spoken English language assessment
in mainstream multiethnic classrooms in England, observed that alternative and
psychometrically oriented assessment derive from different intellectual sources.
Extending this argument, Leung (2004a, 2004b) argues that the evaluation crite-
ria traditionally associated with psychometric testing such as reliability and valid-
ity need to be reinterpreted in SBA. Without such reinterpretation, SBA may be
reduced to the testing of linguistic knowledge through a series of summative mini-
achievement tests encouraging rehearsed dialogues with little or no opportunity

50 DAVISON
for spontaneous language use and “disembedded from the flow of teaching and
learning” (Rea-Dickins, 2006, p. x).
On the other hand, some language testing researchers would argue that tradi-
tional test criteria do apply to alternative assessment schemes such as the SBA;
for example, Clapham (2000) observed
A problem with methods of alternative assessment, however, lies with their validity
and reliability: Tasks are often not tried out to see whether they produce the desired
linguistic information; marking criteria are not investigated to see whether they
‘work’; and raters are often not trained to give consistent marks. (p. 152)
These same debates over technical issues arose, and continue to arise, with the
HKEAA SBA initiative, exemplified by some teachers’ concerns that students
would “cheat”—by memorizing whole chunks of spoken text, by overrehearsing
and/or requesting they redo the same task again and again, or by blindly parroting
their partners’ responses in group interaction. Teachers were also concerned that
other teachers would “cheat” by allowing students to take copious notes or read
aloud, or by giving them the same task again and again, or by simply fabricating
results. Favouritism or bias was also seen as an issue by both teachers and stu-
dents. More common was a lack of confidence among teachers, with many
doubting that they had the required knowledge and skills to carry out the assess-
ment properly. As one teacher commented, “I would like the HKEAA to take up
my marks to see if I have interpreted the criteria correctly.” The notions of peer
debriefing and standardisation were foreign to many teachers, and they saw such
processes as just adding to their workload.
However, despite many concerns about reliability and validity, the teachers
who were involved in the trial of the SBA did observe that they were surprised at
the naturalness and ease with which many of their students were able to commu-
nicate. They commented that as the SBA was based on independent and exten-
sive reading/viewing, students often neither had all read or viewed the same
material nor had the same ideas. Thus when individuals made presentations or
interacted with a group, there was often a genuine information gap, hence real
interest among the students in what others had to say, enhanced by the familiar,
more relaxing surroundings of the classroom. These teacher perceptions were
reinforced by similar comments by the students themselves.
Another factor identified by the informants as contributing to greater authen-
ticity in speaking was the actual assessment criteria and processes. For example,
initially some teachers assumed that to do well on the SBA tasks it was necessary
to rehearse students as if for an external exam (and an external examiner). How-
ever, the trial showed very clearly that students who relied on extensive notes or
memorization did much less well in terms of communication strategies. Their
pronunciation and language also suffered when they tried to memorize material

VIEWS FROM THE CHALKFACE 51
they had not fully mastered, thus leading to a tendency to mispronounce words,
produce unnatural intonation, and lose their train of thought. As the assessment
was undertaken by the class teacher, who was familiar with the range of student
performance and who could ask questions to ensure the text was a student’s own
work, many teachers eventually concluded that there was little possibility of
cheating.
The teachers increasingly felt that the reliability of the assessment was also
enhanced by having a series of assessments (rather than just one) by a teacher
who was familiar with the student and by encouraging multiple opportunities for
assessor reflection and standardisation The exemplars of student work (a CD-
ROM set was developed for use as a training package for teacher scoring and
standardisation) provided a starting point for discussing the set standards. How-
ever, even after the trial some teachers assumed that achieving superficial con-
sensus was the key to reliability rather than understanding and interpreting the
assessment criteria and being able to justify a score to others. As a number of
educational researchers have argued, developing the capacity and confidence to
disagree can actually create more reliable conditions for assessment, because it
allows misunderstandings and inconsistencies in the interpretation of the criteria
to emerge and be challenged in a familiar and supportive environment (Davison,
2004; Moss, 1994). However, in the trial it was clear that Hong Kong teachers
would not naturally adopt such a position; rather, they needed explicit encourage-
ment and modelling, suggesting this was as much a sociocultural issue as a tech-
nical one.
Practical Issues
It is perhaps not surprising, given what is known from the literature on educa-
tional innovation and change (e.g., Cheung, 2001; Cheung & Ng, 2000), that it
was the immediate and practical issues that most concerned the majority of teach-
ers involved in the initial implementation of the SBA. Practical issues and con-
cerns raised by teachers in relation to implementation of SBA included the
following:
• The need for access to appropriate assessment (and extensive reading)
resources.
• The need for activities and techniques as models/resources.
• Concerns about the type of recordings of oral performance that they were
expected to collect.
• Lack of practical support for teachers at the school level.
• Concerns about the adequacy of professional development in SBA.
• Lack of time to implement and discuss assessments.
• Competing demands and priorities in relation to time allocation.

52 DAVISON
A particular source of misunderstanding early on was the technical requirements
of SBA. Many teachers in the trial and subsequent focus group interviews had
concerns about the level of technical resources and expertise that they thought
were required. Some teachers reported that they did not have access to video
cameras within their school. The SBA guidelines have been modified to empha-
sise even further that recording is optional, and that “homemade” audio-recording
is sufficient for feedback and standardisation purposes. However, many partici-
pating teachers felt video-recording was necessary for them to be able to review
students’ performances and make good judgements, especially when scoring
group interaction with more than two students at one time.
Apart from the issues just mentioned, teachers were also concerned about
whether they would be adequately prepared to implement SBA. Although train-
ing sessions were conducted on the overall design of the SBA component, and on
the assessment criteria and standards, teachers also expressed a desire to better
understand the underlying assumptions of SBA, especially how to integrate
assessment into teaching and learning and how to set up effective assessment
tasks. As a teacher associated with HKEdCity, one of the dedicated Web sites
9
for SBA teacher support, accurately noted, “many teachers have an urgent need
to view others’ practices and share experiences. … We can film the good lessons
for teachers and analyse the lessons. We (need to) focus on teaching instead of
assessment only.” In light of this request, more professional development was
scheduled for all teachers teaching F4 English in 2005–06 so that they could be
supported during the whole of the assessment period in the second semester. An
SBA handbook and an introductory DVD and booklet were also produced (see
http://web.hku.hk/∼sbapro/), so teachers could become more familiar with the
assessment process. In addition the system of district- and systemwide support
was extended, with more training provided for the group coordinators of each
cluster of schools. In general, teachers were also concerned that there was not
sufficient recognition at the school level by principals and panel chairs of the
time needed to discuss ideas and standardise assessments. With the official
launch of teacher-support material along with the ongoing training sessions, it
was hoped that awareness of the importance of SBA would be raised and more
support would be given at the school level.
However, the teachers’ major concern with SBA was their perception that it
was the last of too many new initiatives that they had to juggle, along with their
busy schedules and heavy workload. Teachers were not placated by reassurances
from the HKEAA that once SBA became a routine part of classroom activities,
there would be no significant increase in the workload for students and teachers.
9
The other is the author’s Faculty of Education Web site for SBA projects: http://web.hku.hk/
∼sbapro/.

VIEWS FROM THE CHALKFACE 53
Obviously much more input is needed to show teachers how to integrate assess-
ment, teaching, and learning in schools with still very large classes and insuffi-
cient resources, but their concerns about workload and about the reform being
implemented too quickly without adequate support were also very real (for a
summary of media concerns at the time of the introduction of the SBA, see http://
web.hku.hk/∼sbapro/doc/Chinese%20news%20related%20to%20SBA.doc). In
fact, these concerns over workload and time threatened to derail the whole SBA
initiative until the HKEAA was finally given permission to adjust the timeline
and provide more training and resources.
A POSITIVE IMPACT ON TEACHERS, LEARNERS AND SCHOOLS?
Despite these problems and issues with the design and implementation of SBA,
the responses of teachers and students to the underlying philosophy of SBA and
its emphasis on improving the quality of teaching and learning were generally
very positive. A comment from a teacher involved in the trial was typical:
Personally, I enjoyed this trialling experience. I learnt how to judge the stu-
dents through this activity. Moreover, my students tried to do the presenta-
tion based on the guiding questions given to them. Students found this
presentation quite interesting and motivating. They learnt how to speak
confidently and bravely during this assessment activity. They found this
presentation rewarding since they can learn not only from the book but also
through their actual participating experience.
Those who were involved in the trialling indicated that they were now much
more aware of the values underlying SBA, the principles that they needed
to work with to ensure their students gained the most from it, and the various
options now available to them. For example, one teacher in a trial school
commented:
After my students … had finished the presentation, I give them feedback
and try to improve their performance and … I was really amazed by the
response. … I thought that one of my students actually is not very good in
English, but after the presentation she tried very hard to do it again and
again. Feedback really works and I found that my students have improved
a lot.
Other trial school teachers commented on the increase in confidence among
students and the positive effect of the assessment activity on other language
skills:

54 DAVISON
M: I think my student Sandy, just the one, just videotaped. She has great
improvement … She thinks that is useful, very useful and she told me that
… it also helps her reading. I think the student can really improve a lot
after this trialling.
C: Well, my student has shown a great improvement in terms of confidence
and English proficiency. They like talking to each other in English and
they are not being afraid of being videotaped.
A: Yeah, they start, they start reading English books as well.
S: I think is just like what you said, is very good training for confidence and
the students actually articulate what they read and … they found reading
very useful, purposeful, that is something that you can share with some-
one, is not something just happening in your inner self. So I think the one
that I trained out of the two, one girl called, actually she is a very shy girl
after that I found she gained some confidence and she does quite well
even in her writing. So she did well in this writing exam. I’m not sure if it
is the effect of that training experience (but) it seems that she gained
some confidence during this period of time.
Students also commented very favourably on the assessment activities, as the fol-
lowing extract from an interview with students illustrates:
T: What do you think about the assessment task you did in presentation
task 5?
S1: It was quite interesting that we need to think about what the character
needs. We can buy a gift.
S2: I just think it’s easy to handle it.
T: Why?
S2: It’s quite interesting to think for a gift to the character.
T: When you are thinking of a gift is it difficult?
S2: I don’t think so because I can think of many, many gifts to solve the
problem.
T: Did you enjoy working with your partners, why?
S2: Yes, I did because my partners are all my best friends. We didn’t have any
gaps so we did the project perfect.
T: How about you?
S3: I also very enjoy doing the task with my friends as they know me very
much. When I don’t know what can I say they will help me to continue
the conversation.
S4: With the partner I won’t feel nervous.
Overall, one of the more significant benefits of the SBA initiative identified
by students and teachers was the capacity for the students to comment on their

VIEWS FROM THE CHALKFACE 55
own development and to receive constructive feedback immediately after the
assessment had finished, hence improving learning. Teachers commented that
when examinations are externally set and assessed, the only feedback that stu-
dents and teachers receive is a grade at the end of the year, with no opportuni-
ties for interaction with the assessor and no chance to discuss how to improve.
Teachers and students also commented that ongoing assessment encouraged
students to work consistently. At the same time, it was seen as a source of con-
crete data for the evaluation of teaching and assessment practices. In contrast,
examinations were seen as purely summative in purpose, leading to a focus on
exam technique, rather than outcomes, and outside teachers’ control or even
understanding. However, with SBA many teachers and students saw them-
selves as becoming partners in the assessment process. Although SBA involved
only oral language assessment, some teachers reported that the SBA initiative
had increased their level of awareness and skill in assessment planning, which
they found readily transferable to other areas of the English language curricu-
lum, and beyond.
CONCLUSIONS
The HKEAA SBA initiative is a major assessment reform that entails substan-
tial change in school culture and structures as well as in pedagogic expecta-
tions among students, teachers, administrators, parents, and the wider
community. It requires the development of content- and context-appropriate
assessment activities, instruments, and procedures that are explicitly linked to
high-quality teaching and learning and English language teachers who are not
only confident and skilled at making highly contextualized, consistent, and
trustworthy assessment decisions but also effective at involving students in
the assessment process. These are major challenges on both a theoretical and
practical level, so it not surprising that this early research study at the very
beginning of the initiative has been followed by a number of other studies
looking at how teachers (and students) deal with specific aspects of the SBA
initiative, including designing the assessment task, making grouping deci-
sions, intervening in the assessment process, involving students in self- and
peer assessment, giving feedback, making assessment judgments, and using
summative assessments for formative purposes. There are also studies evaluat-
ing the impact of the SBA initiative on students, on teachers, and on the
English language curriculum. The results of this extensive research will be the
ultimate test of whether teachers and students’ initial concerns are a natural
response to a new initiative or an indication of more deep-seated problems
with the assessment reform itself.

56 DAVISON
ACKNOWLEDGMENTS
An earlier version of this article was presented at the 8th Asian Forum of English
Language Testing Associations, Hong Kong, November 2005. I acknowledge the
support of the HKEAA and the Hong Kong Research Grants Council (RGC HKU
7268/04H) for funding this research, and the extensive assistance with the study
of the SBA Consultancy Team. I am particularly grateful for the constant advice
and encouragement of my colleague and coinvestigator, Professor Liz Hamp-Lyons.
REFERENCES
Adamson, B., & Davison, C. (2003). Innovation in English language teaching in Hong Kong primary
schools: One step forwards, two steps sideways. Prospect, 18, 27–41.
Andrews, S. (1994). The washback effect of examinations: Its impact upon curriculum innovation in
English language teaching. Curriculum Forum, 4, 44–58.
Andrews, S., Fullilove, J., & Wong, Y. (2002). Targeting washback: A case-study. System, 30, 207–
223.
Arkoudis, S., & O’Loughlin, K. (2004). Tensions between validity and outcomes: Teachers’ assess-
ment of written work of recently arrived immigrant ESL students. Language Testing, 20, 284–304.
Assessment Reform Group. (1999). Assessment for learning: Beyond the Black Box. Cambridge,
England: University of Cambridge School of Education. Retrieved November 11, 2005, from http://
www.assessment-reform-group.org.uk/CIE3.pdf
Biggs, J. (1995). Assumptions underlying new approaches to educational assessment: Implications
for Hong Kong. Curriculum Forum, 4(2), 1–22.
Black, P. (2001). Formative assessment and curriculum consequences. In D. Scott (Ed.), Curriculum
and assessment (pp. 7–23). Westport, CT: Ablex.
Black, P., Harrison, C., Lee, C., Marshall, B., & Wiliam, D. (2003). Assessment for learning.
New York: Open University Press.
Black, P., & Wiliam, D. (1998). Assessment and classroom learning. Assessment in Education, 5,
7–74.
Breen, M., Barratt-Pugh, C., Derewianka, B., House, H., Hudson, C., Lumley, T., et al. (1997). Pro-
filing ESL children: How teachers interpret and use national and state assessment frameworks.
Canberra, Australia: Department of Employment, Education, Training and Youth Affairs.
Brindley, G. (1998). Outcomes-based assessment and reporting in language learning programmes: A
review of the issues. Language Testing, 15, 45–85.
Brookhart, S. (2003). Developing measurement theory for classroom assessment purposes and uses.
Educational Measurement, Issues and Practices, 22, 5–12.
Carless, D. (1999). Perspectives on the cultural appropriacy of Hong Kong’s Target-Oriented Curric-
ulum (TOC) initiative. Language, Culture and Curriculum, 12, 238–254.
Carless, D. (2004). Issues in teachers’ re-interpretation of a task-based innovation in primary schools.
TESOL Quarterly, 8, 639–662
Chapelle, C. (1999). Validity in language assessment. Annual Review of Applied Linguistics, 19,
254–272.
Cheah, Y. M. (1998). The examination culture and its impact on literacy innovations: The case of
Singapore. Language and Education, 12, 192–209.
Cheng, L. Y. (1998). Does washback influence teaching? Implications for Hong Kong. Language and
Education, 11, 38–54.

VIEWS FROM THE CHALKFACE 57
Cheng, L. Y. (2005). Changing language teaching through language testing: A washback study
(Studies in Language Testing: Volume 21). Cambridge, England: Cambridge University Press.
Cheung, D. (2001). School-based assessment in public examinations: Identifying the concerns of
teachers. Education Journal, 29, 105–123.
Cheung, D., & Ng, D. (2000). Teachers’ stages of concern about the target-oriented curriculum.
Education Journal, 28, 109–122.
Clapham, C. (2000). Assessment and testing. Annual Review of Applied Linguistics, 20, 147–161.
Curriculum Development Council. (1999). English language education secondary syllabus. Hong
Kong: Government Printer.
Curriculum Development Institute. (2002). School policy on assessment: Changing assessment prac-
tices. In Basic education curriculum guide: Building on strength (chap. 5). Hong Kong: Author.
Davison, C. (2004). The contradictory culture of classroom-based assessment: Teacher assessment
practices in senior secondary English. Language Testing, 21, 304–333.
Davison, C., & Hamp-Lyons, L. (in press). “You mean we are the assessors now?” Changing ESL
teacher assessment practices in Hong Kong secondary schools. Australian Review of Applied
Linguistics.
Davison C., & Williams, A. (Eds.). (2002). Learning from each other: Literacy, labels and limita-
tions. Studies of child English language and literacy development K–12 (Vol. 2). Melbourne: Lan-
guage Australia.
Hamp-Lyons, L. (1999). Implications of the “examination culture” for (English language) education
in Hong Kong. In V. Crew, V. Berry, & J. Hung (Eds.), Exploring diversity in the language curric-
ulum (pp. 133–140). Hong Kong: Hong Kong Institute of Education.
Hamp-Lyons, L. (2001). Fairness in language testing. In A. J. Kunnan (Ed.), Fairness and validation
in language assessment, Studies in Language Testing 9 (pp. 99–104). Cambridge, England:
Cambridge University Press.
Hamp-Lyons, L. (2006). The impact of testing practices on teaching: Ideologies and alternatives. In
J. Cummins & C. Davison (Eds.), The international handbook of English language teaching (Vol. 1)
(pp. 487–506). Norwell, MA: Springer.
Leung, C. (2004a). Classroom teacher assessment of second language development: Construct
as practice. In E. Hinkel (Ed.), Handbook of research in second language learning and teaching
(pp. 869–888). Mahwah, NJ: Lawrence Erlbaum Associates.
Leung, C. (2004b). Developing formative teacher assessment: Knowledge, practice and change.
Language Assessment Quarterly, 1, 19–41.
Lynch, B. (2001). Rethinking assessment from a critical perspective. Language Testing, 18,
351–372.
McMillan, J. (2003). Understanding and improving teachers’ classroom assessment decision-making:
Implications for theory and practice. Educational Measurement: Issues and Practice, 22, 34–43.
McNamara, T. (1997). “Interaction” in second language performance assessment: Whose perfor-
mance? Applied Linguistics, 18
, 446–466.
McNamara, T. (2001). Language assessment as social practice: Challenges for research. Language
Testing, 18, 333–349.
Moss, P. A. (1994). Can there be validity without reliability? Educational Researcher, 23, 5–12.
Pryor, J., & Akwesi, A. (1998). Assessment in Ghana and England: Putting reform to the test of prac-
tice. Compare, 28, 263–275.
Pryor, J., & Lubisi, C. (2002). Reconceptualising educational assessment in South Africa: Testing
times for teachers. International Journal of Educational Development, 22, 673–686.
QRS. (2002). NVivo 2.0: Using NVivo in qualitative research. Melbourne, Australia: Author.
Rea-Dickins, P. (2007). Classroom-based assessment: Possibilities and pitfalls. In J. Cummins &
C. Davison (Eds), The international handbook of English language teaching (Vol. 1) (pp. 505–520).
Norwell, MA: Springer.

58 DAVISON
Sadler, D.R. (1989). Formative assessment and the design of instructional systems. Instructional
Science, 18, 119–144.
SBA Consultancy Team. (2005). 2007 HKCEE English examination: Introduction to the school-
based assessment component. Hong Kong: HKEAA.
Stobart, G. (2005). Fairness in multicultural assessment systems. Assessment in Education, 12,
275–287.
Stiggins, R. (2001). The unfulfilled promise of classroom assessment. Educational Measurement,
Issues and Practices, 20, 5–15.
Swain, M. (1984). Teaching and testing communicatively. TESL Talk, 15, 7–18.
Teasdale, A., & Leung, C. (2000). Teacher assessment and psychometric theory: a case of paradigm
crossing? Language Testing, 17, 163–184.
Wiliam, D. (2001). An overview of the relationship between assessment and the curriculum. In
D. Scott (Ed.), Curriculum and assessment (pp. 165–181). Westport, CT: Ablex.
Yung, B. (2001). Examiner, policeman or students’ companion: Teachers’ perceptions of their role in
an assessment reform. Educational Review, 53, 251–260.

VIEWS FROM THE CHALKFACE 59
APPENDIX A
Sample Assessment Tasks
This is one of the sample assessment tasks developed for the school-based assess-
ment (SBA). Please visit the SBA Web site at http://web.hku.hk/∼sbapro/
sba.html for more sample assessment tasks.
Name of Task: New Neighbours
Oral Text-type:
individual presentation interaction
Communication Functions:
describing reporting explaining discussing
classifying comparing persuading others: ______________
Audience--teacher plus:
a student partner
small groups
class
more than one class
Targeted audience:
fellow students
students from other classes
teacher(s)
others: _______________
Role(s) of audience:
giving non-verbal responses
only
questioning/commenting
interacting with no limitations
Where on this continuum would you place the task?
spontaneous,
informal
dialogue, e.g.
small group
interaction
interactive, planned
yet dialogic, e.g.
semi-formal group
report, interactive
factual report
individual long turn
of planned, spoken
text, e.g. news
reporting, story
telling
individual long turn
that is planned,
cohesive, organized,
formal, e.g. spoken
report, a speech
This task is suitable for use with the following genre(s):
print/non-print fiction
print/non-print biography/autobiography
factual books/documentaries on common topics, e.g. sports, hobbies, travel
books/films on real life issues, e.g. environmental, social, economic
Preparation: none
Description of pre-assessment activities:
1. Ask students to think of an interesting character from a story/class reader that you have taught
recently.
2. Ask them to imagine that one of the characters in the story has moved in next door to them.
3. Ask them to think about what life is like with such a neighbour.
4. Hold a discussion with the students and write down what kind of information they should
cover if they were asked to describe an imaginary day they spent with the new neighbour. The
information may include one or more of the followings:
a) Name and gender of the neighbour
b) What does he/she look like?
c) How does she/she dress at home?
d) What is his/her personality?
e) How does he/she treat his family or people around him/her?
f) What is/are the major event(s) in the story that your character takes part in?
g) Do you like this new neighbour? Why/ Why not?
h) How did you spend your day with this new neighbour? What did you do?
i) What did you learn from this new neighbour?
5. For homework, ask each student to write a description of an imaginary day he/she spent with
“the new neighbour”.

60 DAVISON
6. Remind them to draw references from the books. They can’t turn their new neighbour into a
wonderful person, if the descriptions from the book prove otherwise.
7. In the next lesson, ask students to share what they wrote in small groups.
8. Ask students to nominate the most interesting presentation among their group members.
9. Invite a student from each group to share their presentation with the whole class.
Planned SBA Task:
Ask the students to describe an imaginary day in their lives when they spend time with a character
from a book or film they viewed. Ask them to provide some background information about the
book/film they read/viewed before they describe their imaginary day with their new neighbour during
the individual oral presentation.
Tips/comments:
• The personal responses for this task can provide a good basis for discussion in English at a
comfortable level.
•
If students need more opportunities to speak in public, you may invite each student to take
turns sharing their presentation with the whole class.
Sources:
Adapted from: Andy Barfield’s “Getting Personal.” In Bamford, J & Day, R. (Eds) Extensive reading
activities for teaching language, p. 146-148.

61
APPENDIX B
TABLE B1
School-Based Assessment Criteria for Group Interaction
1. Pronunciation & Delivery 2. Communication Strategies
3. Vocabulary & Language
Patterns 4. Ideas & Organisation
6 Can project the voice appropriately
for the context.
Can pronounce all sounds/sound
clusters and words clearly and
accurately.
Can speak fluently and naturally,
with very little hesitation, and
using intonation to enhance
communication.
Can use appropriate body language to
display and encourage interest.
Can use a full range of turn-taking
strategies to initiate and maintain
appropriate interaction and can
draw others into extending the
interaction (e.g., by summarising for
others’ benefit, or by redirecting a
conversation); can avoid the use of
narrowly formulaic expressions
when doing this.
Can use a wide range of
accurate vocabulary.
Can use varied and highly
accurate language patterns;
minor slips do not impede
communication.
Can self-correct effectively.
Can express a wide range of relevant
information and ideas without any
signs of difficulty.
Can consistently respond effectively
to others, sustaining and extending
a conversational exchange.
Can use the full range of
questioning and response levels
(see Framework of Guiding
Questions) to engage with peers.
5
Can project the voice appropriately
for the context.
Can pronounce all sounds/sound
clusters clearly and almost all
words accurately.
Can use appropriate body language to
display and encourage interest.
Can use a good range of turn-taking
strategies to initiate and maintain
appropriate interaction (e.g., by
encouraging contributions from
others’ in a group discussion, by
asking for others’ opinions, or by
responding to questions); can mostly
avoid the use of narrowly formulaic
expressions when doing this.
Can use varied and almost
always appropriate
vocabulary.
Can
use almost entirely
accurate and appropriate
language patterns.
Can express relevant information and
ideas clearly and fluently.
Can respond appropriately to
others to sustain and extend a
conversational exchange.
(Continued)

62
TABLE B1
(Continued)
1. Pronunciation & Delivery 2. Communication Strategies
3. Vocabulary & Language
Patterns 4. Ideas & Organisation
Can speak fluently with only
occasional hesitation, and using
intonation to enhance
communication, giving an overall
sense of natural nonnative language.
Can usually self-correct
effectively.
Can use a good variety of
questioning and response levels
(see Framework of Guiding
Questions).
4 Can project the voice mostly
satisfactorily.
Can pronounce most sounds/sound
clusters and all common words
clearly and accurately; less
common words can be understood
although there may be articulation
errors (e.g., dropping final
consonant clusters).
Can speak at a deliberate pace, with
some hesitation but using
sufficient intonation conventions
to convey meaning.
Can use some features of appropriate
body language to encourage and
display interest.
Can use a range of appropriate turn-
taking strategies to participate in, and
sometimes initiate, interaction (e.g., by
responding appropriately to others’
comments on a presentation, by making
suggestions in a group discussion).
Can use some creative as well as
formulaic expressions if fully
engaged in interaction.
Can use mostly appropriate
vocabulary.
Can use language patterns that
are usually accurate and
without errors that impede
communication.
Can self-correct when
concentrating carefully or
when asked to do so.
Can present relevant literal ideas
clearly with well-organised
structure.
Can often respond appropriately to
others; can sustain and may
extend some conversational
exchanges
However: Can do these things less
well when attempting to respond to
interpretive or critical questions, or
can interpret information and
present elaborated ideas, but at
these questioning levels coherence
is not always fully controlled.

63
3 Volume may be a problem.
Can pronounce all simple sounds
clearly but some errors of sound
clusters; less common words may
be misunderstood unless supported
by contextual meaning.
Can speak at a careful pace and use
sufficient basic intonation
conventions to be understood by a
familiar and supportive listener;
hesitation is present.
Can use appropriate body language to
show attention to the interaction.
Can use appropriate but simple and
formulaic turn-taking strategies to
participate in, and occasionally initiate,
interaction (e.g., by requesting repetition
and clarification or by offering praise).
Can use simple vocabulary and
language patterns
appropriately and without
errors that impede
communication.
Can sometimes self-correct
simple errors.
May suggest a level of
proficiency above 3 but has
provided too limited a
sample.
Can present some relevant ideas
sequentially with some links among
their own ideas and with those
presented by others.
Can respond to some simple questions
and may be able to expand these
responses when addressed
directly.
2 Volume may be a problem.
Can pronounce simple sounds/sound
clusters well enough to be
understood most of the time;
common words can usually be
understood within overall context.
Can produce familiar stretches of
language with sufficiently
appropriate pacing and intonation
to help listener’s understanding.
Can use appropriate body language
when especially interested in the group
discussion or when prompted to
respond.
Can use simple but heavily formulaic
expressions to respond to others (e.g., by
offering greetings or apologies).
Can appropriately use vocabulary
drawn from a limited and very
familiar range.
Can use some very basic
language patterns accurately
in brief exchanges.
Can identify some errors but
may be unable to self-correct.
Provides a limited language
sample
.
Can express some simple relevant
information and ideas, sometimes
successfully, and may expand some
responses briefly.
Can make some contribution to a
conversation when prompted.
1 Volume is likely to be a problem.
Can pronounce some simple sounds
and common words accurately
enough to be understood.
Can use appropriate intonation in the
most familiar of words and
phrases; hesitant speech makes the
listener’s task difficult.
Can use restricted features of body
language when required to respond to
peers.
Can use only simple and narrowly
restricted formulaic expressions, and
only to respond to others.
Can produce a narrow range of
simple vocabulary.
Can use a narrow range of
language patterns in very
short and rehearsed
utterances.
A restricted sample of language
makes full assessment of
proficiency difficult.
Can occasionally produce brief
information and ideas relevant to
the topic.
Can make some brief responses or
statements when prompted.
0 Does not produce any
comprehensible English speech.
Does not use any interactional strategies. Does not produce any recog-
nisable words or sequences.
Does not produce any appropriate,
relevant material.

64
TABLE B2
School-Based Assessment Criteria for Individual Presentation
1. Pronunciation & Delivery 2. Communication Strategies
3. Vocabulary & Language
Patterns 4. Ideas & Organisation
6 Can project the voice
appropriately for the context.
Can pronounce all sounds/sound
clusters and words clearly and
accurately.
Can speak fluently and naturally,
with very little hesitation, and
using intonation to enhance
communication.
Can use appropriate body language to
show focus on audience and to
engage interest.
Can judge timing to complete the
presentation.
Can confidently invite and respond to
questions or comments when
required for the task.
Can use a wide range of
accurate vocabulary.
Can use varied and highly
accurate language patterns;
minor slips do not impede
communication.
Can choose appropriate content
and level of language to
enable audience to follow,
without the use of notes.
Can self-correct effectively.
Can convey relevant information and
ideas clearly and fluently without
the use of notes.
Can elaborate in detail on some
appropriate aspects of the topic, and
can consistently link main points
with support and development.
5 Can project the voice
appropriately for the context.
Can pronounce all sounds/sound
clusters clearly and almost all
words accurately.
Can speak fluently with only
occasional hesitation, and
using intonation to enhance
communication, giving an
overall sense of natural
nonnative language.
Can use appropriate body language to
show focus on audience and to
engage interest.
Can judge timing sufficiently to cover
all essential points of the topic.
Can appropriately invite and respond
to questions or comments when
required for the task.
Can use varied and almost
always appropriate
vocabulary.
Can use almost entirely
accurate and appropriate
language patterns.
Can choose content and level of
language that the audience
can follow, with little or no
dependence on notes.
Can usually self-correct
effectively.
Can convey relevant information and
ideas clearly and well.
Can elaborate on some appropriate
aspects of the topic, and can link
main points with support and
development.

65
4 Can project the voice mostly
satisfactorily.
Can pronounce most sounds/
sound clusters and all common
words clearly and accurately;
less common words can be
understood although there
may be articulation errors
(e.g., dropping final consonant
clusters).
Can speak at a deliberate pace,
with some hesitation but using
sufficient intonation
conventions to convey meaning.
Can use appropriate body language to
display audience awareness and to
engage interest, but this is not
consistently demonstrated.
Can use the available time to
adequately cover all the most
essential points of the topic.
Can respond to any well-formulated
questions that arise.
Can use mostly appropriate
vocabulary.
Can use language patterns that
are usually accurate and
without errors that impede
communication.
Can choose mostly appropriate
content and level of language
to enable audience to follow,
using notes in a way that is
not intrusive.
Can self-correct when
concentrating carefully or
when asked to do so.
Can present relevant literal ideas
clearly and in well-organised
structure.
Can expand on some appropriate
aspects of the topic with additional
detail or explanation, and can
sometimes link these main points
and expansions together effectively.
3 Volume may be a problem.
Can pronounce all simple
sounds clearly but some errors
of sound clusters; less
common words may be
misunderstood unless
supported by contextual
meaning.
Can speak at a careful pace and
use sufficient basic intonation
conventions to be understood
by a familiar and supportive
listener; hesitation is present.
Can use some appropriate body
language, displaying occasional
audience awareness and providing
some degree of interest.
Can present basic relevant points but
has difficulty sustaining a
presentation mode.
Can respond to any cognitively
simple, well-formulated questions
that arise.
Can use simple vocabulary and
language patterns
appropriately and without
errors that impede
communication, but reliance
on memorised materials or
written notes makes language
and vocabulary use seem more
like written text spoken aloud.
Can choose a level of content and
language that enables audience
to follow a main point, but
needs to refer to notes.
Can sometimes self-correct
simple errors.
Can present some relevant literal ideas
clearly, and can sometimes provide
some simple supporting ideas.
Can sometimes link main and
supporting points together.
(
Continued)

66
TABLE B2
(Continued)
1. Pronunciation & Delivery 2. Communication Strategies
3. Vocabulary & Language
Patterns 4. Ideas & Organisation
2 Volume may be a problem.
Can pronounce simple sounds/
sound clusters well enough to
be understood most of the
time; common words can
usually be understood within
overall context.
Can produce familiar stretches
of language with sufficiently
appropriate pacing and
intonation to help the
listener’s understanding.
Can use a restricted range of features
of body language, but the overall
impression is stilted.
Can present very basic points but does
not demonstrate use of a
presentation mode and is dependent
on notes.
Audience awareness is very limited.
Can appropriately use
vocabulary drawn from a
limited and very familiar
range.
Can read notes aloud but with
difficulty.
Can use some very basic
language patterns accurately
in brief exchanges.
Can identify some errors but
may be unable to self-correct.
Can make an attempt to express
simple relevant information and
ideas, sometimes successfully, and
can attempt to expand on a few
points.
Can link the key information
sequentially.
1 Volume is likely to be a
problem.
Can pronounce some simple
sounds and common words
accurately enough to be
understood.
Can use appropriate intonation
in the most familiar of words
and phrases; hesitant speech
makes the listener’s task
difficult.
Body language may be intermittently
present, but communication
strategies appropriate to delivering a
presentation are absent. The
delivery is wholly dependent on
notes or a written text. There is no
evident audience awareness.
Can produce a narrow range of
simple vocabulary.
Can use a narrow range of
language patterns in very
short and rehearsed
utterances.
A restricted sample of
language makes full
assessment of proficiency
difficult.
Can express a main point or make a
brief statement when prompted, in a
way that is partially understandable.
0 Does not produce any
comprehensible English
speech.
Does not attempt a presentation. Does not produce any
recognisable words or
sequences.
Does not express any relevant or
understandable information.

VIEWS FROM THE CHALKFACE 67
APPENDIX C
Assessment Record
TABLE C1
Hong Kong Certificate of Education Examination English Language School-Based
Assessment Component: Assessment Record (Group Interaction)
School Name:
Teacher Name:
Oral Text-type: Group Interaction Assessment date: ___/___/___
Name of text: _______________________________ Category: Print / N-Print
(circle)
Fiction / N-Fic
(circle)
Class: Summary of task:
ADVICE TO TEACHERS
This assessment sheet will assist teachers to allocate marks. There are two stages to this process. The first stage is to make judgments on the student’s performance in
each domain (i.e. pronunciation and delivery, communication strategies, vocabulary and language patterns, and ideas and organisation) with reference to the
Assessment Criteria. You should circle one of the numbers 1- 6 (or 0 if no language was produced) to indicate how well the stud ent performed in each domain. The
second stage is to add up the marks for all domains. The total number of possible marks is 24. Add a comment if possible.
Student 1: No.: Student 2: No.:
CRITERIA FOR THE AWARD OF MARKS
(Circle number for each domain)
1. Pronunciation & delivery
0 1 2 3 4 5 6
2. Communication strategies
0 1 2 3 4 5 6
3. Vocabulary & language patterns
0 1 2 3 4 5 6
4. Ideas & organisation
0 1 2 3 4 5 6
TOTAL: _____ / 24
TEACHER’S
COMMENTS
CRITERIA FOR THE AWARD OF MARKS
(Circle number for each domain)
1. Pronunciation & delivery
0 1 2 3 4 5 6
2. Communication strategies
0 1 2 3 4 5 6
3. Vocabulary & language patterns
0 1 2 3 4 5 6
4. Ideas & organisation
0 1 2 3 4 5 6
TOTAL: _____ / 24
TEACHER’S
COMMENTS
Student 3: No.: Student 4: No.:
CRITERIA FOR THE AWARD OF MARKS
(Circle number for each domain)
1. Pronunciation & delivery
0 1 2 3 4 5 6
2. Communication strategies
0 1 2 3 4 5 6
3. Vocabulary & language patterns
0 1 2 3 4 5 6
4. Ideas & organisation
0 1 2 3 4 5 6
TOTAL: _____ / 24
TEACHER’S
COMMENTS
CRITERIA FOR THE AWARD OF MARKS
(Circle number for each domain)
1. Pronunciation & delivery
0 1 2 3 4 5 6
2. Communication strategies
0 1 2 3 4 5 6
3. Vocabulary & language patterns
0 1 2 3 4 5 6
4. Ideas & organisation
0 1 2 3 4 5 6
TOTAL: _____ / 24
TEACHER’S
COMMENTS
AUTHENTICATION
1. I certify that each student has read/viewed the text above used in this oral assessment, that the text is not a class reader, comic, newspaper, or a set text for other
subjects, and that the work is all the student’s own.
2. I certify that the assessment was undertaken under the conditions specified in the HKEAA guidelines, that I am the students’ English teacher, that I conducted
the assessment and that the task has not been repeated.
Teacher Student 1 Student 2 Student 3 Student 4
Signature
Date

68 DAVISON
TABLE C2
Hong Kong Certificate of Education Examination English Language School-Based
Assessment Component: Assessment Record (Individual Presentation)
School Name:
Teacher Name:
Oral Text-type: Individual Presentation
Assessment date: ___/___/___
Class:
Name of text: _______________________________
Category: Print / N-Print (circle)
Fiction / N-Fic (circle)
Student Name: Summary of task:
Student No.:
ADVICE TO TEACHERS
This assessment sheet will assist teachers to allocate marks. There are two stages to this process. The first stage is to make judgments
on the student’s performance in each domain (i.e. pronunciation and delivery, communication strategies, vocabulary and language
patterns, and ideas and organisation) with reference to the Assessment Criteria. You should circle one of the numbers 1-6 (or 0 if no
language was produced) to indicate how well the student performed in each domain. The second stage is to add up the marks for all
domains. The total number of possible marks is 24. Add a comment if possible.
CRITERIA FOR THE AWARD OF MARKS
(Circle number for each domain)
1. Pronunciation & delivery
0 1 2 3 4 5 6
2. Communication strategies
0 1 2 3 4 5 6
3. Vocabulary & language patterns
0 1 2 3 4 5 6
4. Ideas & organisation
0 1 2 3 4 5 6
TOTAL: _____ /
24
TEACHER’S COMMENTS
Comments on aspects of the student’s work that led to your
assessment and any contextual factors (e.g. amount of rehearsal
or teacher support) that need to be taken into account.
AUTHENTICATION
1. I certify that this student has read/viewed the text above used in this oral assessment, that the text is not a class reader, comic,
newspaper, or a set text for other subjects, and that the work is all the student’s own.
2. I certify that the assessment was undertaken under the conditions specified in the HKEAA guidelines, that I am the student’s
English teacher, that I conducted the assessment and that the task has not been repeated.
Teacher Student
Signature
Date