Establishing Writing Benchmarks for Primary and Secondary English Language Teachers in Hong Kong.

Concerns relating to the English language proficiency of teachers in Hong Kong, expressed in Education Commission Report Number 6 (ECR6, 1995), provided the impetus for what has become commonly known as the Benchmarks Test. Within the context of the overall development of a range of assessment instruments for setting standards (language benchmarks) for primary and secondary teachers in Hong Kong, this paper first describes the origins of the project. It then goes on to look at the four and a half year process of piloting, moderating, ratifying and implementing the test of writing skills for English language teachers. Introduction and background In December 1995, the Education Commission, the major educational policy making body for Hong Kong, published Report Number 6 (ECR6). The Education Commission revealed that it was fully aware of the concerns expressed to Government by companies and organisations in business and commerce. It took note of educational bodies and institutions that claimed that language skills were declining. The Education Commission also acknowledged the need for high level language skills among the workforce in Hong Kong, as it moves away almost entirely from a manufacturing to a service, e-commerce and high-tech industries base, and highlighted a number of areas for action. On the issue of standards of language ability of teachers, about which there were concerns (see e.g., Tsui, 1993; Tsui et al., 1994), two recommendations were made by the Education Commission and passed to the Advisory Committee on Teacher Education and Qualifications (ACTEQ) for action. It tasked ACTEQ with investigating, establishing, and ultimately, implementing benchmarks on two fronts. The first concerned language teachers, that is, teachers of English, Chinese and Putonghua. The second concerned teachers who teach content subjects (history, geography, biology, 1 Editors Note: The Benchmarks Test is now known as the Language Proficiency Assessment of Teachers (LPAT) Establishing writing benchmarks for primary and secondary English language teachers | 129 mathematics etc.) through the mediums of either English or Chinese (see Falvey and Coniam, 1997). The term benchmark is relatively new in educational and language environments although it is now being used increasingly in educational contexts. The term which literally means “a mark on a bench” was used initially for standard setting for physical and material measurements e.g. the length, width and breadth of steel, or the size and strength of screws. Benchmarked items came to represent standards in industry and commerce and, eventually, an International Standards Office was set up to determine and certify benchmarks worldwide. Tucker (1996) has created a useful definition – “Benchmarking is the search for best practices that lead to superior performance” (1996: ix). Concerns about standards have led to the use of the term „benchmarks‟, often with a different connotation to Tucker‟s „best practice‟ definition. The contrary view would define a benchmark as „the minimum acceptable standard‟ rather than Tucker‟s „best practice‟ perspective. However, whether one takes the optimum or minimum definition of benchmarks, the issue of standards in teacher education and certification is certainly not new. In the context of standards for teachers, they normally focus on the concept of „minimum standards‟. As an example, in the US, concerns about standards in education and teaching resulted in the commissioning of the report of the Holmes Group (1986). The concerns were fuelled by the publication of incidents of teacher incompetence. Soled (1995) cites a famous example taken from Time magazine in 1980 where a note to a parent read as follows: Scott is dropping in his studies he acts as if he don’t Care. Scott want to pass in his assignment at all, he had a poem to learn and fell to do it. Such examples are extremely damaging to the profession as a whole and are clearly unacceptable at any level of evaluation and reporting. The report of the Holmes Group (1986) considered the concerns of parents and professionals in education and was instrumental in bringing in teacher assessment, stating that one of its five major goals was to create professionally relevant and intellectually defensible standards for entry into the teaching profession. The US system of teacher assessment includes basic competencies plus pedagogic knowledge. In any country, because of the very nature of P. Falvey & D. Coniam | 130 professional evaluation, teacher assessment remains a high-stakes, sensitive issue. Recent developments in education and teaching call for more accountability and the demonstration of professional competence. This has emerged as a result of increased pressure from parents and professional groups who are dissatisfied with the products of the education system. Soled (1995) notes that, in a survey of public attitudes of the general public in the US, 85% felt that teachers should be required to pass competency tests. Similar attitudes were found by a Hong Kong political party, the Democratic Alliance for the Betterment of Hong Kong, in October 2000 (South China Morning Post 2000) when they polled the public about the need to create language standards for language teachers. Further concern in the US arose in the mid 1980s when the lack of language competence of many International Teaching Assistants (ITAs) was exposed (see Chism and Warner, 1987) for a discussion at the first national conference on ITAs where issues about testing ITAs were debated). The ITAs were used on a regular basis to teach freshmen in US tertiary institutions, even though many had an inadequate background in English for study or teaching purposes, coming from countries where English was a second or foreign language. As a result of that concern, measures were taken to establish standards for the ITAs. Those standards must be reached before ITAs are allowed to tutor or teach US undergraduates: language competency tests (T.E.A.C.H.) are offered five times per year for this purpose. Work has also been carried out on the language standards of language teachers in a number of countries, usually in response to complaints about teacher educational or language competence. Standard setting for teachers has occurred in Guam (Stansfield et al., 1990) and for language instructors in Canada (Sanaoui, 1997, 1998). In Australia, Elder (1994) has discussed the creation of benchmarks for teachers of languages other than English (L.O.T.E.). Returning to the US, in 1998, 60% of 2,000 prospective teachers in the State of Massachusetts failed a pen and paper test of basic literacy and numeracy skills. Showing the constant concern that the public world-wide has about standards the then State Governor, Governor Cellucci stated in a board meeting held 2 nd July 1998, “Education reform is not without conflict and controversy, but we must send a clear message that we are going to hold the line for high standards” (Massachusetts Department of Education, 1998). Even more recently, reports from the US during the race for the Year 2000 Presidency cite both Vice-President Al Gore and presidential candidate Establishing writing benchmarks for primary and secondary English language teachers | 131 George W. Bush talking about educational standards on the same day as they traveled through different states (The Independent, UK, 5 June 2000). It is interesting that both candidates stressed the need for teacher competence and both stated that federal funds would be withdrawn from states which do not insist on and maintain acceptable standards for their teachers. The two Presidential candidates reiterated their views on teacher standards in the first of three country-wide presidential debates on television in mid-October 2000 and Bush maintained his theme of the need to insist on teacher standards right to the end of his campaign. Within the Hong Kong context, ACTEQ commissioned a consultancy study in early 1996 to investigate the feasibility of establishing language benchmarks for lower secondary teachers of English. The consultancy report (Coniam and Falvey, 1996) was accepted by ACTEQ and a comprehensively representative English Language Benchmark Subject Committee (ELBSC) was subsequently established in late 1997 with members drawn from ACTEQ itself, the Hong Kong Examinations Authority (HKEA), principals, department heads, practising teachers and tertiary language teacher educators. The ELBSC worked together or in sub-committees over the next three years agreeing assessment constructs, establishing specifications, creating exemplar tasks, creating scales and descriptors for criterion-referenced task assessment and monitoring the piloting and moderation of the assessment instruments. It was agreed by the ELBSC and accepted by ACTEQ that the assessment battery should consist of a battery of „formal‟ pen and paper tests (i.e., Speaking, Writing, Listening, and Reading and Language Systems), and a performance test of Classroom Language, where teachers would be assessed teaching a live lesson. The culmination of the work of the ELBSC was to be the Pilot Benchmark Assessment (English), or PBAE. The PBAE was a study which comprised as representative as possible a sample of the cohort of lower secondary English language teachers in Hong Kong. The PBAE was intended as a test bed of the prototype test specifications and assessment instruments. It was envisaged that the PBAE would indicate how the minimum standard level – the „benchmark‟ level – matched the perceptions of the ELBSC. Test takers would also participate in surveys to gauge their reactions to the test battery in terms of how they perceived issues such as test validity, test relevance, and test difficulty. As stated in Falvey and Coniam (1997), benchmarking – including language benchmarking – will, by its nature, directly affect the lives and P. Falvey & D. Coniam | 132 careers of thousands of people