****************************************

    The following is a question by the Hon Ma Lik and a written reply by the Secretary for Education and Manpower, Professor Arthur K C Li, in the Legislative Council today (March 16):

Question:

    Since the implementation of the short-term Secondary School Places Allocation ("SSPA") in the 2000/01 school year, students entering secondary schools had their school internal assessment ("IA") results scaled by the Academic Aptitude Test ("AAT") results achieved by the students of their respective primary schools in the 1997/98 to 1999/2000 school years. The scaled scores of all the students in the same school net were ranked and students were subsequently divided into three allocation bands. Band One students were allocated a place first, followed by Band Two and then Band Three students. In February this year, the Working Group on Review of Secondary School Places Allocation and Medium of Instruction for Secondary Schools under the Education Commission published a consultation document, in which a new SSPA was proposed. In this connection, will the Government inform this Council:

(a) of the respective percentages of the ten primary schools which had the lowest percentage of Band One students in their primary six students and the numbers of Band One students in the ten primary schools with the smallest number of such students, in each of the past three years;

(b) whether the existing SSPA can reflect

(i) the efforts of those primary schools whose overall academic results have been on the rise over the past few years; and

(ii) the academic performance of students with extremely high learning abilities;

(c) whether other territories have adopted a secondary school places allocation system similar to the ones currently used in Hong Kong or proposed in the consultation document, including whether such territories have ever used the results of past students in the same school to scale the IA results of the students; if they have, of the details of the systems; and

(d) given that the consultation document proposes to use the Pre-Secondary 1 Hong Kong Attainment Test, in place of the AAT results of past students in the same school, as an instrument for scaling school IA results, of the merits of this proposal?

Reply:

Madam President,

(a) Under the existing Secondary School Places Allocation (SSPA) mechanism, students' allocation bands are determined on a net basis, viz. all participating students in a school net are put into a single order of merit according to their scaled scores, and then divided into three equal allocation bands, with each consisting of one-third of the total number of students in the school net. The banding of students in a school net is for the purpose of determining the order of allocation of Secondary 1 (S1) places. Since the allocation bands are based on the relative performance of the students in a particular school net, they cannot accurately reflect the actual performance of the students nor could they be appropriately used for comparing students' performance on a territory basis. Besides, as the number of students differs from one school net to another, the number of students in each allocation band also varies among school nets. It is therefore not meaningful to make comparison among schools in terms of their number and ratio of students in a particular allocation band.

(b) (i) Under the present SSPA mechanism, primary schools' results in the last three Academic Aptitude Tests (AAT) are used to scale their students' internal assessment (IA) results. Although the AAT was abolished in 2000, the accuracy of using the past AAT results for banding purpose, as demonstrated by research, is still high at about 85%. However, it is acknowledged that the reliability of the past AAT results would diminish over time. Many primary schools and parents consider this as unfair. Progressing and newly established schools feel that their efforts and improvement made have not been duly recognised. Besides, many secondary schools opine that the continual use of the present scaling mechanism would further widen the within-school student diversity, which would increase their difficulties in learning and teaching.

(ii) Since the allocation bands are determined on the basis of IA results of the participating Primary Six students, those with higher learning abilities would have better performance in their IAs, and hence a better chance of being included in a higher allocation band.

(c) As regards the arrangement for students' progression from primary to secondary schools, different countries/regions adopt different modes with regard to their local context and the development of their education systems. While some primarily adopt the principle of vicinity (such as China, South Korea and New Zealand), some (such as Singapore) stream/place students on the basis of highly selective public examination(s). In Hong Kong, our present SSPA mechanism has evolved over several decades, taking into account educational considerations, historical factors and the latest development in education in Hong Kong. The SSPA system currently used in Hong Kong, as well as the changes proposed in the consultation document, which include the use of the pre-S1 Hong Kong Attainment Test (HKAT) results of past students in the same school to scale the IA results of the students, is to address the need for a low-stake scaling tool. As far as we know, no other country or region adopts a system congruent to our SSPA mechanism.

(d) In the consultation document on Review of Medium of Instruction for Secondary Schools and Secondary School Places Allocation, the Working Group (WG) puts forth two options for scaling students' IA results, viz. continual use of the past AAT results, or replacing the AAT results with the pre-S1 HKAT. The WG considers the latter a better choice. The drawbacks of the continual use of the past AAT results for scaling have been mentioned at paragraph (b)(i) above. To use the pre-S1 HKAT as a scaling tool can more accurately reflect the overall performance between schools. The only concern is the possibility of inducing drilling. To reduce the incentive to drilling, the WG proposes that, if this option is accepted, the pre-S1 HKAT results should be sampled biennially and the average of two most recently sampled results would be used to scale the IAs of the coming cohort of P6 students proceeding to S1. Under such a scaling mechanism, students' IA results will continue to form the basis in determining their allocation bands. Since the pre-S1 HKAT results of the students taking the test will not have any direct bearing on their allocation results, incentive to drill would be reduced. As the pre-S1 HKAT is curriculum-based, distortion of the primary school curriculum is therefore less likely. Even if some primary schools would, due to the scaling purpose of the pre-S1 HKAT, place greater emphasis on students' performance in Chinese, English and Mathematics, meaningless drilling can still be reduced. As a matter of fact, the pre-S1 HKAT is an existing assessment tool which secondary schools make reference to in streaming, and planning enhancement/support programmes for S1 new entrants. Students need not sit for an "extra" test, even if it is adopted as a scaling tool.

Ends/Wednesday, March 16, 2005

NNNN