The main problem examined in this research is the lack of explicit criteria by which standards for examinations can be set. An attempt was therefore made to formulate educationally accountable criteria in order to set uniform standards for examinations, with special reference to mathematics. The research first established a comprehensive perspective of the role of examinations within an educational context. This was done by a detailed literature study and a study of the existing ways by which standards are determined. A critical analysis of the types of examinations, as used worldwide, was made. The use of item banks and the role played by the Joint Matriculation Board in the Republic of South Africa, received special attention. There are two dimensions implicit in determining standards for examinations. The horizontal dimension deals with equality - equal opportunity, equal treatment and equal evaluation. The vertical dimension covers the discriminatory aspects of examinations. Both dimensions must be taken into account if uniform, and proper, standards are sought for any formal educational system as a whole. The main conclusion of the study is that reliable and valid criteria are only possible if the curriculum cycle as a whole is taken into account. This implies curricular and instructional validity - a curricular match between the examination and the components constituting the curriculum cycle. Criteria to establi.sh uniform standards for examinations in mathematics and in other subjects, may be subdivided into two sets. These sets are: (i) criteria of curricular relevancy (ii) criteria of a technical nature The criteria of curricular relevancy cover the components of the curriculum cycle. These are identified as * situational relevancy * objectives relevancy * content relevancy and * instructional relevancy Minimum competency skills and objectives should be clearly and specifically defined so that teachers will know what to teach, students will know what to study and examiners will know what to test in an examination. The nature of the educational end product will then be of a more equal standard (Pullin, 1983). The technical criteria are necessary to ascertain validity in the examination as a whole. These embrace controlling mechanisms to ensure the validity, objectivity and reliability of the results. The following aspects were identified as the most important: * the curricular match * formulating of questions * marking * sundry parameters covering the length of the paper, grading of questions, uniformity in allocation of marks, etc. * comparability mechanisms * administrative aspects to ensure security and accuracy * uniformity in determining the final result of ~he examination / Proefskrif (D.Ed.) - Randse Afrikaanse Universiteit, 1986

The main problem examined in this research is the lack of explicit criteria by which standards for examinations can be set. An attempt was therefore made to formulate educationally accountable criteria in order to set uniform standards for examinations, with special reference to mathematics. The research first established a comprehensive perspective of the role of examinations within an educational context. This was done by a detailed literature study and a study of the existing ways by which standards are determined. A critical analysis of the types of examinations, as used worldwide, was made. The use of item banks and the role played by the Joint Matriculation Board in the Republic of South Africa, received special attention. There are two dimensions implicit in determining standards for examinations. The horizontal dimension deals with equality - equal opportunity, equal treatment and equal evaluation. The vertical dimension covers the discriminatory aspects of examinations. Both dimensions must be taken into account if uniform, and proper, standards are sought for any formal educational system as a whole. The main conclusion of the study is that reliable and valid criteria are only possible if the curriculum cycle as a whole is taken into account. This implies curricular and instructional validity - a curricular match between the examination and the components constituting the curriculum cycle. Criteria to establi.sh uniform standards for examinations in mathematics and in other subjects, may be subdivided into two sets. These sets are: (i) criteria of curricular relevancy (ii) criteria of a technical nature The criteria of curricular relevancy cover the components of the curriculum cycle. These are identified as * situational relevancy * objectives relevancy * content relevancy and * instructional relevancy Minimum competency skills and objectives should be clearly and specifically defined so that teachers will know what to teach, students will know what to study and examiners will know what to test in an examination. The nature of the educational end product will then be of a more equal standard (Pullin, 1983). The technical criteria are necessary to ascertain validity in the examination as a whole. These embrace controlling mechanisms to ensure the validity, objectivity and reliability of the results. The following aspects were identified as the most important: * the curricular match * formulating of questions * marking * sundry parameters covering the length of the paper, grading of questions, uniformity in allocation of marks, etc. * comparability mechanisms * administrative aspects to ensure security and accuracy * uniformity in determining the final result of ~he examination / Proefskrif (D.Ed.) - Randse Afrikaanse Universiteit, 1986