WP28/2010

Signalling performance: Continuous assessment and matriculation examination marks in South African schools

Stellenbosch Working Paper Series No. WP28/2010

Author(s): Servaas van der Berg, Debra Shepherd

JEL Classification: I21, D82

Keywords: Economics of Education, assessment, asymmetric information, South Africa

Download: PDF (897 KB)


Abstract:

Economists regard information and feedback as important ways for self-correction in a system. This study analyses one aspect of information and feedback in the South African education system. Continuous assessment (CASS) carries a 25% weight in the final matriculation (Grade 12) mark and, more importantly, provides feedback on performance that affects examination preparation and effort. Weak assessment in schools means that pupils are getting wrong signals that may have important consequences for the way they approach the final examination. Moreover, similarly wrong signals earlier in their school careers may also have affected their subject choice and career planning. This study analyses data on CASS and compares it to the externally assessed matric exam marks for three years for a number of subjects. There are two signalling dimensions to inaccurate assessments: (i) Inflated CASS marks give students a false sense of security that they are well-prepared for the matric exams, thereby leading to unrealistic expectations and diminished effort. (ii) A weak correlation between CASS and the exam marks means poor signalling in another dimension: Relatively good students may get relatively low CASS marks. This indicates poor reliability of assessment, as the examination and continuous assessment should both be testing the same mastery of the national curriculum. The paper analyses the extent of each of these two dimensions of weak signalling in South African schools, by subject, province, socio-economic background of schools, and public versus independent schools. The analysis draws disturbing conclusions for a large part of the school system.