Thwarted by an inappropriate disability indicator: the case of the new student-level electronic EMIS in South Africa

• In Event: Disability inclusion in Education Management Information Systems

Proposal

Background: From 2008 onwards, a new process of identification and assessment of student disability (aligned with the biopsychosocial model of disability) was introduced in schools in South Africa. A powerful new student-level Education Management Information System (EMIS) was launched in 2016. A student-level electronic EMIS allows for more consistent, accurate data on student disability status. By 2023, this EMIS was used to track grade progression of students with and without disabilities over seven years. Such a system could be used to determine what proportion of students with disabilities are receiving reasonable accommodations in school.
However, this presentation will describe that the disability indicators in EMIS have not kept pace with reforms in disability identification. The implications will be described, using national-level data from 1) the annual school census 2011 to 2014, 2) student-level EMIS data for all students in the Eastern Cape in 2018 and, 3) student-level EMIS data for the Gauteng province from 2017 to 2023.
Methods: Categories of disability in EMIS were described and compared with those in the Screening, Identification, Assessment and Support strategy (2008) and policy (2014) and the post-provisioning norms (1998). Total enrolment of students with disabilities in 2011 to 2014 (collected in annual school censuses) was compared with that in the new student-level EMIS in 2018 in the Eastern Cape province. Disability prevalence in schools was estimated and compared with disability prevalence rates from household surveys (which uses the Washington Group Short Set). Statistical techniques were applied to all three sets of data to test whether schools in wealthier areas are more likely to report enrolment of students with disabilities.
Results: In the Eastern Cape province, a much higher number of schools reported the enrolment of students with disabilities in the student-level EMIS than in annual school surveys. We attribute more widespread reporting to the less cumbersome reporting process in the new EMIS. Disability prevalence (using the school disability indicator) is much lower than disability prevalence among children enrolled in school (using the Washington Group Short Set in the national census or household surveys). Schools in wealthier areas are shown to be much more likely to report the presence of at least one student with a disability. The co-existence of two different sets of disability indicators in the education system has led to measurement error as teachers must match up the categories in the screening forms with a quite different list of disability categories in EMIS.
The South African case study powerfully demonstrates how the potential benefits of a student-level EMIS can be thwarted by incoherent and narrowly defined indicators of disability status. In societies with unequal access to health care, using a diagnosis-dependent disability indicator will result in low reporting of disability enrolment in schools in poorer areas. When reforming disability indicators, it is critical that reforms are carried through timeously in all parts of the education system: from EMIS to inclusive education directorates to those involved in planning and resourcing.

Author

• Nicola Deghaye, Stellenbosch University