
Publications in high school systems
Peer-reviewed publications are available from the publications sites provided.
2024
The South African Journal of Industrial Engineering, 35(3), pp. 15–27
The South African high school education system faces numerous challenges, including high dropout rates and unequal educational outcomes, calling for innovative methods to analyse and address these problems. This study employs an integrated approach that merges machine learning and agent-based modelling to simulate learner progression in public high schools, illuminating the critical factors that influence educational outcomes. Using data from the 2019 General Household Survey in South Africa, factor analysis is first conducted to identify and quantify the principal characteristics defining learners. These features then train an XGBoost machine learning model, which is integrated within an agent-based framework to simulate learner progression from Grades 8 to Grade 12. Validating the model against the Learner Unit Record Information and Tracking System dataset resulted in a root square error of 2.95%, which is indicative of the model's ability to predict learner progression. Overall, the model represents a significant advancement in the field of educational simulation, serving as a practical tool for schools to analyse and improve learner outcomes through analytical decision-making.
Working papers in high school systems
The working papers, research projects, theses and dissertations listed are being prepared for publication.
2024, work in progress
South Africa has one of the highest measures of economic inequality and one of the worst performing basic education systems in the world. For every 100 learners who enrol for Grade 1, 50 will be lost from the system before Grade 12 (as learners mostly exit in Grades 10 and 11), 40 will matriculate from Grade 12 and 12 will qualify to enrol into university. Only 4 will eventually graduate with a bachelor’s degree. Of 39 countries surveyed in 2021 by OECD, South Africa had the highest percentage (i.e., 46.1%) of young adults who were neither employed nor enrolled in formal education or training. In 2021 the DHET estimated that less than a third of South African adults (i.e., 32.5%) held NSCs. In light of South Africa’s sordid past the country’s education system has experienced many challenges and the wealthiest 20% of learners and the poorest 80% of learners are still marked by considerable and statistically significant differences in academic performance. This bimodality evident in high school outcomes makes it impossible to define the “average” learner.
​
In this study, simulation modelling will be used to analyse and compare the academic achievement of learners with disparate sets of characteristics. These learners are enrolled into a single education system founded on the assumption that broad policies for the heterogenous average learner could adequately support each individual. By means of specifically agent-based and system dynamics simulation modelling this study aims to provide insight as to how one overarching system supports or fails different classifications of learners. It will provide recommendations for ways in which policies may be customised to serve a learner body as a granular collection instead of a homogenous whole. Industrial engineering principles and tools will be used to formulate the engineering solution which will then validated by means of a case study.
2024
This study aims to refine and validate a causal mapping model for learner progression in the South African public high school system, with a particular focus on the Western Cape and Gauteng provinces. The research addresses the question of which factors influence the successful completion of Grade 12, particularly given the high learner attrition and the dysfunctional state of South Africa's basic education system. Using data from the 2019 General Household Survey, this research employs advanced quantitative techniques including factor analysis, multilevel regression, and structural equation modelling. These methods are used to uncover key factors such as socioeconomic status, health, household conditions, and educational characteristics that significantly affect learner outcomes. The study not only identifies critical factors but also establishes the relational links between them, quantifying their influence on learner progression.
​
Results indicate that household socioeconomic status and educational resources are the strongest predictors of learner success, accounting for over 35% of the variance in Grade 12 completion rates. Additionally, health-related variables and access to social security benefits were shown to have a moderate yet significant influence on learner retention, contributing to a 15% reduction in dropout rates when adequately supported. The findings contribute to a comprehensive understanding of the South African high school system by highlighting the complex interplay between personal, household, and educational factors. This refined causal model offers a validated framework for future policy interventions aimed at improving learner outcomes and reducing high school dropout rates in South Africa.
The impact of pandemic control on the High School Model
2022, work in progress
A systems perspective on education systems design and policy: The High School Model
2020
The South African education system is in crisis. On average, South Africa has the worst education system of all middle-income countries and perform worse than many lower-income African countries. The South African Department of Basic Education released an action plan with 27 goals towards the realization of a functional system by 2030. Two of these goals are considered priority: Improve the average performance of Grade 9 learners in mathematics and ensure that all children remain effectively enrolled in school at least up to the year in which they turn 15. This project aims to build a system dynamics simulation model with which to analyze the impact of this goal on the Western Cape high school system. In so doing we may discover the feedback loops and possible unintended consequences of the goal that intuitively appears helpful but may prove to be unhelpful.