Please use this identifier to cite or link to this item: https://hdl.handle.net/10321/3737
Title: Predicting the success of TVET learners in a higher education engineering programme
Authors: Stops, Rodney Alan 
Issue Date: 1-Dec-2021
Abstract: 
South Africa has experienced radical political and social change since 1994. Recognised for their
role in the transformation process, universities have been and remain at the forefront of this
change. While learners recognise that higher education is paramount to changing their socioeconomic condition, the massification of education along with new and advanced curricula has
presented ongoing challenges for both universities and learners. Coming from diverse
backgrounds and dealing with a variety of academic choices, learners encounter many challenges
to entering Higher Education (HE). University programmes accept learners with varying
competency levels and needs but are required to ensure that graduates meet standards that are
acceptable to both industry and the academia. In the quest for universities to improve the delivery
of educationally sound and industrially relevant programmes, ongoing research is being
conducted and new and innovative ways have had to be developed to solve the problems
associated with larger numbers of underprepared learners. An emerging method being employed
in HE is the use of Data Analytics and Education Data Mining (EDM) techniques to derive
solutions to assist institutions in maximising retention, and through-put rates.
Durban University of Technology (DUT) has, since 1994, accepted learners into the Report 151
National Diploma from Technical Vocational Education and Training (TVET) Colleges. These
learners, in respect to the Articulation Policy for the Post-School Education and Training (PSET)
system of South Africa in terms of Section 8(2)(b) of the NQF Act, 2008 (Act 67 of 2008), are
among those referred to as articulating learners. The perception among DUT staff involved with
the teaching of these learners, is that they are as able to cope with the complexity and quality of
engineering programmes as those learners entering the institution directly after completing their
school leaving Senior Certificate/National Senior Certificate (SC/NSC).
As no previous formal tracking, analysis or research has been conducted to determine the
success or failure of learners articulating from TVET Colleges into DUT in general or into DUT’s Department of Electrical Power Engineering specifically, this research utilised Educational Data
Mining and Inferential Statistics on an engineering learner dataset, to determine hidden patterns
and relationships. Specifically, those relationships that promote progression, correlation, and
selection were investigated. The Waikato Environment for Knowledge Analysis (WEKA) was
employed to do the EDM, and a tool was developed to assist with the selection of learners entering
the department, as well as ranking those entrance requirements that affect course content and
throughput and various statistical methods were employed to conduct a retrospective longitudinal
study. The Knowledge Discovery in Databases Process is used to work with 5 years of learner
data.
Both from the perspective of the progression of learners from semester 1 to semester 2 and the
cohort throughput analysis, the results showed that there were no statistically significant
differences in the performance of learners articulating from TVET colleges into the National
Diploma: Electrical Engineering at DUT and their counterparts from high school admitted directly
in the said qualification. The findings are that learners from a TVET College articulating into an
HEI qualification, specifically the National Diploma: Electrical Engineering at DUT, complete the
course in similar rates and in similar proportions to those learners admitted directly from school.
The statistical analysis indicates that 77.6% of TVET N4 learners are promoted to semester 2,
compared to 70.0% of SC/NSC learners and the EDM prediction tool developed for TVET N4
learners, the IBK classification tool resulted in a 77.61% accuracy, while the ANN classification
tool returned an accuracy of 77.56% for the SC/NSC learners.
Description: 
Submitted in fulfilment of the requirements for the degree of Master of Engineering in the Faculty of Engineering, and the Built Environment at Durban University of Technology, 2021.
URI: https://hdl.handle.net/10321/3737
DOI: https://doi.org/10.51415/10321/3737
Appears in Collections:Theses and dissertations (Engineering and Built Environment)

Files in This Item:
File Description SizeFormat
StopsRA_2021.pdfThesis4.11 MBAdobe PDFView/Open
Show full item record

Page view(s)

312
checked on Jul 16, 2024

Download(s)

709
checked on Jul 16, 2024

Google ScholarTM

Check

Altmetric

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.