Martin Bekker
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Martin Bekker has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Carin Runciman
University of Edinburgh
4 shared publications - Benjamin James Roberts
Human Sciences Research Council (HSRC)
2 shared publications - Narnia Bohler-Muller
Human Sciences Research Council
2 shared publications - Stephen Rule
Outsourced Insight cc
2 shared publications - Estelle Trengove
University of the Witwatersrand
1 shared publication Hraklis Papageorgiou
University of the Witwatersrand
1 shared publication
Latest Publications
- 18 July 2025
- Published by Springer Nature
- p. 224-240
Data Overdose? Time for a Quadruple Shot: Knowledge Graph Construction Using Enhanced Triple Extraction
- 18 July 2025
- Published by Springer Nature
- p. 224-240
Data Overdose? Time for a Quadruple Shot: Knowledge Graph Construction Using Enhanced Triple Extraction
- Open Access
- 29 August 2024
- Published by Academy of Science of South Africa in South African Crime Quarterly
- No. 73,p. 2-11 to 2-24
Protest events have a 'Twitter Signature': Evidence from South Africa's #FeesMustFall
Protest levels in South Africa remain notoriously high. However, Protest Event Analysis (PEA) within the South African context has predominantly relied on media or administrative data, with scant regard for the insights that social media records may provide. Using data gathered during the #FeesMustFall movement, we employ a machine learning model trained on historical South African Twitter data combined with event records from a protest database. Our findings indicate that such events establish a distinctive time-series pattern, suggesting a robust approach to modelling, identifying or predicting protests within South Africa. Moreover, this implies that protests can be characterised using social media sources in ways that can enhance the insights from traditional data sources, and that automated analysis can unearth and assemble PEA databases. Moreover, automated PEA may soon provide valuable information about the actors, level of turmoil, size, duration, grievances, and motivations of protest.
- 29 August 2024
- Published by Academy of Science of South Africa in South African Crime Quarterly
- No. 73,p. 2-11 to 2-24
Open AccessProtest events have a 'Twitter Signature': Evidence from South Africa's #FeesMustFall
Protest levels in South Africa remain notoriously high. However, Protest Event Analysis (PEA) within the South African context has predominantly relied on media or administrative data, with scant regard for the insights that social media records may provide. Using data gathered during the #FeesMustFall movement, we employ a machine learning model trained on historical South African Twitter data combined with event records from a protest database. Our findings indicate that such events establish a distinctive time-series pattern, suggesting a robust approach to modelling, identifying or predicting protests within South Africa. Moreover, this implies that protests can be characterised using social media sources in ways that can enhance the insights from traditional data sources, and that automated analysis can unearth and assemble PEA databases. Moreover, automated PEA may soon provide valuable information about the actors, level of turmoil, size, duration, grievances, and motivations of protest.
- Open Access
- 24 April 2024
- Published by Taylor & Francis in South African Geographical Journal
- Vol. 107 (1) , 1-20
Making ends meet: spatial and socio-economic variations during the 2020-2021 COVID-19 lockdown in South Africa
- 24 April 2024
- Published by Taylor & Francis in South African Geographical Journal
- Vol. 107 (1) , 1-20
Open AccessMaking ends meet: spatial and socio-economic variations during the 2020-2021 COVID-19 lockdown in South Africa
- Open Access
- 30 January 2024
- Published by Academy of Science of South Africa in South African Journal of Science
- Vol. 120 (1/2)
Large language models and academic writing: Five tiers of engagement
Against a backdrop of the rapidly expanding use of large language models (LLMs) across diverse domains, this discussion breaks LLM usage into tiers of use, offering practical guidance to cautiously embrace the benefits of this significant new tool.
- 30 January 2024
- Published by Academy of Science of South Africa in South African Journal of Science
- Vol. 120 (1/2)
Open AccessLarge language models and academic writing: Five tiers of engagement
Against a backdrop of the rapidly expanding use of large language models (LLMs) across diverse domains, this discussion breaks LLM usage into tiers of use, offering practical guidance to cautiously embrace the benefits of this significant new tool.
- Open Access
- 30 October 2023
- Published by Center for Open Science in SocArXiv
Making ends meet: spatial and socio-economic variations during the 2020- 2021 Covid-19 Lockdown in South Africa
To mitigate the ravages of Covid-19, governments across the world imposed limitations on human movement from early 2020 onwards. In South Africa, as elsewhere, such limitations were framed as aiming to achieve a delicate balance between saving lives and livelihoods. One way to test the impact on livelihoods is to look at various socio-economic metrics, and their spatial patterns over time. We draw on data from the UJ-HSRC national Covid-19 Democracy Survey, a weighted, nationally representative sample (n = 38 785), collected from April 2020 to July 2021. Data from four iterations of the survey are used. We assess changes in the socio- economic and psycho-social conditions experienced by the adult population in South Africa, indicative that the experience of socio-economic distress far exceeded the epidemiological impact of the pandemic. Whereas infection rates during this peak period of the pandemic reached about 5% of the population, with provincial variation between 2% and 7%, experience of hunger and/or fear affected more than one-third. We find that feelings of fear were dominant in those areas where livelihoods were most affected by the pandemic, ostensibly caused by uncertainties related to job and income losses reported by a large number of respondents in the economic hubs of the country. Regarding economic distress, results indicated decreases in the large metropolitan areas of Johannesburg and Cape Town, but increases in other parts of the country. The results are significant in providing insight to changing distress levels, which is important for planning for distress relief allocation efforts.
- 30 October 2023
- Published by Center for Open Science in SocArXiv
Open AccessMaking ends meet: spatial and socio-economic variations during the 2020- 2021 Covid-19 Lockdown in South Africa
To mitigate the ravages of Covid-19, governments across the world imposed limitations on human movement from early 2020 onwards. In South Africa, as elsewhere, such limitations were framed as aiming to achieve a delicate balance between saving lives and livelihoods. One way to test the impact on livelihoods is to look at various socio-economic metrics, and their spatial patterns over time. We draw on data from the UJ-HSRC national Covid-19 Democracy Survey, a weighted, nationally representative sample (n = 38 785), collected from April 2020 to July 2021. Data from four iterations of the survey are used. We assess changes in the socio- economic and psycho-social conditions experienced by the adult population in South Africa, indicative that the experience of socio-economic distress far exceeded the epidemiological impact of the pandemic. Whereas infection rates during this peak period of the pandemic reached about 5% of the population, with provincial variation between 2% and 7%, experience of hunger and/or fear affected more than one-third. We find that feelings of fear were dominant in those areas where livelihoods were most affected by the pandemic, ostensibly caused by uncertainties related to job and income losses reported by a large number of respondents in the economic hubs of the country. Regarding economic distress, results indicated decreases in the large metropolitan areas of Johannesburg and Cape Town, but increases in other parts of the country. The results are significant in providing insight to changing distress levels, which is important for planning for distress relief allocation efforts.
- 25 October 2023
- Published by Taylor & Francis
- p. 233-253
Rallying the Nation
In the context of the global COVID-19 pandemic, trust in the institutions of government and between individuals became essential not just for long-term prospects of democratic resilience and durability but also to fight the spread of the virus effectively. In this chapter, we chart institutional and interpersonal trust in pre-pandemic, early pandemic (April and May 2020) and advanced pandemic (September 2020 onwards) phases. In examining institutional trust, we evaluate trust in the national government and the presidency, as the national government coordinated South Africa’s COVID-19 response. We explore data on two critical elements of interpersonal trust, that is, trustingness (the extent to which one feels able to trust others) and ‘other’ trust-focus (the extent to which one feels others are trustworthy) (Zang, 2021). These are important considerations for managing a pandemic or other social crises. Teasing out key lessons for building political trust in South Africa, we gauge the impact of a possible ‘rallying effect’ that may have shaped institutional and interpersonal trust early in the pandemic.
- 25 October 2023
- Published by Taylor & Francis
- p. 233-253
Rallying the Nation
In the context of the global COVID-19 pandemic, trust in the institutions of government and between individuals became essential not just for long-term prospects of democratic resilience and durability but also to fight the spread of the virus effectively. In this chapter, we chart institutional and interpersonal trust in pre-pandemic, early pandemic (April and May 2020) and advanced pandemic (September 2020 onwards) phases. In examining institutional trust, we evaluate trust in the national government and the presidency, as the national government coordinated South Africa’s COVID-19 response. We explore data on two critical elements of interpersonal trust, that is, trustingness (the extent to which one feels able to trust others) and ‘other’ trust-focus (the extent to which one feels others are trustworthy) (Zang, 2021). These are important considerations for managing a pandemic or other social crises. Teasing out key lessons for building political trust in South Africa, we gauge the impact of a possible ‘rallying effect’ that may have shaped institutional and interpersonal trust early in the pandemic.
- Open Access
- 24 October 2023
- Published by Center for Open Science in OSF Preprints
Large Language Models and Academic Writing: Five tiers of engagement
In light of the mass adoption of Large Language Models assistance to academic writing, five tiers of LLM support for academic writing are introduced, each offering a different level of writing support, and each entering the writing (and thought-) process at a different stage. Tiers range from no AI-assistance to complete AI-coproduction. Regarding guidelines for publications and assessment-setters, this piece advocates the most utility, and the most practical and morally defensible position, not at the extremes, yet towards the lower end of the continuum. In addition, it claims that, with some intentionality, the principles of ownership (plus responsibility) and transparency (sharing of prompts) can, and ought to be maintained.
- 24 October 2023
- Published by Center for Open Science in OSF Preprints
Open AccessLarge Language Models and Academic Writing: Five tiers of engagement
In light of the mass adoption of Large Language Models assistance to academic writing, five tiers of LLM support for academic writing are introduced, each offering a different level of writing support, and each entering the writing (and thought-) process at a different stage. Tiers range from no AI-assistance to complete AI-coproduction. Regarding guidelines for publications and assessment-setters, this piece advocates the most utility, and the most practical and morally defensible position, not at the extremes, yet towards the lower end of the continuum. In addition, it claims that, with some intentionality, the principles of ownership (plus responsibility) and transparency (sharing of prompts) can, and ought to be maintained.
- 24 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Online Education Poverty: A Multidimensional Approach
The COVID-19 pandemic prompted universities to shift to emergency remote teaching. This shift foregrounded previously underappreciated dimensions of poverty. Poverty is a complex and multidimensional phenomenon. When considering the needs of university students, conceptions and measures of poverty should include the lack of the resources required to access online education. This research seeks to create a multidimensional poverty index, that can be used to assess the poverty of students with respect to online education. This index will be useful when planning interventions that can assist students. The index is created by combining the well-known Alkire-Foster method with concepts emanating from the digital divide, energy poverty and educational technology. The work will focus on students at the University of the Witwatersrand in South Africa.
- 24 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Online Education Poverty: A Multidimensional Approach
The COVID-19 pandemic prompted universities to shift to emergency remote teaching. This shift foregrounded previously underappreciated dimensions of poverty. Poverty is a complex and multidimensional phenomenon. When considering the needs of university students, conceptions and measures of poverty should include the lack of the resources required to access online education. This research seeks to create a multidimensional poverty index, that can be used to assess the poverty of students with respect to online education. This index will be useful when planning interventions that can assist students. The index is created by combining the well-known Alkire-Foster method with concepts emanating from the digital divide, energy poverty and educational technology. The work will focus on students at the University of the Witwatersrand in South Africa.
- 1 January 2023
- Published by Review of African Political Economy in Review of African Political Economy
- Vol. 50 (175) , 107-115
The EFF as a ‘gateway party’? Briefing based on data from the 2021 South African local government elections
SUMMARY: This briefing offers three contributions concerning the voter profile of the Economic Freedom Fighters (EFF), as witnessed at the 2021 elections in South Africa. The first is that the party’s support base is relatively well educated (compared to the ruling African National Congress). Second, well over 10% of party support comes from relatively high earners, as measured by income levels. Finally, EFF support appears notably ‘fluid’, as indicated by voters switching support to the EFF (mostly from the ANC) and away from the EFF (mostly towards smaller parties), ultimately suggesting an image of the EFF as a ‘gateway party’.
- 1 January 2023
- Published by Review of African Political Economy in Review of African Political Economy
- Vol. 50 (175) , 107-115
The EFF as a ‘gateway party’? Briefing based on data from the 2021 South African local government elections
SUMMARY: This briefing offers three contributions concerning the voter profile of the Economic Freedom Fighters (EFF), as witnessed at the 2021 elections in South Africa. The first is that the party’s support base is relatively well educated (compared to the ruling African National Congress). Second, well over 10% of party support comes from relatively high earners, as measured by income levels. Finally, EFF support appears notably ‘fluid’, as indicated by voters switching support to the EFF (mostly from the ANC) and away from the EFF (mostly towards smaller parties), ultimately suggesting an image of the EFF as a ‘gateway party’.
- Open Access
- 2 October 2022
- Published by Taylor & Francis in Politikon
- Vol. 49 (4) , 297-317
Beyond the binary: examining dynamic youth voter behaviour in South Africa
- 2 October 2022
- Published by Taylor & Francis in Politikon
- Vol. 49 (4) , 297-317
Open AccessBeyond the binary: examining dynamic youth voter behaviour in South Africa