How are Data used to Create Change? The Case of Violent Attacks on Health Care

On 28 March 2021 Myanmar security forces shot protesters in Yangon city. Some sought refuge in the hospital where soldiers and police followed them and opened fire. Unfortunately, this kind of violence on health care is all too common in contexts affected by armed conflict. Since the full-scale invasion of Ukraine in 2022, the World Health Organization (WHO) Surveillance System for Attacks on Health Care (SSA) has verified over 715 attacks on health. A Safeguarding Health in Conflict Coalition (SHCC) and Insecurity Insight review of five years of data on attacks found ‘more than 4,000 unique incidents of violence against health care in situations of armed conflict—on average more than two incidents a day.’ These attacks threaten health workers, individual health outcomes for patients and conflict-affected populations, and jeopardise access to health services. 

The question in the title may seem relatively straightforward. Scholarly and policy analysis about attacks on health care call for more data and better monitoring to document and understand the issue and to ensure accountability (see here and here). These numbers have gotten headlines about Ukraine and COVID-19, helping to raise awareness about these attacks. 

But do numbers do more than raise awareness? What do we know about their influence? Put another way, what is the relationship between data, often numbers, and changes in policy or behaviour? Scholars have examined the efficacy of transnational advocacy and decision-making but few have examined the specific role of data in these processes. 

In a recent publication, my colleague Róisín Read and I consider the relationship between data and change, as part of a broader effort to research the impact of attacks on health care. In our open-access article, we argue that data about attacks on healthcare are indeed necessary for understanding the scope of the problem and for raising awareness. But the continued occurrence of attacks demonstrates that data are insufficient in creating normative, policy, or behavioural change. To investigate the complex and potential role of data in these processes, we focus on two pathways for change. We call the first pathway ‘operational change,’ designed to prevent or mitigate the impact of attacks on health. The second refers to normative change, often pursued via transnational advocacy aiming to achieve a reduction in the frequency of attacks. The former operates at the level of those affected by attacks, while the latter works at the level of those perpetrating attacks.

Our investigation highlights the institutional, political, and social contexts in which data are produced and used, and how these contexts can be as significant as the evidence they provide for decision-making and advocacy efforts. We find that many issues impact on the role of data related to policy or programmatic change, from the technical (eg related to standards and terminology) to issues of bias and the social or institutional networks that shape data collection and use. To be useful, data should be collected with a clear purpose that is meaningful for those collecting, analysing, and using the data. Moreover, the political context impacts on the framing of data and the incentives to under- or overreport, whether about harms or disease. Even the terminology used in collecting data can be a point of political contention. As we write, ‘Data are never neutral; they privilege particular, subjective realities that are especially contested in fractious political contexts.’ 

Additionally, at the levels of operational and normative change, the role of personal and institutional relationships are crucial. For instance, individuals and organizations bring existing biases and frames of reference to bear on the data they encounter. As a result, data that challenge preconceptions are likely to require a higher burden of proof to become credible. Yet personal and institutional connections also strengthen trust in and interpretation of data. This highlights the crucial role of broad-based networks, which can help to build trust in the underlying data. In doing so, these connections enhance the potential for data to influence change.As academics do, we conclude with a call for more research to investigate the often positive but non-linear role of data in change processes. While our specific focus was on the relationship between data attacks on health, we hope these insights assist other efforts to affect decision-making or create behavioural, policy, or normative change.


How can smart cities shape a happier life? The mechanism for developing a “Happiness Driven Smart City”

Smart cities are expected to provide better solutions for the intensified socio-economic and environmental challenges associated with unprecedented urbanisation by embracing advanced information and communication technologies (ICTs), where challenges like climate change, energy crisis, or social inequality could be addressed through the development and application of state-of-the-art technologies (United Nations, 2019). The idealised narrative of the smart city has been widely accepted and turned into global movement at an appreciable speed for economic expansion and societal transformation, with more than 1,000 cities globally having introduced smart city initiatives with Europe, North America, Japan and South Korea in the leading position (Deloitte, 2018).

However, it appears that smart city’s influence on human happiness has been paid only marginal attention amidst the practices of smart city development. It is often assumed that smart cities may carry both opportunities and risks to human happiness due to its technology-oriented essence, but little information is available to demonstrate which specific aspects of human happiness might benefit or get harmed from the introduction of smart city initiatives. In fact, until we understand how human happiness is affected by smart city initiatives, the holistic benefits of smart city development upon urban inhabitants will remain under question, no matter how much prosperity the advocators promise. The cost of rectifying the progress of smart city development would be huge if any strategy or policy leads to unexpected or unwanted directions which might jeopardise human happiness. Therefore, the big question with regard to the influence upon human happiness brought by smart city development is whether and how smart cities act upon human happiness.

This blog introduces the concept of the Happiness Driven Smart City (HDSC), which will be constructed as a three-layer interrelated functioning structure underpinned by a set of Strategic Measures. For this, four procedures will be conducted, 1) to propose the conceptualisation and principle characteristics of the Happiness Driven Smart City; 2) to construct the key factors contributing to the performance of HDSC characteristics; 3) to build-up the HDSC development mechanism; and 4) to develop the underpinning Strategic Measures and specific application toolkit of the HDSC mechanism.

  1. Happiness Driven Smart City: conceptualisation and characteristics

Figure 1 Happiness Driven Smart City: Conceptualisation and characteristics

2. Key factors contributing to characteristic performance in a Happiness Driven Smart City system

Table 1 Key factors contributing to HDSC characteristic performance

HDSC CharacteristicsFactor 1Factor 2Factor 3Factor 4
Efficient and green physical infrastructure (HDSC1)MobilityEnergyPublic Utilities 
Labour-friendly and innovative economy (HDSC2)EmploymentInnovative SpiritEntrepreneurship 
Inclusive and attractive society (HDSC3)EducationHealthSafetyCulture and Leisure
Sustainable and eco-friendly natural environment (HDSC4)Air qualityPollution and Waste Treatment  

3. Mechanism for developing Happiness Driven Smart City

The synthesised characteristics of a Happiness Driven Smart City presented in Section 1 and the corresponding key factors identified in Section 2 are used to build up the mechanism for developing a Happiness Driven Smart City. The mechanism for developing a Happiness Driven Smart City can be portrayed graphically in Figure 2 and is composed of three components, namely, Overarching objective (top-layer), Characteristics (medium-layer) and Factors (bottom-layer). The working mechanism of HDSC shows the structural relationships between these three components. The bottom-layer Factors are the sources of changing the performance of HDSC Characteristics, which in turn contribute to the Overarching objective of developing a Happiness Driven Smart City. The functions of the HDSC system are underpinned by a set of Strategic Measures which act directly upon the Factors in the HDSC system. By applying various Strategic Measures to change Factors, momentum can be gained to improve the performance of medium-layer Characteristics. Consequently, the Overarching objective of developing a Happiness Driven Smart City can be achieved.

Figure 2 Three-layer mechanism system for developing a Happiness Driven Smart City

4. Strategic measures for the HDSC development mechanism

As shown in Figure 2, the key to make a Happiness Driven Smart City happen is the application of a set of Strategic Measures. These Strategic Measures are developed through examining the interrelationship between the three components in the HDSC system, and the examination and establishment of the Strategic Measures is conducted through a comprehensive literature review. Consequently, a set of Strategic Measures have been developed, which are presented by reference to each of the four HDSC Characteristics, as shown in Tables 2, 3, 4 and 5 respectively. In these tables, a shortlist of Strategic Measures is developed in addressing different factors which exert a driving influence upon the concerned HDSC Characteristic.

Table 2 Strategic Measures for improving the performance of HDSC characteristic “Efficient and green physical infrastructure” (HDSC1)

Factors influencing HDSC1Strategic Measures to act on factors for improving the performance of HDSC1
F11:MobilitySM111: To integrate the principles of green transport into the process of developing smart mobility thus the performance of green infrastructure will be improved. By developing a leveraged mobility paradigm between walking, cycling and driving, greenhouse gas emission will be reduced. SM112: To eliminate possible negative effects caused by the misuse and monopoly of smart mobility technologies which would hurt the users’ benefits and affect negatively the development of HDSC in the long run. For example, regulations shall be formulated to ensure smart mobility service providers take responsibilities in delivering a greener and more equal transport system.
F12:EnergySM121: From the industrial aspect, to take consideration of both the optimisation of energy sources and conservation on the consumption end. For example, integrating smart grids with renewable energy resources is a direct and effective solution for improving green grid management; adopting artificial intelligence and big data analysis into building, manufacture and transportation sectors to forecast and minimize energy consumption is proved effective for minimizing energy consumption and consequently mitigating impacts upon environment and climate. SM122: From the individual aspect, to create an advantageous environment with the facility of technology to encourage individual energy saving behaviours. The potential impact of occupants energy saving behaviour on buildings is evidenced and identified as an essential approach to improve energy efficiency in green buildings and communities without jeopardizing the level of comfort.
F13:Public UtilitiesSM131: To explore resource efficiency and utilization opportunities from both service providers and consumers in smart utility network development. For example, artificial intelligence application in leak detection of water distribution pipelines will increase water resource utilization efficiency. The water/gas consumption feedback technology can help to reduce household wasteful behaviour to trigger the decrease of environmental impact in every step along the whole journey of resources processed by public utility facilities.

Table 3 Strategic Measures for improving the performance of HDSC characteristic “Labour-friendly and innovative economy” (HDSC2)

Factors influencing HDSC2Strategic Measures to act on factors for improving the performance of HDSC2
F21:EmploymentSM211: To prepare solutions and resources to minimize the potential disruptive impact on employment during the dramatic change process and to create a labour-friendly employment market environment to help citizens quickly fit into the new economy. Policies to stimulate education evolution for promptly fulfilling high profile new business requirements and training programs to look after disadvantaged labour shall be taken into full implementation. SM212: To enact labour protection and social protection rules and regulations that are suitable and in favour of employees to keep new digital economy development under a labour-friendly premise. It would be ideal if the policymakers are able to take a longer view and make foreseeable actions on the employment influence arising from the digital economy instead of keeping the legal and social regulation system in a passive adaption situation.
F22:Innovative SpiritSM221: To make the smart city as a nexus for open innovation, which should not just refer to industry but also the ways government and other institutions work and collaborate with society, to jointly create an inclusive, labour-friendly and innovative economy. All sources of innovation from different levels and different sectors shall be encouraged to actualize an innovative and diverse economy where technology plays the catalyst role. SM222: To inspire the potential of new bottom-up approaches based on user-generated content through the experiences of the citizens themselves, which enables citizens to build social capital and the capacity required to become co-creators and co-producers of new and innovative services with the means to ensure that they are delivered in more effective and inclusive ways, taking full advantage of new Internet-based technologies and applications.
F23:EntrepreneurshipSM231: To encourage the utilization of new technologies and promote a strong pro-entrepreneurial state ethos where an innovative economy is nurtured and accelerated. SM232: To encourage and facilitate entrepreneurial citizens to trigger, apply and transform emerging technologies and knowledge into new products, new jobs, and new firms, which in turn enables the creation of a labour-friendly innovative economy. SM233: To promote neo-liberal attempts to inclusively and effectively incorporate the local community into the entrepreneurial city, via the approaches of various participatory ICT projects.

Table 4 Strategic Measures for improving the performance of HDSC characteristic “Inclusive and attractive society” (HDSC3)

Factors influencing HDSC3Strategic Measures to act on factors for improving the performance of HDSC3
F31:EducationSM311: To create an easy environment for innovation during the digital transformation process of education through a participatory approach to enhance social inclusiveness and talent attractiveness in the education sector. All stakeholders in this venturous journey, such as teachers, students, administrators, online platform companies, software and hardware providers etc., shall be encouraged to be more active to innovate and design better solutions to meet the new need. SM312: To address the equity of education resources and opportunities to avoid extreme deprivations caused by the digital transformation to ensure a more inclusive development in cities. For example, online education shall take in consideration of the balance between commercial online education programs and public-free courses so as to benefit the widest audience from the well-educated to the low-skilled learners.
F32:HealthSM321: To improve accessibility to health facilities and services to the widest public for disease prevention and health promotion where technological innovation can be applied as a tool for empowering social inclusiveness. For example, more attention and funding shall be paid to mental health care services where big data and artificial intelligence technology can play an innovative role in prediction and diagnosis. SM322: To encourage the application of new technologies aiming to produce new opportunities to improve the health treatment to a more accurate and effective level where patients can receive higher quality and safer medical service with reduced cost and wastage. The improvement of affordability and effectiveness in medical services and health care driven by smart technology will positively affect the inclusiveness of the society.
F33:SafetySM331: To create a safer social environment by taking actions to prevent the occurrence of crime before it happens via the application of new technology-based crime risk prediction analysis approaches. Safety clearly exerts higher impacts on urban attractiveness.
F34:Culture and LeisureSM341: To create an environment that encourages wide public participation, cultural diversity, digital equity in the context of digital technology to contribute to a city’s attractiveness and social inclusion. The development of a cultural policy that aims for cultural participation may also involve other policy areas such as economic and education sectors. SM342: To provide a pleasant environment and necessary support for both offline and online leisure activities to provide greater opportunities for urban residents to enjoy a higher level of life satisfaction and social inclusion.

Table 5 Strategic Measures for improving the performance of HDSC characteristic “Sustainable and eco-friendly natural environment” (HDSC4)

Factors influencing HDSC4Strategic Measures to act on factors for improving the performance of HDSC4
F41:Air QualitySM411: To monitor and forecast air quality more precisely in both vicinity (building and neighbourhood) and city level through collecting and analysing data with the ubiquitous upgraded devices and advanced algorithms, and to satisfy the evolved need for environment protection and human life by taking more targeted and effective solutions. SM412: To enable citizens to have tailored air quality notification to help prevent any exposure risks amongst vulnerable groups to reduce the ecological threat on public health (ibid).
F42:Pollution and Waste TreatmentSM421: To control emissions and effluents through IoT-enabled smart system in various sectors including building, manufacturing industry and logistics etc. SM422: To modernize traditional waste management systems to prevent the negative effects of incorrect operation on both people and the environment during the whole process including waste collection, disposal, recycling, and recovery (ibid). SM423: To enable citizens to track daily personal pollution footprints via smart approaches to better understand and behave upon reducing waste emission and production towards a sustainable and eco-friendly environment.

The proposed theoretical mechanism can be applied with specific assessment criteria to examine to what extent a smart city initiative implemented in a given city has enhanced urban residents’ happiness and has achieved the goal of a Happiness Driven Smart City. The application of the HDSC mechanism can thus help urban governors to understand the status quo of smart city development and to better guide the design of smart cities towards a happiness-driven and human-centred direction.

Note: This blog is based upon a recent publication on the journal of Sustainable Cities and Society.

Development Transformation as the Goal for Digital Transformation

Richard Heeks, Bookie Ezeomah, Gianluca Iazzolino, Aarti Krishnan, Rose Pritchard & Qingna Zhou

There’s a lot of talk currently about digital transformation for development.  Sometimes styled “DX4D”, a quick definition would be radical change in development processes and structures enabled by digital systems.

Digital transformation is thus not a goal.  Instead, development transformation is the goal and starting point.  But what kind of development transformation?

In this post, we summarise what development transformation would mean under different development paradigms, and some implications for digital.  The table below is not exhaustive of the various paradigms and it rather brutally simplifies rich and complex ideas.  However, it does help clarify two key DX4D tenets:

  • Vision Matters: unless you know where you want to get to, digital can’t help take you there.
  • Visions Differ: different paradigms aim for very different destinations and, hence, different journeys in the application of digital.

If you have paradigms you’d like to add or you have improvements to offer on what’s in the table, do let us know.

Development Paradigm Essence? Transformation? Digital Implications?
Neoliberal Markets and market relations are the central foundation for economic development.  They, and not government regulation or vested interests, are the best way to allocate development resources and to generate productivity improvements and growth.  The state acts to support market-driven development. Neoliberalism is thus about the reformatting of politics, society and individuals according to market logics, the pursuit of profits, and individual responsibility principles. Stabilisation to reduce government expenditure.  Liberalisation to roll back state regulation, subsidies and other interventions in markets and the private sector.  Privatisation to transfer ownership from public to private sector. Digital, particularly via platforms, must enable the formation and presence of markets in all sectors, and a step change in market functioning via datafication and machine-readability of market actors and processes. Digital will also enable the development of private sector responsibility for public service delivery, and major improvements in efficiency of remaining public sector functions.
Structuralist Particular socio-economic structures inhibit development.  For dependency variants, it is unequal relationships of exchange between core and periphery, whether understood in terms of countries, regions, or more immaterial geographies.  For Marxist and related anti-capitalist paradigms, it is unequal relationships of exchange between capital and labour. The exploitative socio-economic structures must be broken away from and/or replaced.  From a dependency perspective, at the extreme, this means autarky and a focus on localised systems of production and consumption.  From a Marxist perspective, it means the end of capitalism and its replacement with structures of common ownership. Digital must support radical structural change based around localised production and/or cooperative or similar ownership structures.
Sustainable Ensuring resource usage does not compromise the ability of future generations to meet their own needs; with variants ranging from green growth through to de-growth. Major reductions in resource usage including improvements in efficiency of resource-using processes.  Major reductions in polluting outputs from processes. Internalisation of negative environmental externalities so as to gauge the true cost of economic growth. Digital must support a step-change in resource usage and polluting outputs of all economic and social processes, including those involving digital itself.  Digital must also support environmental mapping and monitoring to track progress of sustainability. 
Human Development Development as freedom; in particular economic, political, social, security and informational freedom for all so that no-one is left behind and all have the opportunity to be and to do what they wish.     Changing contexts so that there is equality of opportunity and equality of choice; especially for those currently denied those opportunities.     Digital must be not just accessible but usable and appropriable by all.  It must then support the ability of all to choose the kind of lives and livelihoods that they value; thus requiring some customisation to individual contexts rather than a blanket equality of access to assets, institutions and livelihoods.
Decolonisation Reversal of the current and legacy negative impacts of colonisation. Enabling sovereignty and “self-determination of indigenous peoples over their land, cultures, and political and economic systems”[1].  Also understood – and drawing on the post-development roots of one strand of decolonisation – as the identification, challenging and revision or replacements of “assumptions, ideas, values and practices that reflect a colonizer’s dominating influence and especially a Eurocentric dominating influence”[2]. Digital must be accessible, usable and appropriable by indigenous peoples, enabling them to exercise self-determination.  Digital sovereignty will enable local “control over digital assets, such as data, content or digital infrastructure, or over the use of those assets”[3] and prevent uncontrolled extraction of value from these assets by others. For the latter understanding of decolonising transformation, digital must empower those who have been disempowered by Eurocentric domination of epistemics and discourse, and enable them to engage with and challenge that domination.  

Photo by Javier Miranda on Unsplash




Is there a silver lining to this covid-19 pandemic? The impact of government transfer on financial inclusion

The toll of covid-19 pandemic is still being counted. Meanwhile, new variants continue to threaten because the reservoir of people infected is still large enough for some random mutations to take hold and spread. Excess deaths, a global measure, were estimated by the WHO to be around 14.9 million in 2020-2021.

With all this immense suffering, is there a ray of hope? After all, governments provided support to family finances, delivered through digital and other means. With deepening financial inclusion (more families owning bank accounts), there are benefits to share. Account holders can smooth consumption and manage risks, banks can arbitrage between savers and inventors, governments can expect increased investment in the economy and better information on the funds circulating, useful for macroeconomic management.

If the outpouring of governments’ responses through digital transfer programmes improves financial inclusion, all is not lost. Examples of these programs have been curated by various groups including the World Bank. I used their database to mark whether the country has any digital transfer program (Gentilini 2022, Gentilini et al 2021, Marin & Palacios 2021) then collated the last three global financial inclusion surveys, the Global Findex, 2014, 2017 and 2021. (I wrote about these surveys before.) There were some 400,000 adults aged ≥15 in 146 countries analysed.

To estimate the digital transfer impact, one can apply a double difference estimator, using the available information before and after the pandemic started in 2020. The estimator assumes parallel trends in financial inclusion: countries with and without such transfer moved side-by-side in the past 7 years. Surely, they did not.

So I applied a doubly robust version, balancing the trends with information on society’s readiness for such programmes (digital adoption index in 2014) which accompanies the World Development Report 2016. Three outcomes are important: account in banks, in any formal financial institution and as mobile money (all binaries).

Digital readiness in 2014 and excess death in 2020-2021

Countries were visited by this pandemic at different digital readiness as shown by the span of digital adoption index in 2014. But initial digital readiness relates to lower excess death later in the pandemic, once readiness passes the median (Figure 1). Beyond the median, excess death follows a steeper decline. The pandemic has witnessed governments and society deliver transfers and various supports using digital technology, all this shows in excess deaths.

Among the survivors: financial inclusion in the pandemic’s wake

Because transfer programmes were delivered through digital means in some countries but not others, what impact do digital ones have in financial inclusion? I first check whether countries’ readiness indeed relates to delivering support via digital transfers. The marginal plot shows that the relation is positive, with more digital adoption in 2014 associates with higher probabilities of digital transfers in 2020.

Finally, putting all this together, the data showed an impact of digital transfer programme of 2% on owning any bank account to 8% on owning a mobile money account in the wake of the pandemic (Figure 3).

Perhaps this deepening financial inclusion is a ray of hope amidst the immense suffering. Governments, businesses and families can draw benefits from being part of a more inclusive financial system in the wake of this pandemic.

Global Development Institute – University of Manchester

United Nations University – World Institute for Development Economics Research


Gentilini U. 2022. Cash Transfers in Pandemic Times : Evidence, Practices, and Implications from the Largest Scale Up in History. World Bank Group.

Gentilini, U., Almenfi, A., Blomquist, J., Dale, P., De La Flor Giuffra, L., Desai, V., Fontenez, M., Galicia Rabadan, G., Lopez, V., Marin Espinosa, A., Natarajan, H., Newhouse, D., Palacios, R., Quiroz, Ana., Rodriguez Alas, C., Sabharwal, G. and Weber, M. (2021) ““Social Protection and Jobs Responses to COVID-19: A Real-Time Review of Country Measures”. World Bank, Version 15 (May 14).

Marin G. and Palacios R. 2021. The Role of Digital in the COVID-19 Social Assistance Response. Washington, D.C. : World Bank Group.

Tampubolon G. 2018. The Puzzle of Digital Financial Inclusion: A Generation Game? ICTs for Development blog. World Bank 2016. World Development Report 2016: Digital Dividends. World Bank.

Digital tourism and marginal providers after the crisis

Bouncing back” in tourism should not be about connecting local providers to platforms but ensuring that available online tools provide inclusive outcomes

As tourism has become global it has become an important part of the economy in a number of countries of the global south. It brings foreign currency into the economy and provides a surprising number of jobs to those who provide services. For all the ethical and environmental issues it poses, in countries such as Thailand, Indonesia, Tanzania and Rwanda, the loss of tourists during the pandemic led to crises. Vast swathes of workers and firms have had to move into other sectors with broader implications for economies.

A major imperative following the crisis has been to ensure that tourism can bounce back. Institutions such as the OECD and the UN, as well as development donors and governments have pushed recovery plans with significant “digital tourism” components. Embracing digital tourism is seen as a quick win. Tourists have grown more used to online platforms – whether that be booking hotels, arranging transportation, sharing tourism experiences or posting reviews. The vision of recovery plans is that if local providers can embrace platforms, not only will they become more efficient, but drive forward tourism demand.

Drawing on recent research examining platform practices of small/marginal tourism service providers in Indonesia and Rwanda [1], we argue these visions for digital tourism may have limits. This research highlights three major considerations: the contexts of the adoption of platforms by tourism firms, the inflexibility of tourism platforms, and how tourism development may better be guided by grassroots online practices.

Platform use by small service providers

With several decades of investment in internet connectivity, the costs and barriers to internet use have been reducing, leading to growing use. This is especially the case for businesses in tourism, where digital tourism is becoming the norm. In both Indonesia and Rwanda, major global platforms such as TripAdvisor, Uber,, Airbnb, Google maps and Traveloka are now well-established.

With the growing ubiquity, we might say that platforms are moving from something that forward-thinking firms opt into, to being non-negotiable for all firms. It is now almost like an infrastructure that firms need to be part of. This is true even amongst more marginal service providers such as tour guides, tiny hotels and those providing cultural activities who would use mobile devices to be part of such platforms.

Whether they want to go online or not, they are aware they are being mapped, rated and discussed online.

The complexity of tourism platforms

At first glance platforms seem to offer significant potential. They are easy to sign up for tourism providers and provide a way to quickly reach and interact with tourists across the globe. They often offer services such as online payments and booking systems that can make operations more efficient.

However, for small tourism providers platforms remain a challenge. While it is easy to join, successfully harnessing these platforms requires a broad range of technical skills. Successful firms need to be adept at website design, digital media skills and social media use to be able to stand out.

Challenges are not just about the capabilities of service providers, platforms are often highly complex and inflexible. For example, in Rwanda, small hotels were spending time and resources trying to move up search rankings on platforms. In Indonesia, some providers of tourism services were trying to negotiate algorithmic pricing systems.

With local support from platforms often non-existent and limited flexibility, small providers in these countries often suffered in competition with larger and foreign providers who were better places to make gains from being online.

Agency of tourism service providers

Even with these significant challenges, small service providers were able to combine digital tools for benefits – using shared calendar software, mobile apps, cloud sharing, online translation and social media to collaborate with customers and better fit with their daily needs.

Moreover, in Indonesia some tourism enterprises have come together to collaborate in more social- or environmentally-orientated online spaces.  In some other countries, we have also seen the success of commercial platforms more attuned to small enterprise needs and activities (e.g. South African platform Nightbridge)

These types of activity are very different to the policy prescriptions of joining the platform “juggernauts” for pandemic recovery. They suggest alternative ways forwards for small tourism providers – by amplifying the bottom-up activities already occurring outside mainstream platforms, and by being aware that service providers are negotiating multiple platforms and online software.

Summary: Bouncing back and “digitalisation”

These experiences of tourism and the goals of pandemic recovery are mirrored in other sectors in the global south. Governments and donors are not sitting back but seeking to play an active part in recovery through support. And, like tourism, one of the areas that are repeatedly mentioned is “digitalisation” – supporting so-called “inefficient” small firms to connect and use digital platforms for economic gain. 

But as this research shows, the reality is that platforms pose challenges. Connecting online is often no longer the major barrier. Rather platforms fit poorly with the skills of small firms and their growing complexity favour better-financed firms. They rarely adapt to the challenges faced in global south contexts.

Blindly shepherding firms towards adopting large platforms may negatively affect small providers. Interventions should rather support more creative uses of technology and leverage the unique relationships and applications that could afford more inclusive outcomes.

[1] This article is based on the recently published paper:

Foster, C., & Bentley, C. (2022). Examining Ecosystems and Infrastructure Perspectives of Platforms: The Case of Small Tourism Service Providers in Indonesia and Rwanda. Communications of the Association for Information Systems, 50

An open-access version is available to download.

Latest Digital Development Outputs (China, Data, Economy/Platforms, Inclusion, Water, Rights, Sustainability) from CDD, Manchester

Recent outputs – on China Digital; Data-for-Development; Digital Economy / Platforms; Digital Inclusion; Digital Water; Rights; and Sustainability – from Centre for Digital Development researchers, University of Manchester:


China’s digital expansion in the global South” presents recordings of nine presentations at a CDD international workshop that discusses the implications for the global South of China’s emergence as a digital superpower.

Understanding the evolution of China’s standardization policy system” (open access) by You-hong Yang, Ping Gao & Haimei Zhou, investigates the evolution of China’s technology standardization policy system in the period from 1978 to 2021.  


A DC State of Mind? A Review of the World Development Report 2021: Data for Better Lives by Hellen Mukiri-Smith, Laura Mann & Shamel Azmeh, reviews the World Development Report (2021) on data governance.


Examining ecosystems and infrastructure perspectives of platforms: the case of small tourism service providers in Indonesia and Rwanda” (open access version available) by Christopher Foster & Caitlin Bentley, analyses tourism platforms from the perspective of small and marginal service providers. It is useful to move away from ideas of platform leaders organising ecosystems from the top-down, towards more emergent behaviours of service providers in multi-platform environments.

Automation and industrialisation through global value chains: North Africa in the German automotive wiring harness industry by Shamel Azmeh, Huong Nguyen & Marlene Kuhn, examines the implications of automation on the global map of production and the position of developing countries in global value chains. Through the case of the German automotive wiring harness industry, we examine the implications of ongoing automation processes on production in North Africa.

Digital public goods platforms for development: the challenge of scaling” (open access) by Brian Nicholson, Petter Nielsen, Sundeep Sahay & Johan Saebo.  We articulate the notion of digital global public goods and examine the development of DHIS2, a global health platform inspired by public goods, focusing on the paradoxes that arise in the scaling process. A presentation of the paper to the Pankhurst Institute, University of Manchester is available on YouTube.


Digital inequality beyond the digital divide: conceptualizing adverse digital incorporation in the global South” (open access) by Richard Heeks, presents a new model to understand how inclusion in – rather than exclusion from – digital systems leads to inequality.

Revisiting digital inclusion: a survey of theory, measurement and recent research” (open access) by Matthew Sharp, sets out a framework of core components of digital inclusion, surveys current measures of digital inclusion, and makes suggestions for how future research could be more rigorous and useful.


Water ATMs and access to water: digitalisation of off-grid water infrastructure in peri-urban Ghana” (open access) by Godfred Amankwaa, Richard Heeks & Alison L. Browne, finds water ATMs to be incremental infrastructures delivering relatively limited and operational-level value, but also producing new and contested socio-material realities.


RaFoLa: A Rationale-Annotated Corpus for Detecting Indicators of Forced Labour” (open access) by Erick Mendez Guzman, Viktor Schlegel & Riza Batista-Navarro, describes a dataset of news articles categorised according to forced labour indicators. The articles were annotated with rationales, i.e. human explanations for placing them under specific categories, to support the development of explainable AI systems.

Hustling day in Silicon Savannah: datafication and digital rights in East Africa” (open access) by Gianluca Iazzolino, Michael Kimani & Maddo, is a cartoon on the winners and losers in Kenya’s booming tech scene. It translates, for a non-academic audience, the authors’ research on how digital technologies are reshaping the informal economy in the global South.


Exploring financing for green-tech SMEs in East Africa: current trends and risk appetite” (open access) by Aarti Krishnan, reviews the financing of green-tech SMEs in East Africa including different financing at different enterprise lifecycle stages, in different sectors, and across different countries.

Applications of Industry 4.0 digital technologies towards a construction circular economy: gap analysis and conceptual framework” by Faris Elghaish, Sandra T. Matarneh, David John Edwards, Farzad Pour Rahimian, Hatem El-Gohary & Obuks Ejohwomu, investigates the interrelationships between emerging digital technologies and the circular economy, concluding with the development of a conceptual digital ecosystem to integrate IoT, blockchain and AI.

Antecedents of Significant Digital Development Research

This post is a cheat because it’s actually summarising a paper on organisational – not digital development – research.

It’s by the leading organisational theorist – and confutation of nominative determinism – Richard Daft, and I read it just before I started my PhD.

Based on a survey of organisational researchers, its findings feel relevant to digital development.  Significant research . . .

– Is an outcome of the researcher’s involvement in the real world

– Is an outcome of the researcher’s own interests, resolve and effort

– Is chosen on the basis of intuition

– Is an outcome of intellectual rigour

– Reaches into an uncertain world to produce something that is clear, tangible and well-understood

– Focuses on real problems

– Is concerned with theory, with a desire for understanding and explanation

Not-so-significant research is the opposite: expedient, quick and easy, lacking personal commitment from the researcher, lacking theoretical thought and effort, and so on.

While planning and clarity mark out the latter stages of significant research, it is the outcome of an organic process of intuition, integration of ideas from different fields or chance meetings, that starts with uncertainty.  Precisely planned, tidy, clean and clearly-defined research most likely leads to small results (research funders please take note!).

That all seems to fit equally-well with digital development research but, of course, these criteria come from a researcher perspective, not that of other stakeholders.  See what you think.

If you’d like to read the paper, it’s not so easy to find:

 – Daft, R. L. (1984). Antecedents of significant and not-so-significant organizational research. In: T.S. Bateman & G.R. Ferris (eds), Method and Analysis in Organizational Research. Reston Publishing, Reston, VA, 3-14.

Or, there’s a firewalled update:

– Daft, R. L., Griffin, R. W., & Yates, V. (1987). Retrospective accounts of research factors associated with significant and not-so-significant research outcomes. Academy of Management Journal30(4), 763-785.

Income of Gig Work vs. Previous Job in Pakistan

Richard Heeks, Iftikhar Ahmad, Shanza Sohail, Sidra Nizamuddin, Athar Jameel, Seemab Haider Aziz, Zoya Waheed, Sehrish Irfan, Ayesha Kiran & Shabana Malik

Does the transition to gig work improve incomes in Pakistan?

Many workers join gig work platforms in the belief that their incomes will improve, but is this borne out in practice?  To investigate, the Centre for Labour Research interviewed 94 workers based on six platforms across three sectors: ride-hailing, food delivery, and personal care.

Of these, 51 were able to tell us what their previous monthly income had been in their most-recent employment prior to joining the platform[1].  Stated income varied from the equivalent of US$60 per month up to U$1,200 per month, and averaged US$220 per month[2].

After moving into gig work, average gross income was slightly higher, at US$240 per month but, as the graph below shows, there was a much more differentiated picture behind the average, with around 40% of respondents earning less gross income (red-bordered blue columns) than they had done previously.

However, as the graph also shows, things looked worse when comparing net income (orange columns).  For the great majority of prior jobs, work-related costs were small (only work-to-home transport, which we calculated based on typical commuting journeys in Pakistan to be just under US$18 per month; i.e. less than 10% of average gross income).  But for gig work – much of which relies on journeys by vehicle and continuous internet connectivity – the costs of petrol, maintenance and data eat heavily into gross income.  In addition, for some (only a few in our Pakistan sample) there are costs of renting their vehicle.[3]

These costs represented, on average, 65% of gross income and knocked average net income for gig workers down to just US$85 per month.  When we compare before-and-after for net income, then, we found more than 70% of our sample were earning less than in their previous job, and 45% earned over US$100 per month less.

This was especially an issue for ride-hailing drivers and it does reflect the particular circumstances during our interview period of late 2021 to early 2022: a drop-off in demand for travel due to Covid, and a steep rise in petrol prices.  Indeed, so bad was the problem that just over a fifth – 21 of the 94 – were reporting negative income.  That is, they were effectively paying to go to work as their costs exceeded their gross income; something to which the platforms responded in May 2022 by dropping the commission taken from drivers to 0%.

While recognising the challenging period for gig workers covered by our fieldwork, nonetheless, this does suggest that – by and large – gig work is not delivering the income boost that workers often hope for.  They may, for example, be lured by gross income figures, not realising how much lower net income will be.  Gig work does provide a livelihood – 40% of our sample were unemployed in the immediate period prior to joining – but it is not really fulfilling its promise.  It also falls far from decent work standards: five-sixths of those we interviewed took home less than a living wage.

If you’d like to know more, please refer to the 2022 Fairwork Report on Pakistan’s gig economy.

[1] Those who stated what their prior employment had been gave the following job descriptions: BPO operator, Teacher (2), Housekeeper, Shopkeeper, Gas company worker (2), Safety officer, Business person, Tanker driver, Ride-hailing driver with another platform (3), Traditional taxi driver (3), Farmer, Builder, Computer operator, Cook, Technician, Shop assistant, Domestic worker, Government worker

[2] This average is some way above the overall average earnings of US$140 per month but well below formal sector average monthly salary of US$480.

[3] For further detail, see this discussion of the breakdown of ride-hailing passenger payments.

Digital public goods platforms for development

Nicholson, B. Nielsen, P. Sahay, S. Saebo, J. Digital public goods platforms for development: The challenge of scaling The Information Society available open access at:

Recently there has been an explosion of research into digital platforms.  To provide an indication of the size of the output, a quick search on Google Scholar provided 3270000 “hits”, 39900 in 2022 alone to date with publications across diverse disciplines including management, information systems, economics and more.   In the realm of ICT4D, discourse has focused on how platforms may enable socio-economic development (Nicholson et al 2021) however there is a paucity of examples of empirical research on how this may be realised.  

Digital platforms are defined according to their principal purpose and identifies two broad categories: transaction platforms and innovation platforms. Transaction platforms refer to a two or multi sided marketplace mediated by the platform.  Innovation platforms act as “foundations upon which other firms can build complementary products, services or technologies” (Gawer, 2009, p. 54).

Most prior empirical research on digital platforms involves commercial, for-profit platforms situated in the regulative institutions of the Global North.  Inherent in this prior work is an assumption of “monetisation” and the capitalist market forces, and little is known about platforms that are donor supported and aimed at socio economic development.    

A forthcoming paper attempts to address the knowledge gap by conceptualising innovation platforms as public goods and asking:

How can innovation platforms be public goods?

A goal of the article is to identify the challenges of simultaneously scaling up digital platforms and developing them into public goods.  Empirically, the focus is on health, specifically the empirical example is the District Health Information System (DHIS2). 

The relevance of public goods in development is well-established in the domain of health.  Initiatives driven by global health organisations such as the World Bank and World Health Organization aim to promote digital public goods. Digital Square, a marketplace initiative in digital health, has developed a Global Goods Guidebook and a Global Goods Maturity Model.  Before and during the pandemic, open-source systems have been launched to support outbreak management, such as the Surveillance Outbreak Response Management and Analysis System (SORMAS). SORMAS intuitively displays features of a public good: it is free of charge, open source, independent from tech companies, and interoperable with other platforms such as DHIS2.

Turning to theory of public goods leads us to the economics discipline and centres on two main principles: non-rivalry and non-exclusion. “Goods” such as crime control, flood defences etc. are provided because of failure of the market mechanism.  Government thus intervenes either financially, through such mechanisms as taxation or licensing, or with direct provision.   Public goods are non-rivalrous, implying that one individual’s consumption of the good does not influence what is available for others. They are also non-excludable, in the sense that no one can be excluded from consumption of a public good. 

This image has an empty alt attribute; its file name is picture1.jpg

Consider a lighthouse where one navigator’s use of the light does not prevent other navigators from doing the same. Many potential public goods exhibit only one of these properties resulting in the tragedy of the commons which can be illustrated with the example of a village pasture. Unrestricted access (non-exclusion) to the commons – pasture belonging to the village as a whole – leads to its degradation (rivalry). However, some scholars question the inevitability of depletion of common pool resources when they are managed in a bottom-up, cooperative way by those most dependant on them.  Under certain conditions, individuals govern themselves collectively, and without market pressures or government regulation, to obtain benefits, even if the temptation to freeride is present.

Global public goods are goods whose benefits cross borders and are global in scope for example eradication of infectious diseases where it is impossible to exclude any country from benefiting and each country will benefit without preventing another.

This image has an empty alt attribute; its file name is picture2.png

The district health information system or DHIS2 supports decentralized routine health management. The architecture is designed with a generic core that enables local innovation and anyone with internet access can at any time download the most recent version of DHIS2, the source code, as well as required libraries and third-party products (such as Chrome or Firefox browsers). DHIS2 also comes with a set of bundled apps, developed by University of Oslo or through its partners in the Global South (such as HISP Tanzania, an independent entity with close collaboration with Oslo) available in an “app store.” It is similar in concept to Apple App Store or Google Play and some DHIS2 apps are also available on these platforms too. The platform architecture allows local innovation as apps, increasing its potential relevance globally.

Due to its openness and flexibility, it is impossible to know the exact number of DHIS2 implementations. It is known that ministries of health and other organizations in more than 100 developing countries use DHIS2, together covering an estimated population of 2.4 billion people.  In November 2020, the ministries of health in 73 countries (primarily developing countries) used DHIS2, out of which 60 were nationwide implementations, and 13 were in the pilot stage. In addition, 22 Indian states used DHIS2. There is also a range of other organizations using DHIS2 independently for reporting in the countries they are operating, including PEPFAR, Médecins Sans Frontières (MSF), International Medical Corps, Population Services International (PSI), and Save the Children.

We can explore the “qualification” of DHIS2 as a public good by considering some of the challenges experienced by developers in Oslo and other implementation sites examined as tensions and paradoxes.  In a seminal paper on paradoxes and theory building, Poole and van de Ven (1989) identify a paradox as “concerned with tensions and oppositions between well-founded, well-reasoned, and well-supported alternative explanations of the same phenomenon” (565). 

Consider the story of the product lead of the DHIS2 analytics team response to the challenge of prioritizing requests by developing a roadmap prioritization matrix. Most use-cases need analytics functionality, and a wide variety of requests are directed to this team. The product lead estimates that the analytics team can only accommodate about half the requests at any stage of the product development cycle. The question facing this individual is: “which requests should be prioritized, coming from whom, and in which release cycle?” The primary implementations of DHIS2 are users from governments in low- and middle-income countries, according to the product lead, who tend to not actively voice their requests for changes in functionality. These groups are constrained by physical separation often across great distance, limiting ability to meet in person and develop social relationships. By contrast, users from donor organizations and other users in the West, tend to have closer proximity and resources to visit Oslo and “make their voices heard,” resulting in greater influence over the DHIS2 functionality development. This mismatch led the product manager to develop this “objective” prioritization methodology.  From the perspective of public goods, the dynamics of donors’ activity affects the rivalry / excludability conditions as their greater influence means that other users are relatively excluded, and access is rivalrous depending on this influence.

There are also paradoxical consequences of scaling at the macro and micro levels.  While the Oslo development team add in their releases of new features for strengthening outputs and analysis towards a generic global platform, the typical user in a district of a developing country requires basic functionalities, and the new features often detract instead of increasing the software’s value for the users.  At the macro-level, the development team are seeking to cater to the universe of users, including district users, researchers, and data analytic experts in multiple country contexts. This requires them to continuously add new features, often for increasingly sophisticated use. This process went counter to the needs at the micro-level of the users in district offices, who want specific and easy to use functionalities for their everyday use.  Thinking again from a theoretical standpoint, the malleability of a digital good compared to the oft cited example of a static lighthouse is clearly evident. The drive towards generic global features at the macro level causes rivalry and excludes some users at the local more micro level.  

Overall, the more macro interests of the donors and drive towards a global generic platform appear incompatible with the smaller players who become increasingly marginalized. Furthermore, their capacity for collective action is limited by structural factors.   This challenges DHIS2’s status as a public good as we can see rivalry and exclusion creeping in.

The problem is not insurmountable, collective action and subsidiarity offer helpful mechanisms of governance. Two main subsidiarity conditions are known to be helpful related to effectiveness and necessity: that action should be taken at the level where it is most effective and that action at the higher level should be taken when lower levels cannot achieve the set goals by themselves. This is in line with ongoing efforts by Oslo to build South-South community-based networks and thereby decentralization into the Health Information System Programme (HISP) network. 


Gawer, A. (2009). Platform dynamics and strategies: from products to services. Platforms, markets and innovation45, 57.

Nicholson, B., Nielsen, P., & Sæbø, J. (2021). Digital platforms for development. Inf. Syst. J.31(6), 863-868.

Digital Inequality Beyond the Digital Divide

How can we understand digital inequality in an era of digital inclusion?

As the open-access journal paper, Digital Inequality Beyond the Digital Divide: Conceptualising Adverse Digital Incorporation in the Global South” explains, the digital divide has been an essential and powerful concept that links digital systems with inequality.

But it is no longer sufficient.  A majority of the global South’s population now has internet access and is included in, not excluded from, digital systems.  Yet, as the figure below illustrates, that inclusion also brings inequalities – the small farmers in digital value chains losing out to large intermediaries; the gig workers whose value and data are captured by their platforms; the communities disempowered when they are digitally mapped.

Figure 1: From an Exclusion-Based to an Inclusion-Based Perspective on Digital Inequality

We need a new conceptualisation to explain this emerging pattern.  I refer to this as “adverse digital incorporation”, defined as inclusion in a digital system that enables a more-advantaged group to extract disproportionate value from the work or resources of another, less-advantaged group.

As shown below, I have inductively built a model of adverse digital incorporation, based around three aspects:

Figure 2. Conceptual Model of Adverse Digital Incorporation

Future digital development research can apply this model deductively to cases of digital inequality, and can further investigate the digitality of adverse digital incorporation. 

For digital development practitioners, the challenge will be to achieve “advantageous digital incorporation”: designing digital interventions that specifically and effectively reduce existing inequalities.  This means going beyond digital equity to digital justice: addressing the underlying and contextual causes of inequality not just its surface manifestations.

For further details, please refer to the paper; “Digital Inequality Beyond the Digital Divide: Conceptualising Adverse Digital Incorporation in the Global South”.