Measuring Barriers to Big Data for Development

How can we measure the barriers to big data for development?  A research paper from Manchester’s Centre for Development Informatics suggests use of the design-reality gap model.

Big data holds much promise for development: to improve the speed, quality and consistency of a wide variety of development decisions[1].  At present, this is more potential than actuality because big data initiatives in developing countries face many barriers[2].

But so far there has been little sense of how these barriers can be systematically measured: work to date tends to be rather broad-brush or haphazard.  Seeking to improve this, we investigated use of an ICT4D framework already known for measurement of barriers: the design-reality gap model.

In its basic form the model is straightforward:

  • It records the gap between the design requirements or assumptions of big data vs. the current reality on the ground.
  • The gap is typically recorded on a scale from 0 (no gap: everything needed for big data is present) to 10 (radical gap: none of the requirements for big data is present).
  • The gap can be estimated via analysis of researchers, or derived directly from interviewees, or recorded from group discussions.
  • It is typically measured along seven “ITPOSMO” dimensions (see below).

As proof-of-concept, the model was applied to measure barriers to big data in the Colombian public sector; gathered from a mix of participant-observation in two IT summits, interviews, and secondary data analysis.
WP62 Graphic v2

 

As summarised in the figure above, the model showed serious barriers on all seven dimensions:

  • Information: some variety of data but limited volume, velocity and visibility (gap size 7).
  • Technology: good mobile, moderate internet and poor sensor availability with a strong digital divide (gap size 6).
  • Processes: few “information value chain” processes at work to put big data into action (gap size 7).
  • Objectives and values: basic data policies in place but lack of big data culture and drivers (gap size 7).
  • Skills and knowledge: foundational but not specialised big data capabilities (gap size 7).
  • Management systems and structures: general IT systems and structures in place but little specific to big data (gap size 7).
  • Other resources: some budgets earmarked for big data projects (gap size 5).

A simple summary would be that Colombia’s public sector has a number of the foundations or precursors for big data in place, but very few of the specific components that make up a big data ecosystem.  One can turn around each of the gaps to propose actions to overcome barriers: greater use of existing datasets; investments in data-capture technologies; prioritisation of value-generation rather than data-generation processes; etc.

As the working paper notes:

“Beyond the specifics of the particular case, this research provides a proof-of-concept for use of the design-reality gap model in assessing barriers to big data for development. Rephrasing the focus for the exercise, the model could equally be used to measure readiness for big data; BD4D critical success and failure factors; and risks for specific big data initiatives. …

We hope other researchers and consultants will make use of the design-reality gap model for future assessments of big-data-for-development readiness, barriers and risks.”

For those interested in taking forward research and practice in this area, please sign up with the LinkedIn group on “Data-Intensive Development”.

[1] Hilbert, M. (2016) Big data for development, Development Policy Review, 34(1), 135-174

[2] Spratt, S. & Baker, J. (2015) Big Data and International Development: Impacts, Scenarios and Policy Options, Evidence Report no. 163, IDS, University of Sussex, Falmer, UK

The Power Dynamics of Big and Open Data

At a recent CDI brown-bag discussion on data-intensive development, we hypothesised a mirror-image power dynamic between big data and open data.

Big Open Data Few Many

Open data has an inherent tendency to redistribute power from the few (who originally hold the data) to the many (who can now access the data).  It supports sousveillance.  Big data has an inherent tendency in the opposite direction.  It gathers data about the many but only the few have the power to capture, store, process, interpret and use that big data.  It supports surveillance.

The extent to which these are inherent affordances of these data systems vs. the extent to which these tendencies are inscribed into those data systems is a matter for further debate.  But what it does suggest is that big data per se is more reproductive than transformative of power inequalities within society.  Think of the way in which major users of big data – social media platforms, e-business multinationals, telecommunication companies – operate.  Their uses of big data reinforce inequality much more than they challenge it.

One way to address this is to reverse the power dynamic flow shown above: big data must become open data.  This could happen in various ways:

  • Big data as open data: big datasets are made openly available online in accessible format (as in all cases, with due consideration for data privacy and security).
  • Big data as shared data: big datasets are made available to particular organisations (e.g. those of civil society).
  • Big data as small data: sub-sets of big datasets are shared with the sources of that data for their use (e.g. the particular communities or groups from which the big data derived).

Reversing Big Data Inequalities

But what will make a reversal happen?  To understand this, we need to study open data motivations: what causes organisations to open their datasets?  Reviewing our knowledge of open data, we could not find examples of intrinsic motivations driving adoption of open data.  Instead, drivers to opening of big datasets seem likely to be extrinsic:

  • For public sector owners of big data, domestic political economy (e.g. local campaigns for access to data; economic benefits from creation of a local data economy) and external political economy (e.g. encouraging foreign investment through a reputation for openness).
  • For private sector owners of big data, government regulation to force opening of datasets, or shareholder/consumer pressure.

Without such extrinsic pressures and the openness that ensues, big data may not deliver its developmental potential.

 

A Research Agenda for Data-Intensive Development

18 July 2016 1 comment

In practice, there is a growing role for data within international development: what we can call “data-intensive development”.  But what should be the research agenda for this emerging phenomenon?

On 12th July 2016, a group of 40 researchers and practitioners gathered in Manchester at the workshop on “Big and Open Data for Development”, organised by the Centre for Development Informatics.  Identifying a research agenda was a main purpose for the workshop; particularly looking for commonalities that avoid fractionating our field by data type: big data vs. open data vs. real-time data vs. geo-located data, etc; each in its own little silo.

IMG_0828

A key challenge for data-intensive development research is locating the “window of relevance”.  Focus too far back on the curve of technical change – largely determined in the Western private sector – and you may fail to gain attention and interest in your research.  Focus too far forward and you may find there no actual examples in developing countries that you can research.

In 2014 and 2015, we had two failed attempts to organise conference tracks on data-and-development; each generating just a couple of papers.  By contrast, the 2016 workshop received two dozen submissions; too many to accommodate but suggesting a critical mass of research is finally starting to appear.

It is still early days – the reports from practice still give a strong sense of data struggling to find development purposes; development purposes struggling to find data.  But the workshop provided enough foundational ideas, emergent issues, and reports-back from pilot initiatives to show we are putting the basic building blocks of a research domain in place.

But where next?  Through a mix of day-long placing of Post-It notes on walls, presentation responses, and a set of group then plenary discussions[1], we identified a set of future research priorities, as shown below and also here as PDF.

DID Research Agenda

 

 

The agenda divided into four sub-domains:

  • Describing/Defining: working out the basic boundaries, contours and contents of the data-intensive development domain.
  • Practising: measuring and learning from the practice of data-intensive development.
  • Analysing: evaluating the impact of data-intensive development through various analytical lenses.
  • Resisting: guiding practical actions to challenge potential state and corporate data hegemony in developing countries.

Given the size and eclectic mix of the group, many different research interests were expressed.  But two came up much more than others.

First, power, politics and data-intensive development: analysing the power structures that shape DID initiatives, and that are inscribed into data systems; analysing the way in which DID produces and reproduces power; analysing what resistance to data hegemony would mean.

Second, justice, ethics, rights and data-intensive development: determining what a social justice perspective on DID would mean; analysing what DID can contribute to rights-based development; understanding how ethical principles would guide civil society interventions for better DID.

We hope, as a research community, to take these and other agenda items forward.  If you would like to join us, please sign up with the LinkedIn group on “Data-Intensive Development”.

 

[1] My thanks to Jaco Renken for collating these.

The Politics of Disconnection: Network Geography, Trump, Sanders, Brexit, et al

Disconnection

Due to advances in transport and digital infrastructure, we live in an increasingly-connected world.  The value of global flows rose from US$5tr in 1990 to US$30tr in 2014[1].  In the same period, international travel grew from 435m to 1.1bn per year.

But this global interconnection – and the economic crash that was its direct result – has led to a powerful counter-reaction, with challenger politics emerging from both right and left.  The figureheads in the global North are various and sometimes curious: Trump, Sanders, Farage, Iglesias, Tsipras, Le Pen, Hofer, and more.  While differing in many policies, they share common ground that boils down to the slogan, “Disconnect!”.

Examples of insurgent policies include:

  • Disconnection from human networks through anti-immigration initiatives.
  • Disconnection from governance networks such as leaving the EU or abandoning free trade agreements.
  • Disconnection from production networks through support for localised production, and disincentives to globalised production.
  • Disconnection from – or at least restrictions on – capital networks through tax and other financial controls.
  • Disconnection from geo-political networks through increasing reticence for overseas military intervention.

There are many other policy examples: British disconnection from international development networks; French disconnection from the euro; etc.

Who is this coming from?  Setting aside the catalysis and aspirations of individual leaders, there are differences but also similarities between the demographics of those disconnecting from the right and those disconnecting from the left[2].  Right-wing disconnectors tend to be older, poorer, less-well-educated; left-wing disconnectors the reverse. But they appear to have two things in common: they are more often from the ethnic majority, and they are more often men.

We can understand these people in terms of positional network geography (see earlier discussion).  Rarely excluded from key global networks, instead these are people who perceive themselves – or can be persuaded to perceive themselves – as adversely incorporated, peripheralised in those networks.  They see a network core that benefits at their expense; they see new, mobile members seeking to join their network and potentially displace them.  For those who are white men perhaps there is particularly a gap between the promise or expectation of benefitting from the growth of global networks, and a perceived reality of not doing so.

As the complexity of the networks into which we are connected grows, and as the number of our network connections grows, we become increasingly connected into contexts that are too complex to either understand or control.  Yet we demand that our politicians control these uncontrollable networks.  And this takes place in an environment of growing digital politics in which form matters more than content.

Combine these two and we encourage the confident assertion of simple solutions: on the right, disconnecting from global flows of labour; on the left, disconnecting from global flows of capital; both disconnecting from global governance networks.

This is reminiscent of the disconnections of the 1920s following the shock of the First World War.  Remind me, how did that work out?

[1] MGI (2016) Digital Globalization, McKinsey Global Institute, San Francisco, CA

[2] http://www.newstatesman.com/politics/2015/02/si-we-can-how-left-wing-podemos-party-rattling-spanish-establishment; https://www.quora.com/Whats-the-demographic-profile-of-a-Bernie-Sanders-supporter; https://yougov.co.uk/news/2016/03/24/eu-referendum-provincial-england-versus-london-and/; http://www.theatlantic.com/politics/archive/2016/03/who-are-donald-trumps-supporters-really/471714/

Network Geography and Global Development: Dependency Redux?

6 June 2016 1 comment

The SDGs can be taken as a marker of transition from international development to global development worldviews.  Among other aspects of global development is a universalisation of development: development is everywhere; not just associated with the global South.  What does this mean for the geography of development?

One response would be to retain a physical, spatial approach to geography, but to move on from the old bipolarity of developed vs. developing; North vs. South.  Alongside moving upward to the global, this would move downward to the local; for example to regions, cities, “pockets of poverty” and the like[i].

An alternative would be to move to a network geography of development.  Social networks and transportation links mean networks have always been fundamental to development.  But telecommunication links – particularly digital links during the 21st century – have significantly accelerated the presence and salience and complexity of networks in development.  These exist as physical networks (such as physical infrastructure grids), virtual networks (such as online communities of practice) and most often as hybrid networks (such as supply chains and their parallel digital representations), so a network geography combines the physical and non-physical.  The explosive growth of networks in development demands greater use of network-based conceptualisations, including network geographies[ii].

Three main geographies can be applied to understand these networks.  First, a processual geography that focuses on the flows between nodes in the network (e.g. flows of aid between networks of development NGOs[iii]).  Second, a structural geography that focuses on the shape of the network (e.g. the impact of different network structures on water governance in Costa Rica[iv]).  Third, a relational geography that combines aspects of both flow and shape (e.g. the resource networks drawn on by development champions[v]).

Conceptualisation of these network geographies has come from a number of sources[vi].  Examples include social capital drawing from new institutional economics, global production networks drawing from economic geography, networked governance drawing from political science, embeddedness drawing from new economic sociology, and complex adaptive systems drawing from complexity theory.

However, if the spatial geography of old is to be supplanted, it will be by a new spatial geography; one that replaces position in the physical world with position in the network (physical, virtual, hybrid).  The positional network geography of development has often used the binary of being either inside or outside the network (e.g. Castells’ notion of the excluded “Fourth World”, or conceptions of the digital divide).  More sophisticated versions have added the category of “have lesses” between the “haves” and the “have nots”; those who are within the network but at the periphery.  Given, at least for digital networks, the dwindling numbers who are truly excluded, this is a more appropriate conceptualisation.  The positional geography of development thus becomes a geography of network position: distinguishing those actors at the core of the network from those at the periphery who are marginal and precarious.

While this might be an idea suitable for the 21st century nature of global development, it has earlier origins.  It sounds very similar to dependency theory, with its ideas of core and periphery.  Though long out of favour, this could provide one approach to a positional network geography of development; a revival supported by some in development studies[vii].

Aspects of dependency theory relevant to a wider network geography of development include:

  • Moving beyond the simple binary of core/periphery to world systems theory’s core/semi-periphery/periphery; or even to the idea of a spectrum of network positions.
  • Associating network position with differentiated roles vis-a-vis the production and capture of value, and with differentiated flows and ownership of resources.
  • Recognising the reproduction of network position through power; particularly the power of innovation, knowledge and technology.

A positional network geography of development would need to move away from dependency’s nation-state-centric approach, recognising many other units of analysis; and it would need to recognise the (constrained) potential for mobility of network position.  Work on global production networks has taken some of these ideas and demonstrated their relevance to another unit of analysis, but this needs to be extended to all forms of networks; not just global but regional and local; not just productive but political and social.  Indeed, one would need to recognise that any development actor lies not within a single network but within multiple networks; potentially with somewhat different positions in each.  These would include locally-embedded as well as disembedded networks.

Further developments needed include:

  • Recognition of the relational, institutional and cognitive/symbolic sources of power within the network; and the potential for network-specific conceptions of power[viii].
  • Recognition of the role played by the new technologies that increasingly mediate, enable and constrain the networks of global development; requiring some socio-materiality to be incorporated[ix].
  • Recognition of the increasing potential for quantification of positionality via social network analysis; a tool which can add absent methodological rigour[x].

 

[i] Horner, R. (2016) Unpacking the emergence of global development, unpublished draft

[ii] Bebbington, A. & Kothari, U. (2006) Transnational development networks, Environment and Planning A, 38(5), 849-866

[iii] Bebbington, A. (2004) NGOs and uneven development, Progress in Human Geography, 28(6), 725-745

[iv] Kuzdas, C., Wiek, A., Warner, B., Vignola, R., & Morataya, R. (2015) Integrated and participatory analysis of water governance regimes, World Development, 66, 254-268

[v] Renken, J. & Heeks, R. (2013) Conceptualising ICT4D project champions, paper presented at ICTD2013, Cape Town, 7-10 Dec

[vi] Heeks, R. & Renken, J. (2015) Investigating the potential of social network analysis in development studies, paper presented at DSA 2015 conference, Bath, 7-8 Sep

[vii] E.g. Fischer, A.M. (2015) The end of peripheries? On the enduring relevance of structuralism for understanding contemporary global development, Development and Change, 46(4), 700-732

[viii] E.g. Castells, M. (2011). A network theory of power, International Journal of Communication, 5, 773-787

[ix] E.g. Contractor, N., Monge, P. & Leonardi, P.M. (2011) Multidimensional networks and the dynamics of sociomateriality: bringing technology inside the network, International Journal of Communication, 5, 682-720

[x] Heeks, R. & Renken, J. (2015) Investigating the potential of social network analysis in development studies, paper presented at DSA 2015 conference, Bath, 7-8 Sep

Open vs. Closed Institutional Logics in Open Development Projects

“Open development” is a concept with some momentum in the ICT4D field, encouraged particularly by support from IDRC[1].  A core challenge has been theorisation of open development, and here I briefly propose and test the idea that institutional logics can offer such a foundation.

As noted in an earlier blog entry, “institutional logics are broad social forces with both material and symbolic elements that shape the way we think and act.  Religion, family, state, and market are typical logics but running through digital development is a conflict between two other logics:

  • Open logic: a cooperative logic that values openly-accessible inputs, participative and collaborative processes, and shared distribution of benefits.
  • Closed logic: a competitive or controlling logic that values restriction of inputs, processes and benefits to particular individuals or groups.”

We can understand these ideas better by applying them to a real open development ICT4D case; selecting here the iDART system – open source software developed in South Africa by Cell-Life to help pharmacists dispense anti-retroviral drugs to those with HIV/AIDS.  The case has been written up by Melissa Loudon and Ulrike Rivett[2], and is here reinterpreted through a logics lens.

Cell-Life originated as an inter-university collaboration in Cape Town and, as such, has been heavily influenced by the institutional logics that operate within academic organisations.  Universities can be understood as sites of conflict between open and closed institutional logics; with the latter traditionally dominant but the former finding voice.

Examples of the constitution of the two institutional logics and their material (resources, processes, structures) and symbolic (culture) elements are shown in the table below, drawn from the case study[3].

  Open Logic Closed Logic
Resources Open source technologies

 

Freely-accessible data and content

Proprietary technologies

 

Restricted data and content

Processes Inclusive production of knowledge and technology

 

Student-centred learning

Exclusive production of knowledge and technology

 

Didactic teaching

Structures Unbounded peer-to-peer, multi-disciplinary networks Mono-disciplinary silos
Culture Universities seen as learning and action research environments

 

Academics seen as facilitators

Universities seen as ivory tower storehouses of knowledge

 

Academics seen as experts

 

With Cell-Life an enclave of open logic within a wider context of closed logic, conflict between the two logics was inevitable.  Examples include:

  • System development processes: system developers with a background of closed processes encountered with some difficulty very different open logic imperatives within the project.
  • Intellectual property rights: the university’s approach to software – proprietary IP that would be commercialised to the benefit of the university – conflicted with the open source approach underpinning Cell-Life’s work.
  • Software market: direct rivalry occurred between open-source iDART software and competing proprietary pharmacy management software.

There were also conflicts over the closed focus on disciplinary silos vs. the open logic of multi-disciplinary action research.

When organisational logics conflict, there are a number of potential outcomes including “decoupling”, “compromise”, and “selective coupling”[4].  In this case, two main outcomes were seen:

  1. Compromise: a hybrid approach that combines aspects of both open and closed logics. System development processes were neither completely open nor closed, but a mix of the two.  Users were involved through feedback on prototypes but the Cell-Life team retained control over the development process, often acting as proxies for users and acting as overall custodians of the system.  Some but not all user revision requests were incorporated.
  2. Protected Niche: Cell-Life created a protected niche of open logic, with barriers created against closed logic. After five years within the university system, Cell-Life was spun-off as a non-profit entity, thus increasing the structural barriers and distance to the dominant closed logic of the university system.  The software itself was developed to focus particularly on low-resource, rural pharmacies; a market niche not targeted by closed-logic-based commercial vendors.

What can we conclude?

First, that the idea of open vs. closed institutional logics is applicable to open development projects.  Institutional logics offers a new language; a new way to describe and explain what has happened on the project.  From this brief analysis, it’s not clear what new insights it provides beyond this; but that may be the nature of this post-hoc, external reinterpretation.  There is certainly a case for pre-hoc application of institutional logics – definitely, to analyse open development; likely, to analyse ICT4D more broadly – to help describe the outcome of conflicting logics; to explain when one logic dominates another; to understand how to deal with conflicting logics in practice; and to identify the role of open development/ICT4D champions as institutional entrepreneurs.

Second, and assuming we could generalise the idea of conflict between open and closed logics, this suggests that achieving true “open development” may be very difficult: there will always be pressure to hybridise.  But not only might fully-open development be unfeasible, it might also be undesirable.  Indeed, both extremes might be undesirable: closed development because it leads to inequality and exclusion, open development because it leads to disincentives to action and potentially-ineffective or chaotic outcomes[5].

One can see the latter in the case study, which restricts the openness of processes not only because of the pressures of closed logic, but also for reasons of project effectiveness and efficiency.  The evidence base here is just preliminary, but could suggest – assuming most development systems lean further to the closed than open end of the spectrum – that the objective should be “more-open development” rather than “fully-open development”.

[1] Reilly, K.M.A. & McMahon, R. (2015) Quality of Openness, IDRC, Ottawa

[2] Loudon, M. & Rivett, U. (2013) Enacting openness in ICT4D research, in: Open Development, M.L. Smith & K.M.A. Reilly (eds), MIT Press, Cambridge, MA, 53-77; an earlier version available as: Loudon, M. & Rivett, U. (2011) Enacting openness in ICT4D research, Information Technologies & International Development, 7(1), 33-46; some details from Rivett, U. & Tapson, J. (2009) The Cell-Life Project: converging technologies in the context of HIV/AIDS, Gateways, 2, 82-97

[3] See also Lounsbury, M. & Pollack, S. (2001) Institutionalizing civic engagement: shifting logics and the cultural repackaging of service-learning in US higher education, Organization, 8(2), 319-339

[4] See Nicholson, B., Malik, F., Morgan, S. & Heeks, R. (2015) Exploring hybrids of commercial and welfare logics in impact sourcing, , in: Openness in ICT4D, P. Nielsen (ed.), Department of Informatics, University of Oslo, Norway, 78-91; which draws on Pache, A.-C. & Santos, F. (2013) Inside the hybrid organization: selective coupling as a response to competing institutional logics, Academy of Management Journal, 56(4), 972-1001

[5] See, e.g., Heeks, R. (2015) The curse of hyper-transparency, ICT4DBlog, 27 Feb; and Dahlander, L. & Gann, D.M. (2010) How open is innovation?Research policy, 39(6), 699-709

Policies for Inclusive Innovation

29 April 2016 3 comments

For inclusive innovation to prosper, new policy initiatives are needed, argues the working paper “Policies to Support Inclusive Innovation”.  Inclusive innovation is the means by which new goods and services are developed for and by marginal groups (the poor, women, the disabled, ethnic minorities, etc).  It is central to addressing inequality in society, and thus increasingly on the agenda of global development actors.  (See here for a more detailed explanation of inclusive innovation.)

But inclusive innovation suffers from a series of failures – of innovation development, design, diffusion and use – that provide a rationale for policy intervention:

  • Formal innovators focus insufficiently on the poor.
  • Informal actors are delinked from innovation systems.
  • Those serving peripheral markets have weak adaptive capacity.
  • Low-income users lack capability to use innovations effectively.
  • Underlying policies and context are weak or absent.

These can then be flipped directly into five main inclusive innovation policy objectives:

  1. Orient Formal Innovation Systems Towards the Poor. Measures will include creating new innovation partnerships, supporting local innovative research, and reducing risk by providing market incentives for inclusive innovation.
  2. Promote Grassroots Innovators. Measures will include linking grassroots actors into formal innovation systems e.g. via intermediaries, and incentivising development and diffusion of grassroots innovations.
  3. Improve Absorptive Capacity of Low-Income Groups. Measures will include building skills to absorb and adapt innovations that meet the needs of marginalised groups, and supporting innovation hubs and clusters.
  4. Drive More Effective Use of Innovations among Low-Income Groups. Measures include supply-side actions to accelerate affordability of innovations, and demand-side actions to build the skills and knowledge necessary for effective use of innovations.
  5. Reduce Structural Barriers to Inclusive Innovation. Measures include altering government regulations that exclude or are biased against low-income actors, including altering sourcing rules.

However, such policy measures will only be enacted if there are broader policy changes.  This firstly means changing the worldview of policy makers so that they understand there is an important two-way connection between innovation and social inclusion; that marginalised actors are both consumers and producers of inclusive innovation.  It also means creating “Inclusive Innovation Policy Collaboratories” that bring together a wide range of stakeholders, and which adopt an experimental and iterative approach to the policy measures outlined above.

The diagram below provides an overview summary of inclusive innovation policy background and recommendations.

Inclusive Innovation Policy Overview

Follow

Get every new post delivered to your Inbox.

Join 923 other followers

%d bloggers like this: