dan mcquillan

lecturer in creative & social computing

Dec 14, 2015

Science for Change Kosovo Year 1

[This reflection on the first year of Science for Change Kosovo, commissioned by Datashift and released under a CC-BY-SA license.]

Science for Change Kosovo (SfCK) is a radical citizen science project. I will try to explain what our kind of citizen science is and why it's important, and why we decided to do citizen science in Kosovo. I will also talk about our preliminary results, our plans for the next year or so, and what I think the wider implications are for data, communities and democracy.

what is citizen science?

There are many forms of citizen science; in most, the participants are simply collecting data or completing tasks for an experiment designed by and for scientists. I say that we're a radical project because we believe people from the communities should be involved at every stage, from framing the research questions to designing the data collection, analysing the data and interpreting the results. This makes us less like mainstream citizen science and more like the Public Lab's idea of civic science (‘Public Lab’). We are also inspired by the idea of environmental justice; the recognition that the impact of pollution is often worse for people who are already disadvantaged by society, which goes with a commitment to supporting them to do something about it. An environmental justice project which acts as a model for Science for Change Kosovo is Global Community Monitor (‘Global Community Monitor’), who train and support disempowered "fenceline" communities harmed by serious air pollution from industrial sources and whose concerns agencies and responsible corporations are ignoring. Our ethos, like Community Based Auditing (Tattersall) in Tasmania, is to be an experiential way for citizens to undertake their own disciplined inquiry into environmental issues affecting them, so that they can assert their rights and obligations as generators of valid knowledge and as agents of change.

why does it matter?

But why should anyone take any notice of bottom-up citizen science? How can it compete in any way with the sophisticated equipment of professional scientists, not to mention their years of training? In practical terms, citizen science can fill critical gaps in knowledge. Official air quality data is often sparse, coming from a limited number of fixed monitoring sites, and has to use mathematical models to fill in the gaps. Statutory data gathering is all about averages; it can't get down to the level of everyday lives and doesn't record the variable exposures of different people, the effects of their choices, or the related impacts on their health. Institutional science also lacks citizen participation and accountability and struggles to work with local knowledge and soft data. It loses out on useful insights, relying on objectivity and distance; a stance which gave science authority years ago but more and more leads to mistrust. At a higher level, the concept of post-normal science (Funtowicz and Ravetz) suggests that orthodox scientific method is badly adapted to situations that combine high risk with high uncertainty (such as climate change) and proposes it be enhanced by an 'extended peer review' that includes all those affected by an issue. As we'll see below, Science for Change Kosovo has already started to fill some of the local gaps in data, and is trying to put in to practice the idea of a participatory peer review.

why Kosovo?

Why, though, did we decide to try citizen science in Kosovo? There are certainly easier places to start, especially in terms of available resources, and Kosovo is faced by many other equally pressing challenges. One reason is that we were already in Kosovo, doing social innovation hackathons with the young people based on the Social Innovation Camp (‘Social Innovation Camp Kosovo’) model. These camps took back-of-the-envelope ideas for digital social change and turned them in to working prototypes over the course of a weekend. The camps were hosted by the UNICEF Innovations Lab, the young participants came from the local Peer Educators Network, and many of the coders were part of Free Libre Open Source Software Kosovo. To understand Science for Change Kosovo it's important to know that we had already built credibility on the ground, brought young people in to contact with social empowerment through DIY tools, and had working partnerships with these key local groups.

Another reason for starting citizen in Kosovo is that it's a very polluted country. The ageing lignite power plants are a major source of NO2 (nitrogen dioxide), SO2 (sulphur dioxide) and particulates (dust), and one of the power stations blew up just as we started the project (Bytyci). Some daily pollutant levels exceed EU and World Health Organisation limits by several times. Kosovo's own environmental protection agency says that current data on air quality levels is poor and incomplete, and there's a lack of capacity for environmental protection at a local level. In a poor country, it's hard even for statutory agencies to get the budget for maintenance and training, and all agencies have to deal with basic problems like power cuts. Air pollution in Kosovo causes hundreds of premature deaths and thousands of emergency hospital visits each year due to respiratory tract infections. Among the countries of the region, Kosovo has the worst health outcomes, ranking behind the rest of the region, in some cases dramatically, on indicators such as life expectancy, maternal death rates or infant and child mortality.

The constitution recognizes environmental protection as one of the principles on which the Republic of Kosovo is based. The Law on Air Protection (no. 2004/30) assigns responsibility for air quality and emissions indicators and sets obligations for protection. So there's a legal framework for accountability which we work with. Kosovo also has a long-standing political aspiration to join the EU; making substantial efforts to tackle pollution will be a condition for accession. Signing up to EU principles also means signing up to measures like the Aarhus Convention which sets out rights to environmental information and to participate in environmental decisions.

The focus of Science for Change Kosovo is young people. Although half the population is under 25, their current participation in decision-making at all levels is limited. The project appeals to those young people who are hungry for change and who are enthusiastic about the potential for participatory tech innovation. Because of our environmental justice principles we are also trying to involve marginalised communities, which in Kosovo includes Roma, Ashkali and Egyptians. There's a Roma community living in Plemetina, right at the base of the Kosova A and B power stations which are the core of Kosovo's air pollution, and one of our key contacts for year 1 of Science for Change was a Roma youth activist living in this community.

year 1

Our project began in June 2014 in Prishtina with a weekend co-design event at the Unicef Innovations Lab in Prishtina. Participants included young people from several parts of Kosova that had experienced severe environmental issues, including Plementina (under the polluting power stations), Prishtina (the capital city, downwind of the power stations and with heavy traffic pollution) and Drenas (near the Ferronickel plant). The participants shared experiences of pollution and quizzed experts on an environmental health panel. There were sessions on methods for air quality measurement, such as diffusion tubes, and we had a member of the Smart Citizen Kit team who introduced their Arduino-based citizen sensing device (‘Smart Citizen’) and trained the young people on how to install it and how to connect it to the online data platform. We discussed the way that science doesn't always have the kind of certainty about environmental impacts that it claims in in public, that there are a lot of disagreements inside science and a lot of arguments about what data is valid and what isn't. Participants planned for campaigning, drawing on global examples like the Arab uprisings and local examples like the student direct action that removed a corrupt Rector from the University of Prishtina. By the end of the weekend the action groups had agreed on a plan for air quality monitoring in Prishtina, Plemetina and Drenas.

One key thing we'd understood from researching other DIY air quality sensing projects is the importance of calibration. While sounding like a pretty unexciting notion, this is one of the main discontinuities between the smoothness of data projects and the awkwardness of material reality. So much of what passes as data journalism and data visualisation takes its data from existing sources, whereas a citizen science project is generating data from interactions with the physical world. And if this data is to be at all meaningful there should be a way to tie it to the material; to set a baseline that's been verified in the lab and in the field. Although the Smart Citizen Kits represented the positive maker-movement trend to open hardware sensing, they were a concern for us because they came without any calibration. So we decided to co-locate our kits with diffusion tubes, and use the fact that both measured NO2 to calibrate SCK data against the the tube analysis from the lab.

The field mobilisation of Science for Change Kosovo was, frankly, impressive. Kosovo can be a very frustrating place, where post-Communist institutional inefficiency overlaps with entrepreneurial corruption, and it's hard to get things done if you don't know the right people. By contrast, the young people in the project self-organised with the help of the UNICEF lab, installing Smart Citizen Kits in houses in each location (‘Deployment of Smart Citizen Kits’) and installing and collecting diffusion tubes ( (‘Deployment of Tubes in Drenas’) over three monthly cycles of data gathering. This included a large group of young Roma who installed and monitored the collection devices in Plemetina (‘Deployment of Tubes in Plemetina’).

Unfortunately, the data from the Smart Citizen Kits was difficult to use. While the datasheets for the sensors on the boards showed a linear log-log correlation between gas levels and the signal, in practice there were severe spikes in the data which blew holes in our ability to compare an averaged reading with the tubes. This was a real shame because the kits were our route to live data; emitting sensor readings every second, they were connected to the net and held out the prospect of a live pollution map, not to mention ideas around live campaigning (eg. triggering tweets to members of parliament each time the pollution exceeded EU levels). On the other hand we had significant readings from some of the NO2 tubes in the capital city, Prishtina. Through the dedicated field work of the volunteers, it looks like we've identified local hotspots which were missed by the government's data and which exceeed statutory limits by a large margin. We will be following these up in the next phase of data gathering.

next steps

Our next goal is to expand our measurement activities to particulates i.e. very small dust particles, which are categorised as PM10 (under 10 microns diameter) and PM 2.5 (under 2.5 microns diameter). Particulates are a form of pollution where the link to serious and deadly health problems is absolutely unambiguous. We have acquired a semi-professional portable detector called the TSI Sidepak which will enable us to take readings at different locations and also on the move. In this way we'll be able to compare the exposure of different activities e.g. driving / walking / cycling, and the way this varies over the course of daily life for different groups of people.

Working with the Sidepak detector to make localised and journey-based PM measurements will enable us to test ameliorative measures, such as alternative walking routes that can reduce peoples' exposure. In doing this we're following the Breathe London project and the work of the air quality & health team at Kings College, London, who have been piloting alternative back street walking routes to school for children in areas of central London.

From December 2015 we'll be running junior citizen science workshops in high schools. Rather than the traditionalist pedagogy that's customary in Kosovo, these will be non-formal, experiential and practice-based workshops. The students will also get to experiment with with the Smart Citizen Kits, as a way to learn about open tech for environmental monitoring. They will deploy the Smart Citizen Kits for indoor air quality measurements in the schools.

data & change

This doesn't mean we're ignoring the need for wider campaigning. The current PM measurements in each locality are tied to holding 'Town Hall' meetings shortly afterwards as a way to inform and engage local people. We are exploring how our citizen science measurements can best be used for advocacy and campaigning, but this needs to be alert to the complexities of the local context. The data we generated from Drenas has already been used to report in the Parliamentary Commission, and the Ministry of Environment of Kosovo has initiated a court case against the heavy metal plant “Ferronikeli” in Drenas. However, these developments are heavily enmeshed in party political battles, which generates a certain amount of cynicism in everyone else. All institutions are perceived to be captured by narrow political interests, and NGOs are often seen as internationalised 'do nothings' who complain endlessly from the sidelines. We're hoping to learn from the EcoGuerilla campaign (‘Lëvizja Eco Guerilla’) from neighbouring Macedonia, where leaked information about pollution from a power plant led to mass protests against PM levels.

We are trying to avoid the attribution of agency to data, or an assumption that participation is the same as empowerment. Many data and open data projects in the wealthy West seem to assume that action will inevitably flow from aggregating data and visualising transparency. Other citizen sensing projects assume that participation of communities in gathering data will increase people's sense of responsibility and lead to the generation of solutions. But the idea that collective measurement leads to collective action seems questionable. In fact there may be a tendency for forms of governmentality like the smart city to re-constitute populations as having a duty to measure their environments, while at the same time producing a society that is, overall, less democratic.

It's the relationship between air quality and democracy which underlies Science for Change. Kosovo is democratically challenged, with different forms of corruption alongside political interference in knowledge production. Orthodox political processes are completely captured by elites and oligarchs. The older generation have a hold on power and it's very hard for young people with a more open, socially progressive outlook to make headway. The emphasis of Science for Change Kosovo is on practices, not simply on what is produced; on the construction of data and what that means about subjectivity and agency, not just on the data as such. The interesting thing about air quality is that it is also politics by other means. We know the air is political, that "the air’s chemical composition reveals a history and a politics in itself" (Nieuwenhuis). Most of the time we do not feel this with an intensity that leads to action. What is it in citizen science that causes an 'affective' response, which 'so amplifies our awareness of the injury which activates it that we are forced to be concerned, and concerned immediately'? (Tomkins and Demos) This is a question we hope that Science for Change Kosovo will help to answer over the next two years.


Bytyci, Fatos. ‘At Least One Killed in Kosovo Power Plant Blast, Supplies Hit’. Yahoo News. N.p., 6 June 2014. Web. 26 Nov. 2015. http://news.yahoo.com/explosion-hits-kosovo-coal-fired-power-plant-injuries-095441081.html.

‘Deployment of Smart Citizen Kits | Facebook’. N.p., 2014. Web. 26 Nov. 2015. https://www.facebook.com/media/set/?set=a.1451488458465993.1073741832.1423629027918603&type=3.

‘Deployment of Tubes in Drenas | Facebook’. N.p., 2014. Web. 26 Nov. 2015. https://www.facebook.com/media/set/?set=a.1448248975456608.1073741831.1423629027918603&type=3.

‘Deployment of Tubes in Plemetina | Facebook’. N.p., 2014. Web. 26 Nov. 2015. https://www.facebook.com/media/set/?set=a.1440501616231344.1073741830.1423629027918603&type=3.

Funtowicz, Silvio O., and Jerome R. Ravetz. ‘Science for the Post-Normal Age’. Futures 25.7 (1993): 739–755. ScienceDirect. Web.

‘Global Community Monitor’. 2015. Web. 26 Nov. 2015. http://www.gcmonitor.org/.

‘Lëvizja Eco Guerilla’. 2015. Web. 26 Nov. 2015. http://ecoguerilla.mk/.

Nieuwenhuis, Marijn. ‘Atemwende, or How to Breathe Differently’. Dialogues in Human Geography March (2015): 90–94. Print.

‘Public Lab: About Public Lab’. N.p., 2015. Web. 26 Nov. 2015. https://publiclab.org/about.

‘Smart Citizen’. N.p., 2015. Web. 26 Nov. 2015. https://smartcitizen.me/kits/.

‘Social Innovation Camp Kosovo’. 2013. Web. 17 Aug. 2014. http://sicampkosovo.org/.

Tattersall, Philip J. ‘What Is Community Based Auditing and How Does It Work?’ Futures 42.5 (2010): 466–474. ScienceDirect. Web.

Tomkins, Silvan S., and E. Virginia Demos. Exploring Affect: The Selected Writings of Silvan S Tomkins. Cambridge University Press, 1995. Print.