Parsha: Difference between revisions
Line 91: | Line 91: | ||
Topic: Testing and developing methods, practices and tools for effective and coherent information flow between researchers, policy makers, and stakeholders. | Topic: Testing and developing methods, practices and tools for effective and coherent information flow between researchers, policy makers, and stakeholders. | ||
2. Rationale | 2. Rationale | ||
Basic idea and merit of the project | Basic idea and merit of the project | ||
Parsha project will take several important environmental and health policy issues | Parsha project will take several important environmental and health policy issues under scrutiny and study available scientific knowledge but also political values and options, and stakeholder concerns and beliefs. We will start from climate policies in Helsinki and disease burden of air pollution in Finland. We will use existing and novel open methods for information synthesis and modelling. Open participation and mutual learning is ensured by large collaboration networks of researchers, policy-makers, and non-governmental organisations. The project will develop a new and effective way of resolving disputes and sharing information about these resolutions. This may improve the way we perform science and policy. | ||
Climate change policy actions need synthesis | Climate change policy actions need synthesis | ||
The city of Helsinki is strongly committed to climate policy by both reducing greenhouse gas emissions and adapting to a changing climate. Recently, we performed a review of important strategy papers and found more than 600 planned climate actions. A major challenge is to make sense out of this diverse pool, select the most effective actions, evaluate their other impacts and desirability among different groups, and implement them coherently. Making a full impact assessment of all actions would be ideal but is not feasible. There is a clear need to develop practices to organise information about extensive topics with large written materials, difficult scientific questions, high political stakes, and widely differing opinions and values. | The city of Helsinki is strongly committed to climate policy by both reducing greenhouse gas emissions and adapting to a changing climate. Recently, we performed a review of important strategy papers (Tuomisto et al. 2014) and found more than 600 planned climate actions. A major challenge is to make sense out of this diverse pool, select the most effective actions, evaluate their other impacts and desirability among different groups, and implement them coherently. Making a full impact assessment of all actions would be ideal but is not feasible. There is a clear need to develop practices to organise information about extensive topics with large written materials, difficult scientific questions, high political stakes, and widely differing opinions and values. | ||
An example of a critical scientific dispute is the actual climate impact of biofuels. If biofuels are found as bad in Finland as a fresh US study demonstrates , this resolution seriously alters the preference order of several policy actions. A useful information system would identify this as a critical issue and tell, how the conclusions would change if one or another conclusion is made. The same need applies to non-scientific disputes about values as well. | An example of a critical scientific dispute is the actual climate impact of biofuels. If biofuels are found as bad in Finland as a fresh US study demonstrates (DeCicco et al. 2016), this resolution seriously alters the preference order of several policy actions. A useful information system would identify this as a critical issue and tell, how the conclusions would change if one or another conclusion is made. The same need applies to non-scientific disputes about values as well. | ||
Air pollution disease burden is a methodological and policy challenge | Air pollution disease burden is a methodological and policy challenge | ||
Many environmental issues are health issues as well. During the recent years, several assessments have been performed about health impacts of fine particles, the major pollutant in Finland as well as globally ( | Many environmental issues are health issues as well. During the recent years, several assessments have been performed about health impacts of fine particles, the major pollutant in Finland as well as globally (Hänninen & Knol 2011, Asikainen et al. 2013, Lelieveld et al. 2015, Brauer et al. 2016). However, at the same time, the whole method used to calculate premature deaths caused by air pollution has been challenged, and the discussion is still ongoing (Morfeld, Héroux REFs). The discussion is very much about the detailed methods, mathematics, and interpretations of concepts, being far too complex for any policy maker to follow. | ||
Again, there is a need for an information system that can be used to organise the knowledge and produce an understandable, policy-relevant synthesis that is consistent with the methodological discussion and conclusions. Also, the conclusions should be reflected in the actual mathematical tools assessors are using in their health impact assessments. Parsha project is based on an existing system, Opasnet web-workspace, that partly offers these functionalities but that requires several steps of development and research to fulfil this need. | Again, there is a need for an information system that can be used to organise the knowledge and produce an understandable, policy-relevant synthesis that is consistent with the methodological discussion and conclusions. Also, the conclusions should be reflected in the actual mathematical tools assessors are using in their health impact assessments. Parsha project is based on an existing system, Opasnet web-workspace, that partly offers these functionalities but that requires several steps of development and research to fulfil this need. | ||
Line 114: | Line 111: | ||
Resource description framework (RDF) is a system developed by W3C, an organisation for standardising the Internet (REF about W3C##). RDF can describe rich spaces of information, e.g. contents of an encyclopedia, in a systematic way by defining items and relations that describe properties of the items (REF ## about RDF or Wikidata). RDF is extensively used in e.g. defining the contents of Wikipedia using the Wikidata RDF database. The database enables rich queries of the content. Such properties are needed also for describing complex policy issues. | Resource description framework (RDF) is a system developed by W3C, an organisation for standardising the Internet (REF about W3C##). RDF can describe rich spaces of information, e.g. contents of an encyclopedia, in a systematic way by defining items and relations that describe properties of the items (REF ## about RDF or Wikidata). RDF is extensively used in e.g. defining the contents of Wikipedia using the Wikidata RDF database. The database enables rich queries of the content. Such properties are needed also for describing complex policy issues. | ||
Directed acyclic graphs (DAGs) are an effective graphical way to describe items and their relations. They are extensively used in many disciplines, but they are especially useful in e.g. describing causal relations of knowledge crystals ( | Directed acyclic graphs (DAGs) are an effective graphical way to describe items and their relations. They are extensively used in many disciplines, but they are especially useful in e.g. describing causal relations of knowledge crystals (Pearl 2009). If these relations are estimated as conditional subjective probabilities, the information in the system can be updated using Bayes’ rule, and the system is called a Bayesian belief network (BBN). | ||
RDF and BBN are becoming more commonly used, but they have not been used together to describe complex policy situations in such a way that scientific issues, values and disputes would have all been described in a single, coherent system. In addition, pragma-dialectic argumentation theory gives rules for resolving disputes (van Eemeren | RDF and BBN are becoming more commonly used, but they have not been used together to describe complex policy situations in such a way that scientific issues, values and disputes would have all been described in a single, coherent system. In addition, pragma-dialectic argumentation theory gives rules for resolving disputes (van Eemeren & Grootendorst 2004). Our innovation is that the essence of these resolutions can be described using RDF and thus take argumentation as an integral part of the system. This combination of novel techniques is unique and gives promises of important breakthroughs. Whether this combination works as expected, will be studied during the project. | ||
Although we describe causal systems and people’s reasoning in the same information system, our aim is not to build artificial intelligence that solves policy problems automatically. Rather, the objective is to build an information system where all relevant information - including political discussion - can be organised, synthesised, criticised and made readily available in useful format for policy making. | Although we describe causal systems and people’s reasoning in the same information system, our aim is not to build artificial intelligence that solves policy problems automatically. Rather, the objective is to build an information system where all relevant information - including political discussion - can be organised, synthesised, criticised and made readily available in useful format for policy making. | ||
Open practices are needed in science. PUT THIS FIRST?#The society has changed, and the legitimacy of science requires openness and discussion with citizens, not just one-directional informing of scientific wonders. ( | Open practices are needed in science. PUT THIS FIRST?#The society has changed, and the legitimacy of science requires openness and discussion with citizens, not just one-directional informing of scientific wonders. (Väliverronen 2016a & 2016b). | ||
There is a lot of current activity related to open publishing of articles; open data (Open linked data##); policy relevance of scientific information ( | There is a lot of current activity related to open publishing of articles; open data (Open linked data##); policy relevance of scientific information (Jussila 2012, Klemola ym. 2014, Raivio 2014, Vihriälä 2016, VNK 2011); research focussed on rapid societal utility (Strategic Research Council of the Academy; Prime Minister’s Office’s research program VN-TEAS); citizen science; public participation and co-creation in policy development; and experimentation of different policies m(Tietokäyttöön##). Between and connected to all of these, there exists an ecologic niche for this project: there is an urgent need to study how information should and could flow in the society and produce a consistent, comprehensive approach that is able to capture important ideas, values, and causal understanding related to different activities in the society. It must also be able to distinguish which relevant claims can and which cannot be defended based on science, bearing in mind that there are also other than scientific worldviews in the society. | ||
Open assessment takes two steps toward shared understanding TOO LONG | Open assessment takes two steps toward shared understanding TOO LONG | ||
In our previous work with open assessment, we have already shown that its information structures are effective and applicable in impact assessments and policy analyses. (Pohjola et al State of the art, Tuomisto: Urgenche case##, Avoin PTK ##) A central idea is knowledge crystal (Tuomisto | In our previous work with open assessment, we have already shown that its information structures are effective and applicable in impact assessments and policy analyses. (Pohjola et al State of the art, Tuomisto: Urgenche case##, Avoin PTK ##) A central idea is knowledge crystal (Tuomisto 2015) a distinct web page that has a clear research question and that aims to answer it by co-creating a synthesis of scientific data. Importantly, knowledge crystals are re-usable objects and they are expected to develop in time when they are used in new impact assessments. Knowledge crystals are also effective means to communicate, as they have a plain-text summary in the beginning and go into more and more technical details, data, and analysis code for experts in the end. This approach enables an information structure, where all information relevant to a particular research question is located in one place. This is an improvement to the prevalent scientific publishing system, where information is published in static articles in a fragmented way with limited error correction functionalities (Tragedy of error##). | ||
However, impact assessments have two major limitations, both of which will be tackled in this project. First, performing an assessment requires a lot of work to synthesise scientific information into a quantitative causal description of policy-relevant issues. Therefore, it is often not available in the time frame of rapid political decisions. Second, assessments typically focus on causal chains with established scientific knowledge, thus leaving many important aspects untouched, because solid scientific data is not available or the aspects are inherently based on values rather than facts. | However, impact assessments have two major limitations, both of which will be tackled in this project. First, performing an assessment requires a lot of work to synthesise scientific information into a quantitative causal description of policy-relevant issues. Therefore, it is often not available in the time frame of rapid political decisions. Second, assessments typically focus on causal chains with established scientific knowledge, thus leaving many important aspects untouched, because solid scientific data is not available or the aspects are inherently based on values rather than facts. | ||
Line 146: | Line 143: | ||
Therefore, a major research question is this: is it possible to develop a guidance for describing exhaustive shared understanding, and can that guidance be followed in practical policy situations? | Therefore, a major research question is this: is it possible to develop a guidance for describing exhaustive shared understanding, and can that guidance be followed in practical policy situations? | ||
Basic scientific concepts may be unfamiliar to citizens or even university students, preventing learning of related facts. An effective way to avoid this problem is refutational text that specifically aims to identify and explicate typical misconceptions. ( | Basic scientific concepts may be unfamiliar to citizens or even university students, preventing learning of related facts. An effective way to avoid this problem is refutational text that specifically aims to identify and explicate typical misconceptions. (Södervik 2016). Shared understanding is a systematic approach to produce refutational texts about policy issues. | ||
3. Objectives and expected results | 3. Objectives and expected results | ||
3 A Objectives of the research | 3 A Objectives of the research | ||
Parsha project will study several policy cases relevant for environment and health, review scientific knowledge and synthesise political discussion. All this information is described as shared understanding of the cases. The bold aim is to produce such a comprehensive and well-structured description that most political and scientific aspects raised can be reproduced from there. Such description would enable understanding and learning what and why should be decided, given the valuations of a particular participant. If this aim is achieved, shared understanding may become a key method to produce and disseminate policy-relevant information. This may result in positive feedback loop of improved understanding, when citizens learn to use the method to demand clarity in policy, and experts and policy-makers learn to use it to offer clarity and rationale for decisions. | |||
Parsha is based on i) a social innovation of hearing every viewpoint systematically without aiming at consensus; ii) an information science innovation of a structured synthesis from which viewpoints can be reproduced and discrepancies analysed; and iii) a technological innovation of an interface to facilitate participation, knowledge retrieval, learning, and policy support. Our hypothesis is that scientific knowledge prevails in this process because of its internal consistency, while vague and flawed reasoning is identified as such and turned down. | |||
Parsha is based on i) a social innovation of hearing every viewpoint systematically without aiming at consensus; ii) an information science innovation of a structured synthesis from which viewpoints can be reproduced and discrepancies analysed; and iii) a technological innovation of an interface to facilitate participation, knowledge retrieval, learning, and policy support. Our hypothesis is that scientific knowledge prevails in this process. | |||
Objectives and related research questions: | Objectives and related research questions: | ||
Substantive: To produce shared understanding and policy guidance for the city of Helsinki climate mitigation policies and strategies, accounting for health, climate and other impacts. | Substantive: To produce shared understanding and policy guidance for the city of Helsinki climate mitigation policies and strategies, accounting for health, climate and other impacts; fine particle disease burden and related policies; and other case studies. | ||
Can the guidance for Helsinki be extended or generalised to the whole of Finland? | Can the guidance for Helsinki be extended or generalised to the whole of Finland? | ||
Are Finnish fine particle policies effective? | Are Finnish fine particle policies effective? | ||
Participatory: To develop practices that are tempting and rewarding to policy-makers, experts, and stakeholders so that participation in the production of shared understanding is seen useful. | |||
What factors increase or hinder participation? | |||
Are people motivated to participate and produce meaningful content? | |||
Can researchers produce a common “scientific perspective” using participation rules? | |||
Can the work be crowdsourced and in what conditions? | |||
Methodological: To test, implement, and further develop the method of shared understanding. | Methodological: To test, implement, and further develop the method of shared understanding. | ||
Can shared understanding be described as an information system? What essential parts does it have? | Can shared understanding be described as an information system? What essential parts does it have? | ||
Line 164: | Line 164: | ||
How can conflicting opinions be systematically described to everyone’s satisfaction? | How can conflicting opinions be systematically described to everyone’s satisfaction? | ||
What tools are needed to combine slow modelling and rapid political discussion? | What tools are needed to combine slow modelling and rapid political discussion? | ||
Can shared understanding be produced without compromising scientific integrity? | |||
Technological: To produce web-based tools for creating and managing descriptions of shared understanding. | Technological: To produce web-based tools for creating and managing descriptions of shared understanding. | ||
Are people willing to participate using such an interface? | Are people willing to participate using such an interface? | ||
What problems do the current typical interfaces have, and how can they be improved? | What problems do the current typical interfaces have, and how can they be improved? | ||
Are the developed technologies scalable to much wider use? | Are the developed technologies scalable to much wider use? | ||
3 B Effects and impact beyond academia | 3 B Effects and impact beyond academia | ||
Shared understanding would offer methods to improve the knowledge base of policies. Methods are likely to find support from the society, as Parsha goes along several societal megatrends (openness, participation, co-creation, evidence-based policy). It is likely to alleviate confrontation, because everyone is better heard than before and political benefits of shouting louder decrease. In some cases, popular political opinions and options would not hold against systematic criticism, reducing the risk that the actual decisions turn out bad to the society. | |||
This is actually nothing more than the scientific method applied to policy questions. There is strong evidence that the scientific method is a very powerful and effective tool to produce solid information. This is no news. However, the scientific and societal merit of the Parsha project is that it has identified important areas in the society where this method is not satisfactorily applied, and has produced innovations that can be rigorously studied in policy-relevant situations. It should be noted that the scientific method is not ideally used in the field of science either. Therefore, we also study how to produce and publish scientific information in a more criticisable and applicable way. | |||
If successful, we can learn new things about how science and policy can be performed together in a more effective and robust way by using shared information systems and practices. The societal impacts of Parsha may extend far from the substantive topics studied in the cases. This is facilitated by maximising openness and the potential for re-use of its information content. It may even change the way we think about scientific publishing: continuous participatory improving of the description of our understanding, rather than producing distinct, static objects called articles. | |||
3 C Publication plan | 3 C Publication plan | ||
Openness is a core principle in Parsha. All work is done online, and all scientific data and information is produced into standardised information objects, enabling their effective re-use. This is called passive openness (enabling but not promoting the flow of information). Naturally, traditional research articles are also published, always in open access journals. To illustrate: this very research plan was written on a public website, and we will publish it in the open access journal Research Ideas and Outcomes after it has been submitted to the Academy of Finland. | |||
Openness is a core principle in Parsha. All work is done online, and all scientific data and information is produced into standardised information objects, enabling their effective re-use. This is called passive openness (enabling but not promoting the flow of information). Naturally, traditional research articles are also published, | |||
In Parsha, we will also do active openness by promoting the information produced and case studies planned or performed. We will also actively seek to learn from other information providers, as exemplified by the research visits to the GovLab and RIVM, and by inviting contributions to the web-workspace. This will be done through the large networks of the participants | In Parsha, we will also do active openness by promoting the information produced and case studies planned or performed. We will also actively seek to learn from other information providers, as exemplified by the research visits to the GovLab and RIVM, and by inviting contributions to the web-workspace. This will be done through the large networks of the collaborators and participants, and by actively utilising workshops and social media, e.g. blogs, Facebook groups, and Twitter. | ||
4. Research methods and material, support from research environment | 4. Research methods and material, support from research environment | ||
Methods | Methods | ||
Line 223: | Line 188: | ||
Workshops, experiments literature reivews | Workshops, experiments literature reivews | ||
Science collected. Modelling | Science collected. Modelling | ||
Voiko tehdä kokeen: otetaan useita keisejä joista määritellään miten niitä pitäisi jäsentää ja millaisia tuloksia voi olla. Sitten arvotaan ne joihin lähdetään mukaan. Lopussa katsotaan oliko osallistumosella vaikutusta vai ei. | |||
Haasteena tulee olemaan uskottavien koeasetelmien rakentaminen. | |||
Tämä kai on normaiivista tutkimusta. Mutta pitäisikö olla myös deskriptiivis? Mitä se olisi? Katsotaan miten ihmiset käyttäytyvät? Seurataan kun he käyttävät työkaluja? | |||
Koska hanke on vasta vuoden päässä ei kannata liian monta keisiä kuvata liian tarkasti. Vaan kerrotaan millä oeruaatteilla niitä valitaan. Ja mitä niissä tutkitaan. | |||
Materials | Materials | ||
Trials: same information presented to the audience in two different ways randomly. Are there differences? (How to measure?) | Trials: same information presented to the audience in two different ways randomly. Are there differences? (How to measure?) | ||
Research environment | Research environment | ||
THL and its environmental health units in Kuopio is a highly skilled research environment. We have previously performed research and assessments on related topics (Asikainen et al, 2013; Hänninen and Knol, 2011; Tuomisto et al, 2014, Pohjola, 2014, Neittaanmäki, 2016). The technical infrastructure is up-to-date, and we have many statisticians and modellers to consult within the institute if needed. The project as a whole consists of a wide variety of expertise from government to | |||
Critical points and risk management | Critical points and risk management | ||
In theory, Parsha is simply applying established principles in new kind of information and policy support work. In practice, we will run into problems related to complexity, vagueness, power, and compliance. Risks are reduced by wide collaboration, openness (“if you fail, do it early and learn from it”), and reliance on robust open source technology and tested methods. There are several alternative techniques available, so dead-ends are unlikely. Even if the methodology as a whole fails to mature and become directly applicable during the Parsha project, it will undoubtedly teach us a lot about how information flows between science and policy and what factors hinder the flow. This is already a valuable result. From substantive point of view, there is a lot of scientific information available about climate policies, fine particles, disease burden methods, and other relevant topics. This is not a limiting factor, but rather the methodological capability to organise and synthesise it. Fortunately our team has expertise and skills to handle exactly this kind of topics. | |||
The risk of failure can be managed by limiting the scope of work. We aim at generic methods and results with wide applicability, but the width or complexity of material may prevent progress. Then we will focus on one particular aspect or policy issue at a time, and when methods produce tangible results, step back and analyse problems seen in the generic case. | |||
5. Ethical issues | 5. Ethical issues | ||
Parsha project does not study patients or collect private data. Ethical permissions are not needed. Participation is voluntary and based on informed consent. THL guidance on good research practice will be followed. | |||
6. Implementation: schedule, budget, distribution of work | 6. Implementation: schedule, budget, distribution of work | ||
Schedule for the research | Schedule for the research | ||
EMEMBER THE INTERFACE WORK: WHO DOES IT? IT EXPERTISE NEEDED? BUDGET? | |||
Budget | Budget | ||
THL 600248 (420173 € from Academy + 180075 € own funding) | THL 600248 (420173 € from Academy + 180075 € own funding) | ||
Line 277: | Line 231: | ||
Case studies: | Case studies: | ||
The city of Helsinki has more than 600 action points in various climate strategies and other policy papers. For only a few of them, a systematic evaluation with cost estimates and climate impacts have been done. There is a clear need to organise the information, disputes, roles and actions into a coherent description. In other words: Shared understanding is needed. This will be produced using consultation of target groups within the Helsinki administration and also NGOs and other stakeholders who have opinions about the actions. | The city of Helsinki has more than 600 action points in various climate strategies and other policy papers. For only a few of them, a systematic evaluation with cost estimates and climate impacts have been done. There is a clear need to organise the information, disputes, roles and actions into a coherent description. In other words: Shared understanding is needed. This will be produced using consultation of target groups within the Helsinki administration and also NGOs and other stakeholders who have opinions about the actions. | ||
There are several recent disease burden estimates of air pollution for both Finland and globally. However, there are scientific disputes about the best epidemiological methods (Morfeld REF) and scoping for assessments | There are several recent disease burden estimates of air pollution for both Finland and globally. However, there are scientific disputes about the best epidemiological methods (Morfeld REF) and scoping for assessments. These issues will be clarified by producing shared understanding on both the method (already an conference abstract: Tuomisto et al 2016) and the situation in Finland. This will be based on work and results of currently on-going Akatemia funded project BATMAN, where estimates of disease burden caused by particulate matter in Finland are updated with spatial and source specific exposure data. Furthermore, in BATMAN the effect of several emission mitigation policies are evaluated and ranked. Based on the results of BATMAN and earlier national projects Seturi and ISTE, a shared understanding on common disease burden methodology is produced. This is reached by evaluating how the used disease burden methodology and selected boundaries of assessments effects on importance of particulate matter as an environmental hazard in Finland and whether the conclusions are affected by methodological differences or not. --# : Should we describe the participatns in THL? --Jouni (talk) 21:01, 20 September 2016 (UTC) | ||
Vaccination coverage and acceptability is an issue where researchers say that the benefits of several vaccines are indisputable; yet they are disputed in the society. We will produce a shared understanding of this issue and identify the main arguments and values that contradict. It will also be a test about wether people agree to participate in a process with people who have strongly opposing views about personally very important issues. We may also learn how well viewpoints can be described based on previously written material only, without actual participation. | Vaccination coverage and acceptability is an issue where researchers say that the benefits of several vaccines are indisputable; yet they are disputed in the society. We will produce a shared understanding of this issue and identify the main arguments and values that contradict. It will also be a test about wether people agree to participate in a process with people who have strongly opposing views about personally very important issues. We may also learn how well viewpoints can be described based on previously written material only, without actual participation. | ||
New case studies will be decided later in the project, because it would be impossible to predict, which topics will have policy relevance after three years. | New case studies will be decided later in the project, because it would be impossible to predict, which topics will have policy relevance after three years. | ||
Line 283: | Line 237: | ||
Many information structures already exist for open assessments and open policy practice (Tuomisto and Pohjola 2007, Pohjola 2013). Based on our experience, those structures work well in impact assessments, which are mainly systematic, science-based quantitative modelling exercises describing causal relations. However, they are not flexible enough for describing non-causal inferences, heuristics and value judgements important in people's opinions about policy issues. There is a need for more flexible information structures. | Many information structures already exist for open assessments and open policy practice (Tuomisto and Pohjola 2007, Pohjola 2013). Based on our experience, those structures work well in impact assessments, which are mainly systematic, science-based quantitative modelling exercises describing causal relations. However, they are not flexible enough for describing non-causal inferences, heuristics and value judgements important in people's opinions about policy issues. There is a need for more flexible information structures. | ||
We have piloted with promising results a resource description framework using triple approach. Triples have a subject, a predicate (a relation between items) and an object describing something about the subject. With a small number of different relations we have successfully described complex environmental health issues. So, we will expand the use of this approach and see how it works in the case studies and how the participants can understand own and other people's viewpoints based on triple descriptions. | We have piloted with promising results a resource description framework using triple approach. Triples have a subject, a predicate (a relation between items) and an object describing something about the subject. With a small number of different relations we have successfully described complex environmental health issues. So, we will expand the use of this approach and see how it works in the case studies and how the participants can understand own and other people's viewpoints based on triple descriptions. | ||
In addition, we will experiment pragma-dialectical argumentation theory developed in the University of Amsterdam (van Eemeren, Grootendorst). It has been implemented in several fields ⇤# : Where, describe a few examples. --Jouni (talk) 21:01, 20 September 2016 (UTC), but not | In addition, we will experiment pragma-dialectical argumentation theory developed in the University of Amsterdam (van Eemeren, Grootendorst). It has been implemented in several fields ⇤# : Where, describe a few examples. --Jouni (talk) 21:01, 20 September 2016 (UTC), but not previously in producing statements for shared understanding. | ||
Another topic for investigation is a method to incorporate differing views into a single description. We anticipate that most of the time it will be possible to describe different views with the same | Another topic for investigation is a method to incorporate differing views into a single description. We anticipate that most of the time it will be possible to describe different views with the same statements but using different qualities or truth values. For example, anti-vaccine activists and pediatricians agree that "vaccines are dangerous to children" is a meaningful statement and relevant in the context; the dispute comes from the fact that the former group thinks it is true while the latter group thinks it is false. Interestingly, both groups may agree on a related statement: "If vaccines are dangerous to children, they should not be used." It is useful to see that the difference is in premises, not the reasoning. Although such an example may seem trivial, we think that in many political cases similar clarifications are not made and the precise reasons for disputes are not understood. Another use of such an information structure is that it raises issues like "What do we actually mean by 'dangerous'?", thus directing further discussions into directions that improve shared understanding. | ||
Workpackage 3: Technical development | Workpackage 3: Technical development | ||
We already have an open web-workspace Opasnet, where assessors can perform impact assessments. However, technical development is needed to include a triple database and other functionalities for non-causal reasoning and statements and expressions of value judgements. This is necessary to operationalise the work produced in WP2. | We already have an open web-workspace Opasnet, where assessors can perform impact assessments. However, technical development is needed to include a triple database and other functionalities for non-causal reasoning and statements and expressions of value judgements. This is necessary to operationalise the work produced in WP2. | ||
Line 295: | Line 249: | ||
Second, RIVM (The Environment and Health Institute in the Netherlands) has extensive expertise in policy support and the science-policy interface. | Second, RIVM (The Environment and Health Institute in the Netherlands) has extensive expertise in policy support and the science-policy interface. | ||
7. Research team and collaborative partners | 7. Research team and collaborative partners | ||
Participants and their special interests (with funding): | Participants and their special interests (with funding): | ||
THL: shared understanding method, disease burden estimates. All researchers are experts in environmental health, impact assessment, disease burden, and quantitative modelling. Jouni Tuomisto: chief researcher | THL: shared understanding method, disease burden estimates. All researchers are experts in environmental health, impact assessment, disease burden, and quantitative modelling. Jouni Tuomisto: chief researcher ##PITÄISI KIRJOITTAA MEISTÄ ESITTELYT…. Otto Hänninen: senior researcher … Arja Asikainen: researcher… Päivi Meriläinen: resarcher … N.N.: information technology expert ... | ||
Collaborators and their special interests (with letter of commitment but no direct funding from the Academy) | Collaborators and their special interests (with letter of commitment but no direct funding from the Academy) | ||
City of Helsinki (Environment Center): climate change mitigation and adaptation policy and open data: Jari Viinanen: graduated as M.Sc in field of energy and environment from Lappeenranta University of Technology. Since 2002 he has been working as environmental inspector in City of Helsinki Environment Centre. He is a specialist of urban climate change mitigation and adaptation and has worked in climate change projects developing strategies and writing reports. Jari Viinanen has also done air quality and waste management expert work in relation to air quality programmes and street dust research. | City of Helsinki (Environment Center): climate change mitigation and adaptation policy and open data: Jari Viinanen: graduated as M.Sc in field of energy and environment from Lappeenranta University of Technology. Since 2002 he has been working as environmental inspector in City of Helsinki Environment Centre. He is a specialist of urban climate change mitigation and adaptation and has worked in climate change projects developing strategies and writing reports. Jari Viinanen has also done air quality and waste management expert work in relation to air quality programmes and street dust research. | ||
Line 310: | Line 262: | ||
8. Research careers, fulfilment of the mobility requirement and researcher training | 8. Research careers, fulfilment of the mobility requirement and researcher training | ||
Advancing the research career of the applicant or other researchers to be funded | Advancing the research career of the applicant or other researchers to be funded | ||
9. Mobility plan for the funding period | 9. Mobility plan for the funding period | ||
10. Bibliography | 10. Bibliography | ||
Asikainen A, Hänninen O, Pekkanen J (2013). Ympäristöaltisteisiin liittyvä tautitaakka Suomessa. Ympäristö ja Terveys-lehti 5: 68–74. http://www.julkari.fi/handle/10024/110739 | |||
Brauer M, Freedman G, Frostad J, et al.van Donkelaar A, Martin RV, Dentener F, Van Dingenen R, Estep K, Amini H, Apte JS, Balakrishnan K, Barregard L, Broday DM, Feigin V, Ghosh S, Hopke PK, Knibbs LD, Kokubo Y, Liu Y, Ma S, Morawska L, Sangrador JLT, Shaddick G, Anderson HR, Vos T, Forouzanfar MH, Burnett RT, Cohen A. (2016). Ambient air pollution exposure estimation for the Global Burden of Disease 2013. Environ. Sci. Technol. 50: 79–88. doi: 10.1021/acs.est.5b03709. | |||
DeCicco, J.M., Liu, D.Y., Heo, J. et al (2016). Carbon balance effects of U.S. biofuel production and use. Climatic Change 138: 667. doi:10.1007/s10584-016-1764-4) | |||
Hänninen O, Knol A (eds.) (2011). European perspectives on Environmental Burden of Disease; Estimates for nine stressors in six countries. THL Reports 1/2011, Helsinki, Finland. ISBN 978-952-245-413-3. http://www.thl.fi/thlclient/pdfs/b75f6999-e7c4-4550-a939-3bccb19e41c1 | |||
Jussila H. (2012). Päätöksenteon tukena vai hyllyssä pölyttymässä? Sosiaalipoliittisen tutkimustiedon käyttö eduskuntatyössä. Sosiaali- ja terveysturvan tutkimuksia 121, Helsinki. | |||
Klemola K., Uusi-Illikainen J., Askola T. (2014). Tiedolla johtamisen käsikirja - Julkisrahoitteiset palvelut. Sitra, Helsinki. ISBN 978-951-563-888-5 (PDF) | |||
Lelieveld J, Evans J.S., Fnais M., Giannadaki D., Pozzer A (2015). The contribution of outdoor air pollution sources to premature mortality on a global scale. Nature 525: 367–371. doi:10.1038/nature15371 | |||
Neittaanmäki P, Timo Huttula, Juha Karvanen, Tom Frisk, Jouni Tuomisto, Antti Simola, Tero Tuovinen, Janne Ropponen. Unicorn–Open science for assessing environmental state, human health and regional economy. Research Ideas and Outcomes 2: e9232 (16 May 2016) doi: 10.3897/rio.2.e9232 | |||
Pearl, J. (2009). Causality: Models, Reasoning, and Inference. 2nd edition. Cambridge University Press New York, NY, USA.. ISBN:052189560X 9780521895606 | |||
Pohjola M: Assessments are to change the world – Prerequisites to effective environmental health assessment. Doctoral dissertation. THL, 2013. http://urn.fi/URN:ISBN:978-952-245-883-4 | |||
Raivio K. (2014). Näyttöön perustuva päätöksenteko – suomalainen neuvonantojärjestelmä. Valtioneuvoston kanslian raporttisarja 3/2014. ISBN 978-952-287-135-0 (PDF). | |||
Södervik I. (2016). Understanding biological concepts at university – Investigating learning in medical and teacher education. Annales Universitatis Turkuensis ser. B 421, Turku. | |||
Tuomisto JT. (2015). Massadata kansanterveyden edistämisessä. Duodecim 131:2179–87. | |||
Tuomisto JT., Rintala J., Tuomisto M., Orden P. (2014). Helsingin ohjelmalliset energiatehokkuus- ja ilmastotoimenpiteet ja -tavoitteet. http://fi.opasnet.org/fi/Helsingin_ohjelmalliset_energiatehokkuus-_ja_ilmastotoimenpiteet_ja_-tavoitteet | |||
Van Eemeren F.H., Grootendorst R. (2004). A Systematic Theory of Argumentation: The Pragma-dialectical Approach. Cambridge: Cambridge University Press. | |||
Vihriälä V. (ed) (2016). Taloustutkimus päätöksenteon tukena. Etla, Helsinki. ISBN 978-951-628-672-6 | |||
VNK (2011). Poliittisen päätöksenteon tietopohjan parantaminen - tavoitteet todeksi. Politiikkatoimien vaikuttavuusarvioinnin kehittämistyöryhmän raportti. Valtioneuvoston kanslian julkaisusarja 8/2011. Helsinki. ISBN 978-952-5896-61-9 (PDF) | |||
Väliverronen E. (2016a). Julkinen tiede [Public Science]. Vastapaino. ISBN 978-951-768-537-5. | |||
Väliverronen E. (2016b). Tiede tarvitsee avoimuutta. Tieteessä tapahtuu 5: 1-2. | |||
}} | }} |
Revision as of 11:22, 27 September 2016
This page is a encyclopedia article.
The page identifier is Op_en7744 |
---|
Moderator:Jouni (see all) |
This page is a stub. You may improve it into a full page. |
Upload data
|
Parsha is a research project applied from the Academy of Finland in September 2016. Its main objective is to test and implement the method of shared understanding on pressing environmental and health issues and other policy relevant problems. The method of shared undestanding is closely related to open assessment (a scientific method to evaluate impacts of policy decisions) and open policy practice (guidance to evaluate and manage a decision support process that involves an open assessment). However, the method of shared understanding (link goes to a Finnish description), is especially focussing on developing a description of issues, statements, and values presented by participants of a complex political decision situation. This goes beyond the description of scientific knowledge (which is the aim of open assessment) and incorporating that into an "official" decision making process (which is the aim of open policy practice). The method of shared understanding aims to understand and describe also aspects and values that researchers say are wrong or decision makers say are irrelevant or unfavourable. The purpose of such an exercise is to understand, discuss, and mediate societal opinions that may lead to controversies, political opposition, or even conflicts.
Abstract
Open data and practices are becoming common in the society. This trend has also brought problems: new internet tools enable to distribute also false and even malevolent information and to distort policy making. Scientific policy support practices are not well equipped to tackle this challenge. An example is prevalent quasi-scientific climate scepticism.
Shared understanding is a situation, where participants know, what the important issues are and where there are agreements and disagreements and why. In Parsha project it is considered a prerequisite for rational, slow thinking and societal policy making. ----#: . Minusta kannattaa tarkentaa sitä, mitä tarkoitetaan "slow thinking" ja mitä hyötyä siitä on. Päätöksentekoa halutaan yhteiskunnassa kuitenkin periaatteessa nopeuttaa ja sujuvoittaa eikä hidastaa, tästä voi tulla vähän väärä käsitys. Tästähän on tarkoitus tulla menetelmä todelliseen yhteiskunnalliseen päätöksentekoon. --Signatiu (talk) 12:04, 23 September 2016 (UTC) (type: truth; paradigms: science: comment)The main objective of Parsha is to test and implement methods and tools for producing shared understanding on pressing environmental and health issues and other policy-relevant problems.
Shared understanding will be produced in several policy-relevant, controversial situations, starting with climate change policies in Helsinki and disease burden disputes about air pollution. New topics will be chosen for the latter part of the project based on future needs. The methods to be used are based on open assessment and open policy practice, which have been developed and successfully used by our team and which evaluate impacts of future policy decisions using scientific information. In Parsha, the focus will additionally be on systematically describing and analysing values and statements not necessarily based on science.
Parsha is based on i) a social innovation of hearing every viewpoint systematically without aiming at consensus; ii) an information science innovation of a structured synthesis from which viewpoints can be reproduced and discrepancies analysed; and iii) a technological innovation of an interface to facilitate participation, knowledge retrieval, learning, and policy support. Our hypothesis is that scientific knowledge prevails in this process.
Parsha will extensively use participation and co-creation in its studies and actively share its results. It will use experiments to test performance of the innovations: whether they actually can collect, synthesise and describe information from participants to their satisfaction; whether the synthesis is informative and correctly interpreted by others; whether the disputes can be identified and their impact described; and whether such analyses actually help policy makers avoid emotional, fast thinking. If they do, many complex environmental and health problems may come closer to a solution.
Public summary
In English
The trend of open data and practices has also problems: internet tools help distribute false information and distort policy making. Scientific policy support needs better practices.
We test and implement methods for producing shared understanding on policy-relevant problems. Experimental case studies include science-based open assessments about climate change policies in Helsinki and disease burden disputes about air pollution. They will be augmented by systematically describing and analysing values and statements by stakeholders.
Methods are based on i) hearing every viewpoint systematically without aiming at consensus; ii) a structured synthesis, from which viewpoints can be reproduced and discrepancies analysed; and iii) an interface to facilitate participation, learning, and policy support. Hypothesis: scientific knowledge prevails in this process, helping policy makers avoid emotional, fast thinking. If it does, many environmental and health problems come closer to a solution.
In Finnish
Avoimuustrendissä on myös ongelmia: internetin työkalut auttaa levittämään virheellistä tietoa ja vääristämään yhteiskunnan päätöksentekoa.
Me testaamme ja sovellamme menetelmiä jaetun ymmärryksen tuottamiseen poliittisista kysymyksistä. Kokeelliset tapaustutkimukset liittyvät mm. tutkimuspohjaisiin avoimiin arviointeihin ilmastopolitiikoista Helsingissä ja kiistoihin ilmansaasteiden tautitaakasta. Näitä täydennetään kuvaamalla ja analysoimalla systemaattisesti sidosryhmien väitteitä ja arvostuksia.
Menetelmät perustuvat i) kaikkien näkökulmien kuuntelemiseen ilman pyrkimystä yhteisymmärrykseen; ii) jäsennettyyn synteesiin, josta näkökulmat voidaan toisintaa ja epäjohdonmukaisuuksia tutkia; ja iii) käyttäjärajapintaan, joka tukee osallistumista, oppimista ja päätöstukea. ----#: . Jatkaisin: ja lopputuloksena tämä auttaa päätöksentekoa tavalla x, vaikka yhteisymmärrystä ei synnykään (tarvitsee selittää, miksi yhteisymmärrystä ei tässä tarvita -> päätöksentekijä arvottaa lopulta). --Signatiu (talk) 12:07, 23 September 2016 (UTC) (type: truth; paradigms: science: comment)Hypoteesi: tieteellinen tieto hallitsee tätä prosessia auttaen päättäjiä välttämään tunnepohjaista, nopeaa ajattelua. Jos niin käy, monet ympäristö- ja terveysongelmat lähestyvät ratkaisua.
Description of tasks
The duration of Parsha project is four years, starting from September 1, 2017. The work is divided into the following workpackages and tasks.
Workpackage 1: Participation
This workpackage develops the social innovation of hearing and documenting all viewpoints. It is very much organised based on co-creation of case studies. The work is facilitated by collaborators. ----#: . Can we add Future Earth Suomi, OKFFI, and others here? --Jouni (talk) 21:01, 20 September 2016 (UTC) (type: truth; paradigms: science: comment)
Case studies:
- The city of Helsinki has more than 600 action points in various climate strategies and other policy papers. For only a few of them, a systematic evaluation with cost estimates and climate impacts have been done. There is a clear need to organise the information, disputes, roles and actions into a coherent description. In other words: Shared understanding is needed. This will be produced using consultation of target groups within the Helsinki administration and also NGOs and other stakeholders who have opinions about the actions.
- There are several recent disease burden estimates of air pollution for both Finland and globally. However, there are scientific disputes about the best epidemiological methods (Morfeld REF) and scoping for assessments (Seturi vs ISTE?). These issues will be clarified by producing shared undeerstanding on both the method (already an conference abstract: Tuomisto et al 2016) and the situation in Finland ----#: . Should we describe the participatns in THL? --Jouni (talk) 21:01, 20 September 2016 (UTC) (type: truth; paradigms: science: comment)
- Vaccination coverage and acceptability is an issue where researchers say that the benefits of several vaccines are undisputable; yet they are disputed in the society. We will produce a shared understanding of this issue and identify the main arguments and values that contradict. It will also be a test about wether people agree to participate in a process with people who have strongly opposing views about personally very important issues. We may also learn how well viewpoints can be described based on previously written material only, without actual participation.
- New case studies will be decided later in the project, because it would be impossible to predict, which topics will have policy relevance after three years.
Workpackage 2: Information structures
Many information structures already exist for open assessments and open policy practice (Tuomisto and Pohjola 2007, Pohjola 2013). Based on our experience, those structures work well in impact assessments, which are mainly systematic, science-based modelling exercises describing causal relations. However, they are not flexible enough for describing non-causal inferences, heuristics and value judgement important in people's opinions about policy issues. There is a need for more flexible information stuctures.
We have piloted with promising results a resource description framework using triple approach. Triples have a subject, a predicate (a relation between items) and an object describing something about the subject. With a small number of different relations we have successfully described complex environmental health issues. So, we will expand the use of this approach and see how it works in the case studies and how the participants can understand own and other people's viewpoints based on triple descriptions.
In addition, we will experiment pragma-dialectical argumentation theory developed in the University of Amsterdam (van Eemeren, Grootendorst). It has been implemented in several fields ⇤--#: . Where, describe a few examples. --Jouni (talk) 21:01, 20 September 2016 (UTC) (type: truth; paradigms: science: attack), but not previeously in producing statements for shared understanding.
Another topic for investigation is a method to incorporate differing views into a single description. We anticipate that most of the time it will be possible to describe different views with the same stataments but using different qualitiers or truth valuess. For example, anti-vaccine activists and pediatricians agree that "vaccines are dangerous to children" is a meaningful statement and relevant in the context; the dispute comes from the fact that the former group thinks it is true while the latter group thinks it is false. Interestingly, both groups may agree on a related statement: "If vaccines are dangerous to children, they should not be used." It is useful to see that the difference is in premises, not the reasoning. Although such an example may seem trivial, we think that in many political cases similar clarifications are not made and the precise reasons for disputes are not understood. Another use of such an information structure is that it raises issues like "What do we actually mean by 'dangerous'?", thus directing further discussions into directions that improve shared understanding.
Workpackage 3: Technical development
We already have an open web-workspace Opasnet, where assessors can perform impact assessments. However, technical development is needed to include a triple database and other functionalities for non-causal reasoning and statemements and expressions of value judgements. This is necessary to operationalise the work produced in WP2.
However, it is also important to enhance the user experience. It should be able to easily document own viewpoints and link them to other relevant information already in the system. Also, search functionalities, linkages of related items, and at least some automatic or semi-automatic reasoning based on given premises and statements should be possible. It is also important to visualise an issue using a particular group's viewpoints or the differences of conclusions between two different groups.
Workpackage 4: Integration and mobility
In the complex information system described here, it is obvious that most necessary innovations have been done outside the project. Therefore, rather than only focussing on the development of own ideas further, it is crucial to integrate to other projects and institutions that offer tested practices, knowledge and tools for synthesising scientific information and offering science-based policy support. Therefore, Parsha will collaborate with top institutes in the world and implement their ideas, and also continuously screen development elsewhere and evaluate new methods.
During the first years of Parsha, we will visit two key institutes and collaborate with them to learn more and try out new practices developed in the project. Both visits will last approximately three months each, and they are designed to bring some important insight to be used later in the project.
First, The Governance Lab in The New York University has wide experience in testing information and policy practices.
Second, RIVM (The Environment and Health Institute in the Netherlands) has extensive expertise in policy support and the science-policy interface.
Research plan
- Research plan directly from Docs without any formatting.
Show details |
---|
Participation in shared understanding of climate and other policies (Parsha) 1. Project information Principal investigator (PI): Jouni Tuomisto, Department of Health Protection, National Institute for Health and Welfare (THL) Project title: Participation in shared understanding of climate and other policies (Parsha) Site of research: THL: , Kuopio, Finland. Date of research: 1 Sept 2017 - 31 August 2021. Date of research plan: 28 Sept 2016 Funding: Academy of Finland: 420173 €, THL: 180075 €, total: 600248 € Academy committee: Biosciences and the environment. Collaborators: The research and case studies will be performed in close collaboration with scientific, policy, and participation experts in Kuopio (National Institute for Health and Welfare), Helsinki (City of Helsinki, Open Knowledge Finland, Future Earth Finland, Prime Minister’s Office, Oxford Research), Utrecht (National Institute for Public Health and the Environment RIVM), and New York City (The Governance Lab, NYU). See chapter 7 Collaborators for details. Context: Societal decision making related to climate change mitigation, environmental health, and other contemporary issues. Topic: Testing and developing methods, practices and tools for effective and coherent information flow between researchers, policy makers, and stakeholders. 2. Rationale Basic idea and merit of the project Parsha project will take several important environmental and health policy issues under scrutiny and study available scientific knowledge but also political values and options, and stakeholder concerns and beliefs. We will start from climate policies in Helsinki and disease burden of air pollution in Finland. We will use existing and novel open methods for information synthesis and modelling. Open participation and mutual learning is ensured by large collaboration networks of researchers, policy-makers, and non-governmental organisations. The project will develop a new and effective way of resolving disputes and sharing information about these resolutions. This may improve the way we perform science and policy. Climate change policy actions need synthesis The city of Helsinki is strongly committed to climate policy by both reducing greenhouse gas emissions and adapting to a changing climate. Recently, we performed a review of important strategy papers (Tuomisto et al. 2014) and found more than 600 planned climate actions. A major challenge is to make sense out of this diverse pool, select the most effective actions, evaluate their other impacts and desirability among different groups, and implement them coherently. Making a full impact assessment of all actions would be ideal but is not feasible. There is a clear need to develop practices to organise information about extensive topics with large written materials, difficult scientific questions, high political stakes, and widely differing opinions and values. An example of a critical scientific dispute is the actual climate impact of biofuels. If biofuels are found as bad in Finland as a fresh US study demonstrates (DeCicco et al. 2016), this resolution seriously alters the preference order of several policy actions. A useful information system would identify this as a critical issue and tell, how the conclusions would change if one or another conclusion is made. The same need applies to non-scientific disputes about values as well.
Again, there is a need for an information system that can be used to organise the knowledge and produce an understandable, policy-relevant synthesis that is consistent with the methodological discussion and conclusions. Also, the conclusions should be reflected in the actual mathematical tools assessors are using in their health impact assessments. Parsha project is based on an existing system, Opasnet web-workspace, that partly offers these functionalities but that requires several steps of development and research to fulfil this need. Persistent disputes about climate friendliness of biofuels and methods of calculating and interpreting attributable risks show that there is a need for a) systematic and detailed discussion between disagreeing parties and b) an open, neutral repository for resolutions and reasonings of these discussions. RDF, DAG and BBN is a novel combination for organising information Information systems have developed and are developing rapidly, offering functionalities that were just a dream ten years ago. However, these are typically developed for needs within disciplines, while here we are talking about systems and practices that should be shared by two completely different disciplines: science and policy. A premise of Parsha is that this difference is a major reason for inefficient use of science in policy, and shared information systems and practices would be a major breakthrough. Therefore, we now present a few powerful systems and discuss their potential in this interdisciplinary task. Resource description framework (RDF) is a system developed by W3C, an organisation for standardising the Internet (REF about W3C##). RDF can describe rich spaces of information, e.g. contents of an encyclopedia, in a systematic way by defining items and relations that describe properties of the items (REF ## about RDF or Wikidata). RDF is extensively used in e.g. defining the contents of Wikipedia using the Wikidata RDF database. The database enables rich queries of the content. Such properties are needed also for describing complex policy issues. Directed acyclic graphs (DAGs) are an effective graphical way to describe items and their relations. They are extensively used in many disciplines, but they are especially useful in e.g. describing causal relations of knowledge crystals (Pearl 2009). If these relations are estimated as conditional subjective probabilities, the information in the system can be updated using Bayes’ rule, and the system is called a Bayesian belief network (BBN). RDF and BBN are becoming more commonly used, but they have not been used together to describe complex policy situations in such a way that scientific issues, values and disputes would have all been described in a single, coherent system. In addition, pragma-dialectic argumentation theory gives rules for resolving disputes (van Eemeren & Grootendorst 2004). Our innovation is that the essence of these resolutions can be described using RDF and thus take argumentation as an integral part of the system. This combination of novel techniques is unique and gives promises of important breakthroughs. Whether this combination works as expected, will be studied during the project. Although we describe causal systems and people’s reasoning in the same information system, our aim is not to build artificial intelligence that solves policy problems automatically. Rather, the objective is to build an information system where all relevant information - including political discussion - can be organised, synthesised, criticised and made readily available in useful format for policy making.
There is a lot of current activity related to open publishing of articles; open data (Open linked data##); policy relevance of scientific information (Jussila 2012, Klemola ym. 2014, Raivio 2014, Vihriälä 2016, VNK 2011); research focussed on rapid societal utility (Strategic Research Council of the Academy; Prime Minister’s Office’s research program VN-TEAS); citizen science; public participation and co-creation in policy development; and experimentation of different policies m(Tietokäyttöön##). Between and connected to all of these, there exists an ecologic niche for this project: there is an urgent need to study how information should and could flow in the society and produce a consistent, comprehensive approach that is able to capture important ideas, values, and causal understanding related to different activities in the society. It must also be able to distinguish which relevant claims can and which cannot be defended based on science, bearing in mind that there are also other than scientific worldviews in the society. Open assessment takes two steps toward shared understanding TOO LONG In our previous work with open assessment, we have already shown that its information structures are effective and applicable in impact assessments and policy analyses. (Pohjola et al State of the art, Tuomisto: Urgenche case##, Avoin PTK ##) A central idea is knowledge crystal (Tuomisto 2015) a distinct web page that has a clear research question and that aims to answer it by co-creating a synthesis of scientific data. Importantly, knowledge crystals are re-usable objects and they are expected to develop in time when they are used in new impact assessments. Knowledge crystals are also effective means to communicate, as they have a plain-text summary in the beginning and go into more and more technical details, data, and analysis code for experts in the end. This approach enables an information structure, where all information relevant to a particular research question is located in one place. This is an improvement to the prevalent scientific publishing system, where information is published in static articles in a fragmented way with limited error correction functionalities (Tragedy of error##). However, impact assessments have two major limitations, both of which will be tackled in this project. First, performing an assessment requires a lot of work to synthesise scientific information into a quantitative causal description of policy-relevant issues. Therefore, it is often not available in the time frame of rapid political decisions. Second, assessments typically focus on causal chains with established scientific knowledge, thus leaving many important aspects untouched, because solid scientific data is not available or the aspects are inherently based on values rather than facts. The first problem can be alleviated by openly available, ready-made, reusable information objects that shorten the time to make an assessment. As mentioned earlier, we already have a structure for such objects and a web-workspace for their production. Experiences so far are promising. Solutions to the second problem are less well developed. Ideally, it should be possible to describe any policy-relevant aspect with the same level of detail and scrutiny as a quantitative impact assessment model. However, this is not possible in practice. A hypothesis in Parsha is that it is enough to describe an aspect in some meaningful relation with other relevant aspects, as long as all aspects can be exhaustively described. What set of relations is enough and what attributes of an aspect should be described in a shared understanding, that is a major research question in this project. Shared understanding connects science and policy. WAY TOO LONG To understand what kind of research is needed, let’s first look at ideal shared understanding. It is a written description of all participants’ claims, values, and scientific issues that are relevant to the policy issue at hand. They are described and connected to each other with logical, causal and other relations. From this description it is possible to reproduce anyone’s viewpoint in detail to their satisfaction. It is also possible to analyse discrepancies between any two participants’ viewpoints. Importantly, the scientific viewpoint, based on data and refutation of implausible hypotheses, can also be described and used in comparing different viewpoints. Based on an ideal description, it is possible to infer what actions each participant would support or oppose and why if they were the decision-maker. This brings arguably important benefits. First, it offers a single rich source of reliable, policy-relevant information and thus reduces the power of malevolent distribution of false claims. Second, it reduces the need for politics. In the current political system, a lot of effort is put to fighting over power, but in shared understanding participants can focus on the actual substance because their views are heard based on the merit of their ideas rather than whether they are in the position to decide. This project is based on the assumption that such ideal shared understanding would improve decision support by facilitating decisions that are according to the values of the people and based on rational, slow thinking. The research questions arise from this premise. How can such a shared understanding be produced? What participatory practices motivate people to share their views with others? What information structures are needed to describe it to the participants’ satisfaction? What interfaces guide the users to useful information and help them contribute meaningfully? How much of the information work can be outsourced to non-experts without loss of quality? Exhaustiveness is a key concept. A successful online documentation of shared understanding contains all relevant aspects related to a policy issue. Then it can be used to improve signal-to-noise ratio, because then it is possible to say to a person trying to repeat old arguments: “That was discussed already, and the resolution can be found from here.” But if the description is not exhaustive, we cannot know whether a new aspect has actually been discussed, and we have to engage in the same discussion again, which is of course inefficient. It should be noted, that the current policy-making is, in this respect, very inefficient, because most lines of text written or read about a policy issue are mere repetition of a limited number of key aspects, with possibly minor variation. (REF?##) Therefore, a major research question is this: is it possible to develop a guidance for describing exhaustive shared understanding, and can that guidance be followed in practical policy situations? Basic scientific concepts may be unfamiliar to citizens or even university students, preventing learning of related facts. An effective way to avoid this problem is refutational text that specifically aims to identify and explicate typical misconceptions. (Södervik 2016). Shared understanding is a systematic approach to produce refutational texts about policy issues. 3. Objectives and expected results 3 A Objectives of the research Parsha project will study several policy cases relevant for environment and health, review scientific knowledge and synthesise political discussion. All this information is described as shared understanding of the cases. The bold aim is to produce such a comprehensive and well-structured description that most political and scientific aspects raised can be reproduced from there. Such description would enable understanding and learning what and why should be decided, given the valuations of a particular participant. If this aim is achieved, shared understanding may become a key method to produce and disseminate policy-relevant information. This may result in positive feedback loop of improved understanding, when citizens learn to use the method to demand clarity in policy, and experts and policy-makers learn to use it to offer clarity and rationale for decisions. Parsha is based on i) a social innovation of hearing every viewpoint systematically without aiming at consensus; ii) an information science innovation of a structured synthesis from which viewpoints can be reproduced and discrepancies analysed; and iii) a technological innovation of an interface to facilitate participation, knowledge retrieval, learning, and policy support. Our hypothesis is that scientific knowledge prevails in this process because of its internal consistency, while vague and flawed reasoning is identified as such and turned down. Objectives and related research questions: Substantive: To produce shared understanding and policy guidance for the city of Helsinki climate mitigation policies and strategies, accounting for health, climate and other impacts; fine particle disease burden and related policies; and other case studies. Can the guidance for Helsinki be extended or generalised to the whole of Finland? Are Finnish fine particle policies effective? Participatory: To develop practices that are tempting and rewarding to policy-makers, experts, and stakeholders so that participation in the production of shared understanding is seen useful. What factors increase or hinder participation? Are people motivated to participate and produce meaningful content? Can researchers produce a common “scientific perspective” using participation rules? Can the work be crowdsourced and in what conditions? Methodological: To test, implement, and further develop the method of shared understanding. Can shared understanding be described as an information system? What essential parts does it have? Are the contents of the information system clear to the readers and do they convey the original message? How can conflicting opinions be systematically described to everyone’s satisfaction? What tools are needed to combine slow modelling and rapid political discussion? Can shared understanding be produced without compromising scientific integrity? Technological: To produce web-based tools for creating and managing descriptions of shared understanding. Are people willing to participate using such an interface? What problems do the current typical interfaces have, and how can they be improved? Are the developed technologies scalable to much wider use? 3 B Effects and impact beyond academia Shared understanding would offer methods to improve the knowledge base of policies. Methods are likely to find support from the society, as Parsha goes along several societal megatrends (openness, participation, co-creation, evidence-based policy). It is likely to alleviate confrontation, because everyone is better heard than before and political benefits of shouting louder decrease. In some cases, popular political opinions and options would not hold against systematic criticism, reducing the risk that the actual decisions turn out bad to the society. This is actually nothing more than the scientific method applied to policy questions. There is strong evidence that the scientific method is a very powerful and effective tool to produce solid information. This is no news. However, the scientific and societal merit of the Parsha project is that it has identified important areas in the society where this method is not satisfactorily applied, and has produced innovations that can be rigorously studied in policy-relevant situations. It should be noted that the scientific method is not ideally used in the field of science either. Therefore, we also study how to produce and publish scientific information in a more criticisable and applicable way. If successful, we can learn new things about how science and policy can be performed together in a more effective and robust way by using shared information systems and practices. The societal impacts of Parsha may extend far from the substantive topics studied in the cases. This is facilitated by maximising openness and the potential for re-use of its information content. It may even change the way we think about scientific publishing: continuous participatory improving of the description of our understanding, rather than producing distinct, static objects called articles.
In Parsha, we will also do active openness by promoting the information produced and case studies planned or performed. We will also actively seek to learn from other information providers, as exemplified by the research visits to the GovLab and RIVM, and by inviting contributions to the web-workspace. This will be done through the large networks of the collaborators and participants, and by actively utilising workshops and social media, e.g. blogs, Facebook groups, and Twitter. 4. Research methods and material, support from research environment Methods Workshops, experiments literature reivews Science collected. Modelling Voiko tehdä kokeen: otetaan useita keisejä joista määritellään miten niitä pitäisi jäsentää ja millaisia tuloksia voi olla. Sitten arvotaan ne joihin lähdetään mukaan. Lopussa katsotaan oliko osallistumosella vaikutusta vai ei. Haasteena tulee olemaan uskottavien koeasetelmien rakentaminen. Tämä kai on normaiivista tutkimusta. Mutta pitäisikö olla myös deskriptiivis? Mitä se olisi? Katsotaan miten ihmiset käyttäytyvät? Seurataan kun he käyttävät työkaluja? Koska hanke on vasta vuoden päässä ei kannata liian monta keisiä kuvata liian tarkasti. Vaan kerrotaan millä oeruaatteilla niitä valitaan. Ja mitä niissä tutkitaan. Materials Trials: same information presented to the audience in two different ways randomly. Are there differences? (How to measure?) Research environment THL and its environmental health units in Kuopio is a highly skilled research environment. We have previously performed research and assessments on related topics (Asikainen et al, 2013; Hänninen and Knol, 2011; Tuomisto et al, 2014, Pohjola, 2014, Neittaanmäki, 2016). The technical infrastructure is up-to-date, and we have many statisticians and modellers to consult within the institute if needed. The project as a whole consists of a wide variety of expertise from government to Critical points and risk management In theory, Parsha is simply applying established principles in new kind of information and policy support work. In practice, we will run into problems related to complexity, vagueness, power, and compliance. Risks are reduced by wide collaboration, openness (“if you fail, do it early and learn from it”), and reliance on robust open source technology and tested methods. There are several alternative techniques available, so dead-ends are unlikely. Even if the methodology as a whole fails to mature and become directly applicable during the Parsha project, it will undoubtedly teach us a lot about how information flows between science and policy and what factors hinder the flow. This is already a valuable result. From substantive point of view, there is a lot of scientific information available about climate policies, fine particles, disease burden methods, and other relevant topics. This is not a limiting factor, but rather the methodological capability to organise and synthesise it. Fortunately our team has expertise and skills to handle exactly this kind of topics. The risk of failure can be managed by limiting the scope of work. We aim at generic methods and results with wide applicability, but the width or complexity of material may prevent progress. Then we will focus on one particular aspect or policy issue at a time, and when methods produce tangible results, step back and analyse problems seen in the generic case.
Tietorakenne. Kuvataan ensimmöinen versio tietorakenteesta, jossa on tärkeimmät uudet relaatiot. Rakkennetaan tietovarasto sen muotoiselle tiedolle. Nämä kansi samaan työpakettiin. Mallinrakennus klmasropolitiikoista. Kuvataa peuskausaalimalli mutta myös siihen liittyvät arvot ym. Mallinrakennus tautitaakasta oienhiukkasia esimerkkinä käyttäen. Tässä voimakas yhteistyö batman hankkeen kanssa. Ei saa toistaa mitä siinä tehdään vaan pitää luvata lisää. Tarkista mitä luvataan. Tähän kartoitus suomalaisita pienhiuklaspolitiikoista ja jaettu ymmärrys niistä. Haastatellaan ministeriöitä ja puolueita sekä ympäristöterveyträ kunnissa suunnitelmien selvittämiseksi. Workpackage 1: Participation This workpackage develops the social innovation of hearing and documenting all viewpoints. It is very much organised based on co-creation of case studies. The work is facilitated by collaborators. --# : Can we add Future Earth Suomi, OKFFI, and others here? --Jouni (talk) 21:01, 20 September 2016 (UTC) Case studies: The city of Helsinki has more than 600 action points in various climate strategies and other policy papers. For only a few of them, a systematic evaluation with cost estimates and climate impacts have been done. There is a clear need to organise the information, disputes, roles and actions into a coherent description. In other words: Shared understanding is needed. This will be produced using consultation of target groups within the Helsinki administration and also NGOs and other stakeholders who have opinions about the actions. There are several recent disease burden estimates of air pollution for both Finland and globally. However, there are scientific disputes about the best epidemiological methods (Morfeld REF) and scoping for assessments. These issues will be clarified by producing shared understanding on both the method (already an conference abstract: Tuomisto et al 2016) and the situation in Finland. This will be based on work and results of currently on-going Akatemia funded project BATMAN, where estimates of disease burden caused by particulate matter in Finland are updated with spatial and source specific exposure data. Furthermore, in BATMAN the effect of several emission mitigation policies are evaluated and ranked. Based on the results of BATMAN and earlier national projects Seturi and ISTE, a shared understanding on common disease burden methodology is produced. This is reached by evaluating how the used disease burden methodology and selected boundaries of assessments effects on importance of particulate matter as an environmental hazard in Finland and whether the conclusions are affected by methodological differences or not. --# : Should we describe the participatns in THL? --Jouni (talk) 21:01, 20 September 2016 (UTC) Vaccination coverage and acceptability is an issue where researchers say that the benefits of several vaccines are indisputable; yet they are disputed in the society. We will produce a shared understanding of this issue and identify the main arguments and values that contradict. It will also be a test about wether people agree to participate in a process with people who have strongly opposing views about personally very important issues. We may also learn how well viewpoints can be described based on previously written material only, without actual participation. New case studies will be decided later in the project, because it would be impossible to predict, which topics will have policy relevance after three years. Workpackage 2: Information structures Many information structures already exist for open assessments and open policy practice (Tuomisto and Pohjola 2007, Pohjola 2013). Based on our experience, those structures work well in impact assessments, which are mainly systematic, science-based quantitative modelling exercises describing causal relations. However, they are not flexible enough for describing non-causal inferences, heuristics and value judgements important in people's opinions about policy issues. There is a need for more flexible information structures. We have piloted with promising results a resource description framework using triple approach. Triples have a subject, a predicate (a relation between items) and an object describing something about the subject. With a small number of different relations we have successfully described complex environmental health issues. So, we will expand the use of this approach and see how it works in the case studies and how the participants can understand own and other people's viewpoints based on triple descriptions. In addition, we will experiment pragma-dialectical argumentation theory developed in the University of Amsterdam (van Eemeren, Grootendorst). It has been implemented in several fields ⇤# : Where, describe a few examples. --Jouni (talk) 21:01, 20 September 2016 (UTC), but not previously in producing statements for shared understanding. Another topic for investigation is a method to incorporate differing views into a single description. We anticipate that most of the time it will be possible to describe different views with the same statements but using different qualities or truth values. For example, anti-vaccine activists and pediatricians agree that "vaccines are dangerous to children" is a meaningful statement and relevant in the context; the dispute comes from the fact that the former group thinks it is true while the latter group thinks it is false. Interestingly, both groups may agree on a related statement: "If vaccines are dangerous to children, they should not be used." It is useful to see that the difference is in premises, not the reasoning. Although such an example may seem trivial, we think that in many political cases similar clarifications are not made and the precise reasons for disputes are not understood. Another use of such an information structure is that it raises issues like "What do we actually mean by 'dangerous'?", thus directing further discussions into directions that improve shared understanding. Workpackage 3: Technical development We already have an open web-workspace Opasnet, where assessors can perform impact assessments. However, technical development is needed to include a triple database and other functionalities for non-causal reasoning and statements and expressions of value judgements. This is necessary to operationalise the work produced in WP2. However, it is also important to enhance the user experience. It should be able to easily document own viewpoints and link them to other relevant information already in the system. Also, search functionalities, linkages of related items, and at least some automatic or semi-automatic reasoning based on given premises and statements should be possible. It is also important to visualise an issue using a particular group's viewpoints or the differences of conclusions between two different groups. Workpackage 4: Integration and mobility In the complex information system described here, it is obvious that most necessary innovations have been done outside the project. Therefore, rather than only focussing on the development of own ideas further, it is crucial to integrate to other projects and institutions that offer tested practices, knowledge and tools for synthesising scientific information and offering science-based policy support. Therefore, Parsha will collaborate with top institutes in the world and implement their ideas, and also continuously screen development elsewhere and evaluate new methods. During the first years of Parsha, we will visit two key institutes and collaborate with them to learn more and try out new practices developed in the project. Both visits will last approximately three months each, and they are designed to bring some important insight to be used later in the project. First, The Governance Lab in The New York University has wide experience in testing information and policy practices. Second, RIVM (The Environment and Health Institute in the Netherlands) has extensive expertise in policy support and the science-policy interface. 7. Research team and collaborative partners Participants and their special interests (with funding): THL: shared understanding method, disease burden estimates. All researchers are experts in environmental health, impact assessment, disease burden, and quantitative modelling. Jouni Tuomisto: chief researcher ##PITÄISI KIRJOITTAA MEISTÄ ESITTELYT…. Otto Hänninen: senior researcher … Arja Asikainen: researcher… Päivi Meriläinen: resarcher … N.N.: information technology expert ... Collaborators and their special interests (with letter of commitment but no direct funding from the Academy) City of Helsinki (Environment Center): climate change mitigation and adaptation policy and open data: Jari Viinanen: graduated as M.Sc in field of energy and environment from Lappeenranta University of Technology. Since 2002 he has been working as environmental inspector in City of Helsinki Environment Centre. He is a specialist of urban climate change mitigation and adaptation and has worked in climate change projects developing strategies and writing reports. Jari Viinanen has also done air quality and waste management expert work in relation to air quality programmes and street dust research. The Governance Lab (New York University, USA): Experimenting of social methods. Beth Noveck: (born 1971) graduated from Harvard University with an AM magna cum laude, and University of Innsbruck with a PhD. She graduated from Yale Law School with a JD. She directs The Governance Lab and its MacArthur Research Network on Opening Governance, which is designed to improve governance in governments and elsewhere. Beth Noveck is currently the Jerry Hultin Global Network Professor at New York University’s Tandon School of Engineering. She was formerly the Jacob K. Javits Visiting Professor at the Robert F. Wagner Graduate School of Public Service and a visiting professor at the MIT Media Lab. She is a professor of law at New York Law School and a Senior Fellow at the Yale Law School Information Society Project. She served in the White House as the first United States Deputy Chief Technology Officer and director of the White House Open Government Initiative (2009-2011). UK Prime Minister David Cameron appointed her senior advisor for Open Government, and she served on the Obama-Biden transition team. She’s also designed or collaborated on Unchat, The Do Tank, Peer To Patent, Data.gov, Challenge.gov and the Gov Lab’s Living Labs and training platform, The Academy. National Institute for Public Health and the Environment, RIVM (The Netherlands): science-based policy support. Erik Lebret, MSc, PhD (1953) is Chief Science Officer "integrated risk assessment" at RIVM. His areas of expertise are environmental epidemiology, exposure assessment, and integrated environmental health impact assessment. He has more than 30 years of expertise on supporting policy makers and stakeholders in the use of scientific knowledge. In addition, as endowed professor, he works for the Institute for Risk Assessment Sciences (IRAS) at Utrecht University on Environmental Health Impact Assessment. Erik Lebret has been a principal investigator for various European research projects (EU Framework Programmes) and team leader of a consortium that carried out research into the effects of air pollution on the respiratory systems in children in six Central and Eastern European countries. In addition to his work at RIVM, he regularly participates in national and international expert and advisory committees for the Health Council of the Netherlands, the World Health Organisation and several foreign expert committees, as well as several committees at the US EPA. Open Knowledge Finland ry: open science; participation in open society. Heidi K. Laine: leader of the Open Science Working Group. Supports open sharing of research data, code, protocols, teaching material, publications, and other resources, and citizen science. Promotes standards of openness in Finnish academia and interactions between academic institutions and wider society. Heidi Laine obtained her master’s degree in social sciences and is currently a University of Helsinki Doctoral Candidate in the Humanities and Social and Behavioural Sciences. Laine’s research is an experiment in Open Science in the domain of qualitative social science. In addition Laine has expertise concerning research ethics and integrity in the context of Open Science. Oxford Research: expert advice in policy support: Arttu Vainio: is currently working as CEO of Oxford Research Oy to offer services in the fields of research, evaluation and consulting with the Finnish Oxford Research team and together with the Oxford Research offices in Denmark, Sweden, Norway and Latvia and promoting cooperation with customers and partners as well as creating networks both in public and private sectors. He graduated as MSc in field of Economics from University of Vaasa in 1990 and had his Postgraduate degree LSc in 1996. Currently he is writing a doctoral thesis on the use of evaluation in European Regional Development Fund programmes in Finland for Radboud University Nijmegen in Netherlands. Arttu Vainio has long experience of planning, managing and carrying out demanding research assignments. Good knowledge of research methods and long experience especially in the field of evaluation studies. Specialized in EU co-funded programmes and projects especially regarding the Structural Funds (ESF and ERDF) and other EU programmes (e.g. LEADER, INTERREG). Existing transnational cooperation networks especially in Northern Europe and wide networks in Finland including personal connections to ministries, regional administration and research organisations. Future Earth Suomi: co-creation of policy-relevant science. Tanja Suni: obtained her PhD in atmospheric physics in 2004 at the University of Helsinki. She then continued her post-doc in Australia before moving to work for international global change programmes. In 2011-2015, Dr Suni led the International Project Office for the iLEAPS (Integrated Land Ecosystem – Atmosphere Processes Study) core project of the International Geosphere-Biosphere Programme. Since 2014, she has participated in developing the new global change research programme Future Earth as Executive Director of the European network of national Future Earth platforms and Secretary General of the Finnish national committee for Future Earth, Future Earth Finland. Her work concentrates on developing research culture to better answer the grand challenges of sustainability: this requires solutions-oriented, interdisciplinary research on sustainability issues, stakeholder engagement, and transforming institutional structures such as funding and merit systems to help scientists contribute to societal challenges in full. Prime Minister’s Office: Kaisa Lähteenmäki-Smith: is currently working in Finnish Prime minister’s office as a specialist in science policy in Policy analysis unit. She had her PhD in political science from University of Turku, Finland in 1999 and has previously worked on R&D projects dealing with research and innovation, as well as regional development, in national, Nordic and European context. Since finishing her PhD on Europeanisation and regionalisation, she has worked with European aspects of Finnish regional policy at the Finnish Ministry of the Interior and thereafter in an international research context at Nordregio (Nordic Centre for Spatial Development) in Stockholm on project ranging from structural reform, sustainable development and programme evaluation to the governance aspects of European Structural Funds. Since November 2007 she was a consultant at Net Effect Ltd and thereafter at Ramboll Management Consulting, specializing in policy and programme evaluation, as well as institutional evaluation. Her recent evaluation projects have included the evaluation of Portuguese-American RDI-programmes, institutional evaluation of VTT Technical Research Centre of Finland, institutional evaluation of Finnish Innovation Fund, MIDE programme (Aalto University's multidisciplinary programme for digitalization and energy) and the Strategic Centres for Science, Technology and Innovation (SHOK). Manchester University: Colin Talbot: a professor of Government in Politics and holds the Chair of Government in the School of Social Sciences. He completed MSc in Public Sector Management at Southbank University (with Distinction) and a PhD at the London School of Economics. He has been a Chair of Public Policy and Management at Glamorgan University, a Chair of Public Policy at Nottingham and a Chair of Public Policy and Management at Manchester Business School. His main area of expertise in is public services and public management reform. He has completed major international comparative studies on the creation of arms-length agencies (for the UK government and ESRC); of the use of performance reporting systems (for the National Audit Office); and of budget participation and scrutiny systems (for the Scottish Parliament). Colin has advised Parliamentary Committees on performance and public spending issues for the Treasury, Public Administration and Welsh Affairs Committees. He is currently looking at the interface between academia and policymaking through a series of studies and experimental virtual "Policy Labs". Colin has advised a wide range of international public sector organisations and/or carried out research in countries as diverse as Canada, Jamaica, Mexico, Tanzania, India, Bangladesh, Japan, Hong Kong, Malta and Sweden and spoken at conferences and seminars in over two dozen countries. 8. Research careers, fulfilment of the mobility requirement and researcher training Advancing the research career of the applicant or other researchers to be funded
|
See also
Some important pages related to the project:
- Drive folder for the application (DL 28.9.2016)
- Project description in Harava (password-protected)
- Relevant methods in Opasnet:
- Academy instructions:
- Pekka Neittaanmäki, Timo Huttula, Juha Karvanen, Tom Frisk, Jouni Tuomisto, Antti Simola, Tero Tuovinen, Janne Ropponen. Unicorn – Open science for assessing environmental state, human health and regional economy. Research Ideas and Outcomes 2: e9232 (16 May 2016) doi:10.3897/rio.2.e9232 [1] [2]
- Jouni Tuomisto; John Evans; Arja Asikainen; Pauli Ordén. 2016. Discussion rules as a method to resolve scientific disputes. In: Abstracts of the 2016 Epidemiology (ISEE). Abstract P2-377, ID: 3552. Research Triangle Park, NC:Environmental Health Perspectives; http://dx.doi.org/10.1289/ehp.isee2016.