You are on page 1of 54

RESEARCH

SERIES
vol 05

STAKEHOLDER
ANALYSIS
INTERNATIONAL CITIZEN SCIENCE STAKEHOLDER
ANALYSIS ON DATA INTEROPERABILITY

Claudia Gbel, Museum fr Naturkunde Berlin, Germany


Victoria Y. Martin, Cornell Lab of Ornithology, Ithaca, NY, USA1
Mnica Ramrez-Andreotta, Department of Soil, Water, and
Environmental Science and Division of Community, Environment & Policy,
University of Arizona, Tucson, AZ, USA
STAKEHOLDER
ANALYSIS
INTERNATIONAL CITIZEN SCIENCE STAKEHOLDER
ANALYSIS ON DATA INTEROPERABILITY

Claudia Gbel, Museum fr Naturkunde Berlin, Germany


Victoria Y. Martin, Cornell Lab of Ornithology, Ithaca, NY, USA1
Mnica Ramrez-Andreotta, Department of Soil, Water, and
Environmental Science and Division of Community, Environment & Policy,
University of Arizona, Tucson, AZ, USA

Authors are listed in alphabetical order.


Each author contributed equally to this report.
STAKEHOLDER ANALYSIS:
International Citizen Science Stakeholder Analysis on Data Interoperability.

EXECUTIVE SUMMARY
Commons Lab
Science and Technology Innovation Program
Woodrow Wilson International Center for Scholars
One Woodrow Wilson Plaza
1300 Pennsylvania Avenue, N.W.
Washington, DC 20004-3027

https://www.wilsoncenter.org/program/science-and-technology-innovation-program

Authors: Claudia Gbel, Victoria Y. Martin, Mnica Ramrez-Andreotta


Cover Design: Jeremy Swanston
Layout Design: Kathy Butterfield

2017, The Woodrow Wilson Center: This work is licensed under Creative Commons
Attribution-NonCommercial-NoDerivs 3.0 Unported License: http://creativecommons.
org/licenses/by-nc-nd/3.0/

This report may be reproduced in whole, or in part, for educational and non-commercial
uses, pursuant to the Creative Commons Attribution Non-Commercial-No-Derivs-3.0-
Unported License found at http://creativecommons.org/licenses/by-nc-nd/3.0/ and
provided this copyright notice and the following attribution is given.

Claudia Gobel, Victoria Y. Martin and Monica Ramirez-Andreotta. Stakeholder Analysis:


International Citizen Science Stakeholder Analysis on Data Interoperability Final Report.
Washington, DC: Woodrow Wilson International Center for Scholars (2016).

Users may not use technical measures to obstruct or control the reading or further
copying of the copies that they make or distribute. Nongovernmental users may not
accept compensation of any manner in exchange for copies. The Woodrow Wilson
Center is open to certain derivative uses of this product beyond the limitations of the
included Creative Commons License, particularly for educational materials targeted at
expanding knowledge on the Commons Labs mandate. For more information, please
contact STIP@wilsoncenter.org.

Available for download free of charge at

https://www.wilsoncenter.org/research/Commons%20Lab

This report is funded by the Alfred P. Sloan Foundation and published through the
Woodrow Wilson International Center for Scholars, in Washington, DC.

The views and opinions expressed in this report are those of the authors and are not pre-
sented as those of any of the sponsoring organizations or financial supporters of those
organizations. Any errors and omissions are the responsibility of the authors and editors.

iv
The Woodrow Wilson International Center for Scholars is the national, living
U.S. memorial honoring President Woodrow Wilson. In providing an essential link
between the worlds of ideas and public policy, the Center addresses current and
emerging challenges confronting the United States and the world. The Center promotes
policy-relevant research and dialogue to increase the understanding and enhance the
capabilities and knowledge of leaders, citizens, and institutions worldwide. Created
by an act of Congress in 1968, the Center is a nonpartisan institution headquartered
in Washington, D.C.; it is supported by both public and private funds.

Conclusions or opinions expressed in Center publications and programs are those


of the authors and speakers. They do not necessarily reflect the views of the Center
staff, fellows, trustees, advisory groups, or any individuals or organizations that provide
financial support to the Center.

For more information about the Centers activities and publications, please visit us
on the Web at www.wilsoncenter.org.

Wilson Center Leadership


The Honorable Jane Harman, Director, President and CEO

Board of Trustees
The Honorable Thomas R. Nides, Vice Chair, Morgan Stanley

PUBLIC MEMBERS:
The Honorable William D. Adams, Chairman, National Endowment for the
Humanities
The Honorable Thomas E. Price, Secretary, U.S. Department of Health and Human
Services
The Honorable David Ferriero, Archivist of the United States, National Archives and
Records Administration
The Honorable Carla D. Hayden, Librarian of Congress
The Honorable Rex W. Tillerson, Secretary, U.S. Department of State
The Honorable Elisabeth DeVos, Secretary, U.S. Department of Education
The Honorable David J. Skorton, Secretary, Smithsonian Institution

PRIVATE CITIZEN MEMBERS:


Peter J. Beshar, Executive Vice President & General Counsel, Marsh & McLennan
Companies
Thelma Duggin, President, The AnBryce Foundation
Barry S. Jackson, Managing Director, The Lindsey Group, Strategic Advisor,
Brownstein Hyatt Farber Schreck
The Honorable David Jacobson, Former U.S. Ambassador to Canada and Vice Chair,
BMO Financial Group
Nathalie Rayes, National Public Relations, Director of Grupo Salinas, Executive
Director, Fundacin Azteca America
Earl W. Stafford, Chief Executive Officer, The Wentworth Group, LLC
Jane Watson Stetson, Philanthropist
The Honorable Louis Susman, Former U.S. Ambassador to the United Kingdom and
Manager, CBI Holdings
EXECUTIVE SUMMARY

vi
The Science and Technology Innovation Program (STIP)
analyzes the evolving implications of such emerging technologies as synthetic
biology, nanotechnology, and geoengineering. STIPs research goes beyond
laboratory science to explore new information and communication technolo-
gies, sensor networks, prediction markets, and serious games. The program
provides critical yet nonpartisan research for the policymaking community
and guides officials in the design of new governance frameworks. It gauges
crucial public support for science and weighs the overall risks and benefits
of technology for society at large.

The Commons Lab of STIP seeks to mobilize public


participation and innovation in science, technology and
policy.

Commons Lab Staff


Anne Bowser, CoDirector, Commons Lab
Elizabeth Tyson, CoDirector, Commons Lab

Blog: https://www.wilsoncenter.org/blogs/ctrl-forward
Facebook: https://www.facebook.com/WilsonSTIP
Twitter: https://twitter.com/WilsonSTIP

The Commons Lab of the Science and Technology Innovation Program is sup-
ported by the Alfred P. Sloan Foundation.
ACKNOWLEDGEMENTS

The Authors are indebted to our inter- Drs. Mimi Arandjelovic and Hjalmar
viewees for contributing their precious Kuehl (Chimp&See / MPI-EVA Pan Af-
time, experience, expertise, and ideas to rican Project: The Cultured Chimpan-
this study, to the reviewers for their criti- zee), Blint Balzs (Environmental So-
cal comments and thoughtful sugges- cial Science Research Group), Pieter
tions, and to the Wilson Center staff, van Boheemen (Waag Societys Open
Anne Bowser and Elizabeth Tyson for Wetlab, Public laboratory for biotech-
their visionary leadership in establishing nology), project manager from commu-
this study, and their careful guidance nity mapping project on urban air qual-
throughout the process. We would ity in the EU, Douglas Tait (Community
especially like to thank: Theresa Crim- Driven Science: Understanding the Im-
mins (USA National Phenology Net- pacts of Coal Seam Gas Development
work), Darlene Cavalier (Arizona State in Australia), Jemina Stuart-Smith (Red-
University and SciStarter), Ben Ger- map Australia), Ross Goldingay (Lis-
hardstein (U.S. Agency for Toxic Sub- more Koala monitoring), Rose Herben
stances and Disease Registry), Grand (EstuaryWatch), Peter Brenton (Atlas
Lake Watershed Mercury Study, Steve of Living Australia), Renae Sayers (Fire-
Kelling (eBird), Connie Walker and balls in the Sky), and Dr. Debra Stokes
Mark Newhouse (Globe at Night), Ha- (Southern Cross University, Australia).
git Keysar (Public Lab Organizer), Da- This research was made possi-
cha Atienza Ariznavarreta (BioBlitzes ble through the generosity of the
at NHM Barcelona), Andrea Sieber Alfred P. Sloan Foundation.
(Landscape and You-th & BrotZeit),
ABOUT THE AUTHORS
CLAUDIA GOBEL
Claudia Gbel is project manager at the European
Citizen Science Association and guest researcher
at Museum fr Naturkunde Berlin. Her background is
in science and technology studies. She works on the
organization of citizen sciences, links to open science
and policy engagement.

VICTORIA Y. MARTIN
Victoria Martin is an environmental social scientist with
research experience in a wide range of environmental
impact and management issues in Australia and New
Zealand. Her recent PhD thesis focused on public
engagement in marine citizen science in Australia.
Victoria is now a Rose Postdoctoral Research Fellow
with the Cornell Lab of Ornithology, USA, where she
is continuing her research into citizen science.

MONICA RAMIREZ-ANDREOTTA
Dr. Mnica Ramrez-Andreotta is an assistant profes-
sor of Soil, Water and Environmental Science with a
join appointment in the College of Public Health and
Director of the Integrated Environmental Science and
Health Risk Laboratory at the University of Arizona.
Her research programs include: developing a fun-
damental understanding of the fate and transport of
contaminants in plant-soil systems, building citizen
science programs to increase public participation
in environmental health research, creating low cost
environmental monitoring tools to improve exposure
estimates, and designing data visualizations and risk
communication strategies to improve environmental
health literacy.
CONTENTS

FOREWORD / 1

EXECUTIVE SUMMARY / 3

INTRODUCTION / 7

AIMS AND METHODS / 11


Sampling Procedure / 12
Interview Protocol and Data Analysis / 12

RESULTS / 17
Stakeholders of citizen science projects / 17
Nature of stakeholder involvement / 17
Potential for engaging new stakeholders / 18
Internal and external stakeholders / 18
Data sharing, accessibility, use & reuse / 20
Sharing and accessibility of data and knowledge / 20
Use and reuse of data and knowledge / 22
Data Standardization and Interoperability / 23
What is currently being done by projects and rationale / 23
Overall understanding and perceptions of interoperability / 24
Concerns regarding interoperability / 24
DISCUSSION / 29
Interoperability is only slowly becoming a topic of concern in the citizen science
community / 29
Heterogeneity of data sharing practices and adoption of standards for
interoperability / 30
A broader concept of interoperability is needed to work across disciplines and
project types / 30
Advancing interoperability both facilitates and rests on involvement of stakeholders
/ 31
Limitations of the Study / 33

OVERALL RECOMMENDATIONS TO THE FIELD AND FUTURE RESEARCH


/ 35
Future research and practical needs / 36

ENDNOTES / 39
EXECUTIVE SUMMARY

xii
FOREWORD
I first began thinking about citizen sci- developed by, the diverse global citizen
ence data standards during the sum- science community.
mer of 2013, as a Research Intern with
the DataONE Public Participation in To address these concerns Claudia
Scientific Research (PPSR) Working Goebel, Elizabeth Tyson, and I conceived
Group. DataONE was interested in what of this Stakeholder Analysis at the 2015
we called Project Metadata, perhaps meeting of the European Citizen Science
more accurately defined as a set of key Association (ECSA). Through this report
terms that could help databases like we find empirical evidence for the impor-
SciStarter, the Cornell Lab of Ornithology, tance of data standards in citizen science,
and CitSci.org share records. It quickly for example learning that some authorities
became clear that the value of developing may not use citizen science data because
a common, shared vocabulary for talking of uncertainty about data quality assur- ANNE BOWSER
about citizen science extends far beyond ance and quality control measures, and a CoDirector,
the ability to exchange database records. lack of data standardization practices (p. Commons Lab
Standardized metadata documentation 27). Armed with this knowledge, citizen Science and
promotes the re-use of information by al- science projects can decide to adopt the Technology
lowing researchers outside of the immedi- standards endorsed by formal authorities Innovation Program
ate project team to make decisions about to make it easier for their data to be used. Wilson Center
fitness for use. And developing a com- Or they may deliberately create their own
mon data model can allow researchers standards. Some citizen science commu-
working on different scales and research nities who use bucket sampling to mea-
domains to exchange information, thus sure air quality, for example, design their
scaling the impact of any single citizen own protocols to highlight the absence of
science activity. existing standards for certain pollutants or
point out discrepancies between existing
The promise of citizen science data in- standards for monitoring.1
teroperability is significant, and recog-
nized by a number of organizations around This report uncovered a number of bar-
the globe. By 2015 key initiatives were led riers to the adoption of data standards
by the European Citizen Observatory Web in citizen science. Some projects may
(COBWEB) project, which began devel- doubt the applicability of standards to
oping a common data model through the their research goals, or fear losing relevant
Open Geospatial Consortium (OGC), and information if standards are too general
the Atlas of Living Australia(ALA), which or vague (p. 23). Once articulated, these
began designing an Australian citizen challenges can be addressed by future
science project database and structured initiatives. We offer this report as a road-
data collection protocol supported by map to help coordinate and inform future
the BioCollect tool. We were excited by work on data interoperability, so that any
these initiatives but also concerned that standards produced can be valued or
without comprehensive planning, data at least understood by a diverse range
standards advanced by a handful of tech- of citizen science stakeholders around
nologists and community leaders would the globe.
ultimately be imposed on, rather than co-

1
1 Ottinger, Gwen. Social Movement-Based Citizen Science. Cavalier,
Darlene and Kennedy, Eric, Eds., The Rightful Place of Science: Citizen Sci-
ence, Tempe, AZ: Arizona State University, 2016.
EXECUTIVE SUMMARY

2
EXECUTIVE SUMMARY
This study examines citizen science proj- diversity data from a regional network
ects, with particular emphasis on the and data collected by a public health
stakeholders involved in data and knowl- program. This would enable scientists to
edge generation and the use and reuse advance research on a range of topics,
of information, with the aim of inform- like by using plant phenology to predict
ing data and metadata standardization allergy seasons, understand how lead
and interoperability initiatives. It draws concentrations in drinking water relate to
on semi-structured interviews with 16 lead levels in childrens blood, or combine
CS projects with different governance data on avian presence, distribution, or
models, disciplines, and project aims in health air quality data.
the United States (U.S.), Europe, and
Australia, along with two citizen science Main findings: The projects reported
project catalogues and two potential data a broad spectrum of stakeholders who
users in government and academia. In engage in project decision-making, pro-
this study, interoperability is defined as vide support, and/or use project data.
the ability of different information technol- Six stakeholder groups were identified
ogy systems and software applications in the interviews: (1) Civil society organi-
to communicate, exchange data, and use zations, informal groups and community
the information that has been exchanged. members; (2) Academic and research
It rests on the development of standards organizations; (3) Government agen-
and seeks to give parties outside the cies and departments; (4) Participants;
immediate project team, i.e. external and (5) Formal learning institutions such as
potential stakeholders, access to data schools; and, (6) Businesses or industry.
and knowledge. Barriers to the involvement of a wide
range of stakeholders included: a lack of
Achieving greater interoperability would awareness by these stakeholders about
enable data generated by different citizen the project, or vice versa; difficulty in
science projects to more easily be re- accessing or knowing how to access
used by diverse parties (e.g., volunteers, potential stakeholders; and, time and
researchers, and decision-makers); by resource constraints for doing so.
combining data sets of different scales
(e.g., local, regional, national, and Perspectives on data standardization
global scales); and, by combining dif- and interoperability efforts were het-
ferent types of data (e.g., to help answer erogeneous across projects. Benefits
transdisciplinary research questions). were seen by many interviewees, al-
For example, water quality information though overall the understanding of what
collected by one local citizen science interoperability means was limited. Many
project could be combined with diverse projects either used research domain
types of environmental monitoring or bio- specific, regulatory, or community

3
COMMONS LAB | STAKEHOLDER ANALYSIS

standards while making their data acces- interoperability both facilitates and rests on
sible via domain specific databases, or the involvement of internal, external and, if
shared online tools. Some projects also possible, potential stakeholders.
built infrastructure for data interoperability.
For projects dealing with qualitative data Recommendations: (1) Citizen science
or focusing more on non-data related ac- project managers should invest time at
tivities like knowledge sharing or devising the outset of projects to identify stake-
experimental designs, it was largely unclear holders who could use, and potentially
what data interoperability could mean to reuse data and knowledge. Existing data
them and how they would be impacted. and metadata standards should be used
Main concerns raised regarding the pro- for data management whenever possible.
motion of interoperability in citizen science (2) Interoperability initiatives should be
included limited applicability due to the transparent, open to all types of citizen sci-
natural science bias of standards; costs of ence projects, involve internal and external
adopting standards; doubts about adapt- stakeholders, and consult potential stake-
ability to local circumstances; resistance holders. Shared citizen science data and
to curtailing stakeholder participation and metadata standards should be adaptable
passing of burden; fear of losing relevant to the needs of citizen science projects,
information and decreasing data quality; leverage existing community standards,
and, difficulties in agreeing on common and be open to review and extension. Effort
metadata terms. should be dedicated to building a con-
cept of interoperability that goes beyond
Conclusions: Interoperability is only data, the natural sciences and informa-
slowly becoming a topic of concern in tion technologies. Future interoperability
the citizen science community. The het- initiatives should embrace plurality in a
erogeneity of data sharing practices and comprehensive way. (3) Citizen Science
adoption of diverse types of standards associations and other networks should
represent major challenges for interop- offer capacity building on interoperability
erability initiatives. A broad concept of and facilitate the adoption of data and
interoperability is needed to work across metadata standards.
disciplines and project types. Advancing

4
Photo Credit: Gardenroots Citizen Science Project

Citizen Science at Work


A community gathering for data sharing through the Gardenroots project. A university
student discusses site-specific soil and plant quality data with a community garden
group. Under Gardenroots, community members collected soil and culturally relevant
crop samples from their community garden. Ensuring that projects can co-design pro-
tocols on the community level while still generating data that can be re-used by external
researchers is a key challenge of interoperability projects.

5
EXECUTIVE SUMMARY

6
INTRODUCTION 1

Mobilizing data for research and making the DataONE Public Participation in
data count as evidence for decision- Scientific Research (PPSR) working
making are among the major driving group began working on PPSR CORE,4
forces of citizen science and other forms initially designed as a metadata standard
of participatory research.2, 3 Data and for describing key facets of citizen sci-
information access, licenses, owner- ence projects to help existing project re-
ship, and quality are central topics for positories exchange records. The follow-
citizen science endeavors. Considering ing year, researchers in Europe started
the diversity of emerging projects along developing SWE4CitizenScience, a
with the specificities of how they are common data model that once imple-
implemented locally, it is no wonder that mented and deployed will support data
interoperability - or, the ability of different interoperability between almost all types
information technology systems and soft- of crowdsourcing projects. Actors from
ware applications to communicate, ex- around the globe are involved in these
change data, and use the information that projects, including working groups
has been exchanged - is key to making in the U.S.- based Citizen Science
citizen science research transformative Association (CSA), the European
and more widely accessible to science, Citizen Science Association (ECSA), the
policy, and society at large (see Figure Open Geospatial Consortium (OGC),
1 for a definition of interoperability and Biodiversity Information Standards
other important concepts). (TDWG), The Committee on Data for
Science and Technology (CODATA),
In the past years, various activities have a COST Action on citizen science,
emerged with the aim of fostering in- platforms like CitSci.org, SciStarter
teroperability, both by standardizing and the Atlas of Living Australia, the
the information used to describe citi- Joint Research Center of the European
zen science projects, and by developing Commission, the Global Biodiversity
joint data and metadata standards and Information Facility, and the Woodrow
protocols for data collection and shar- Wilson International Center for Scholars.
ing to support exchange and reuse. In
2012, U.S. researchers affiliated with

7
COMMONS LAB | STAKEHOLDER ANALYSIS

Citizen science programs each have differ- health issues, at national and even global
ent technical requirements, project goals, scales. However, any standardization effort
interests, and stakeholders. Adopting a also poses challenges, including techni-
common and shared vocabulary for de- cal hurdles in working with diverse and
scribing citizen science projects and citi- heterogeneous communities of practice as
zen science data (hereafter: data stan- well as inclusiveness questions regarding
dards) can allow information collected by, transparency and openness for co-creation
for example, a local air quality monitoring and co-design in the process of defining
project to be understood and re-used by both common and project-specific prin-
researchers and communities working ciples and tools.
on other related topics, such as public

Figure 1: Key Concepts

Key Concepts

This study seeks to explore the range of stakeholders of citizen science projects to
better engage them in using data standards to promote interoperability in citizen
science.

In the context of this study, a stakeholder is an individual or organization that


contributes to realizing a citizen science project, has a vested interest in a citizen
science project, and/or benefits from the research activities and data produced.

Data standards include common vocabularies, information formats, and protocols


that may be used to exchange information about citizen science projects, datasets,
and data.

Interoperability is the ability of different information technology systems and


software applications to communicate, exchange data, and use the information
that has been exchanged.

8
Adopting a common and shared vocabulary
for describing citizen science projects and
citizen science data (data standards) can allow
information collected by, for example, a local air
quality monitoring project to be understood and re-
used by researchers and communities working on
other related topics, such as public health issues,
at national and even global scales. However, any
standardization effort also poses challenges,
including technical hurdles in working with diverse
and heterogeneous communities of practice
as well as inclusiveness questions regarding
transparency and openness for co-creation
and co-design in the process of defining both
common and project-specific principles and tools.

Photo Credit: Mosquito Alert Spain

Global Citizen Science


The Global Mosquito Alert is a consortium of citizen science projects committed to
developing a common protocol and shared platform for mosquito monitoring. This is one
example of a citizen science data interoperability initiative that will benefit from involving
diverse stakeholders around the globe.

9
EXECUTIVE SUMMARY

10
AIMS AND METHODS 2

In the context of ongoing work on data who could potentially benefit from (re)
standardization in citizen science, this using the data and knowledge created.
exploratory study aims to (1) understand 3. Gather citizen science project leaders
which stakeholders or parties may be perspectives on:
impacted by these efforts, and (2) identify
potential consequences of standardiza- a. Data generation, management, own-
tion and interoperability. The focus is on ership, accessibility, and sharing.
individuals, groups, and organizations b. Data standardization and interoper-
involved in citizen science activities, as ability efforts.
well as those affected by or potentially
A stakeholder analysis is a methodology
benefiting from the data and knowledge
originating in the social sciences used to
generated in such projects. On this basis
identify parties that are or will be affected
the study seeks to inform ongoing and
by, and thus have a stake in a political
future standardization and interoperabil-
program or particular issue, such as an
ity efforts as well as to facilitate broader
oil spill or a plan to construct a factory in
participation by global citizen science
a given location. While it is a common ap-
communities.
proach in business and natural resource
management, it has become popular as
To accomplish this goal, specific
a preliminary study for designing and
objectives are to:
implementing broad stakeholder engage-
1. Survey a purposive sample of citizen ment mechanisms in recent years.6 In the
science projects representing a full context of this study, a stakeholder is
spectrum of governance models and understood as an individual or organiza-
participatory approaches, as well tion that contributes to realizing a citizen
as projects from different scientific science project, has a vested interest in
disciplines.5 a citizen science project, and/or benefits
2. Identify the various types of stakehold- from the research activities and data
ers involved in citizen science efforts, produced. While a broad spectrum of
and explore additional stakeholders contributions, e.g. money, time, equip-

11
COMMONS LAB | STAKEHOLDER ANALYSIS

ment, etc., is considered, special attention were chosen to provide an international


is paid to the generation and use of data, perspective in the global citizen science
knowledge and similar outputs. landscape. However, due to the limited
number of interviews, no comparisons are
offered between geographic regions.
SAMPLING PROCEDURE
Given the complexities and variability within The majority of projects selected during
and between citizen science projects, and the initial round of sampling focused on the
the exploratory nature of this study, a quali- collection of data through direct observa-
tative study design was selected. While tion. The initial sample was then extended
this precludes a broad study of a large to support a broader perspective on in-
number of projects, the method instead al- teroperability. In addition to data collection
lows for exploration of the issues in greater through direct observation, other projects
depth than can be achieved through quan- surveyed use qualitative methodologies
titative research. Potential projects for the or are concerned with collecting physi-
study were identified through emails to cal samples, use sensors to collect data,
regional citizen science networks, online analyze data online, or conduct lab experi-
searches, and researchers familiarity with ments. In addition to initial interviews with
projects through their work. The project citizen science projects, four additional
leaders were contacted with a request for interviews were conducted to explore the
an interview, all of whom agreed. perspective of key stakeholders identified
during the first round of interviews. These
Based on a purposive sampling strategy,7 additional interviews were conducted with
a total of 16 projects8 were selected to representatives from research infrastruc-
represent different categories of citizen tures including a national citizen science
science governance models across dif- database (Atlas of Living Australia) and an
ferent scientific disciplines and regions international citizen science project reposi-
(Table 1). The five governance models9 are: tory (SciStarter), both of which are involved
contractual, contributory, collaborative, in ongoing interoperability initiatives, as
co-created and collegial. These models well as an environmental scientist and a
represent different roles and responsi- government employee (both as potential
bilities members of the public may play in data users).
citizen science research. This well-recog-
nized typology enables the study to cover Interview Protocol and Data
a wide range of activities and stakeholders.
Analysis
Regarding scientific disciplines, projects
were selected to ensure representation The stakeholder analysis interview protocol
of various fields from natural sciences, was guided by a review of relevant literature
social sciences and humanities in order and discussions with research domain ex-
to include different methodologies and perts, including an exploratory stakeholder
research outputs. The selection of scien- mapping exercise at a citizen science data
tific disciplines for the sampling framework workshop held by the Joint Research
was also informed by previous research.10 Centre of the European Commission.12 The
Three regions (Europe, USA and Australia) interview questions were developed by the

12
INTERNATIONAL CITIZEN SCIENCE STAKEHOLDER ANALYSIS ON DATA INTEROPERABILITY | RESEARCH SERIES

Table 1. Sampling framework*

Discipline

Citizen Biology Environmental & Earth Social Sciences Astronomy


Science/ PPSR Sciences & Humanities and
Governance Planetary
Model Sciences

Contractual CSG impacts (AU)

Contributory Koala monitoring (AU) EstuaryWatch (AU) Fireballs in


USA National Phenology the Sky (AU)
Network - Natures Note-
book (U.S.)
eBird (U.S.)
BioBlitz Barcelona (EU)

Collaborative Redmap (AU) Globe at


Chimp&See (EU) Night (U.S.)

Co-created Public laboratory for Grandlake Watershed Community


biotechnology (EU) Mercury Study (US) supported
Community mapping of agriculture study
urban air quality project Hungary (EU)
(Anonymous, EU) Science and Youth /
BrotZeit (EU)

Collegial Civic View from Above


- Public Labs (U.S.) DIY
aerial photography toolkit
in spaces of political
conflict (project takes
place in Israel/Palestine)

Other interviews CS Databases and Project Repositories:


SciStarter (U.S.)
Atlas of Living Australia (AU)
Potential External Stakeholders (not currently using CS data):
Environmental scientist (AU)
Health agency (U.S.)

*Projects included in the study according to Shirk et al.'s11 PPSR framework, scientific disciplines and regions (Australia =AU,
Europe = EU, United States of America = U.S.).

13
COMMONS LAB | STAKEHOLDER ANALYSIS

authors and Wilson Center staff to cover three authors then compared their research
several themes, including for each project: notes and early analysis to iteratively de-
a description of the project; stakeholders; fine the categories of stakeholders pre-
data generation and management prac- sented below, as well as highlight recurring
tices; degree of interoperability with other themes and major tensions. In addition to
projects; and the projects perspective on interview questions and research journals,
data standardization in citizen science (a a selection of exemplary and broad stake-
full interview protocol may be found in the holder groups from the literature review14
Online Appendix). Each interview lasted was used to prompt further reflection by
between 45 and 90 minutes. Responses the authors. The stakeholders mentioned
were recorded in a research journal, with by interviewees were then grouped and
initial results categorized by each author sample categories adapted.
individually using thematic analysis.13 The

While some barriers to data openness and


accessibility should be addressed, others highlight
the fact that (partial) non-openness is functional,
rather than dysfunctional, for the operation of
citizen science projects and thus should not
simply be treated as barriers to access.

14
Interview Participants

Europe United States Australia:


Barcelona, Spain Tucson, Arizona (2) Lismore, Australia (3)
Klagenfurt, Australia Philadelphia, PA Hobart, Tasmania, Australia
Leipzig, Germany San Francisco, CA Melbourne, Australia
Budapest, Hungary Boston, MA Canberra, Australia
Amsterdam, Netherlands Ithaca, NY Perth, Australia
Brussels, Belgium New Orleans, LA

Interview participants in the United States, Europe, and Australia. We invite future researchers to
expand the scope of inquiry to new geographic areas.

15
EXECUTIVE SUMMARY

16
RESULTS 3

STAKEHOLDERS OF ferent aspects of a project, performing


roles such as project leads, decision-
CITIZEN SCIENCE
makers, supporters, and/or users of the
PROJECTS data or knowledge) to a low level of
The citizen science projects in this study involvement (which describes stake-
reported a broad spectrum of stakehold- holders who participate in only one or
ers who engaged in project decision- two roles in the project).
making, provided support (in the form of
time, funding, expertise, equipment etc.), Stakeholder groups such as Civil soci-
and/or were users of the project data. ety organizations, informal groups and
Six stakeholder groups were identified community members, and Academic
in the interviews. These groups were: and research organizations reported
(1) Civil society organizations, informal the highest level of involvement in the
groups and community members; (2) citizen science projects in this study.
Academic and research organizations; Group (1) refers to formally organized
(3) Government agencies and depart- (incorporated) and less formally or in-
ments; (4) Participants; (5) Formal learn- formally organized stakeholders from
ing institutions such as schools; and, (6) civil society, including activist and ad-
Businesses or industry. vocacy groups, as well as individual
community members. Civil society or-
Nature of stakeholder ganizations, informal groups and com-
involvement munity members participated in differ-
ent ways for all governance models.
These stakeholder groups exhibited
This group was found to support, col-
varying levels of involvement, accord-
lect and/or use the data or knowledge
ing to how frequently they were men-
in all other governance models, and
tioned and their types of contribu-
lead or make decisions in collaborative
tions. The spectrum of involvement in
and co-created projects. In all gover-
projects ranged from a high level of
nance models, members of communi-
involvement (which describes stake-
ties, or community groups also collect-
holders who participate in many dif-
ed data, provided indirect support such

17
COMMONS LAB | STAKEHOLDER ANALYSIS

as access to land and used project data number of roles they played in the proj-
or knowledge. The study also observed ects was variable, and usually fewer than
that the community members/groups who the group leading the project.
participated in project decision-making
for collaborative, co-created and collegial The final two stakeholder groups were
projects also led many aspects of project more removed from the core functions of
design and implementation. For example, the citizen science projects. Formal edu-
in one collegial project, community mem- cational institutions were not involved in
bers decided what tools to use, received the contractual project. However, these
training online or through a local organiz- institutions supported contributory and
er, and then created their own projects, co-created projects (especially for data
which is the inverse of a contributory or collection) and used citizen science data
contractual project. Academic and other or knowledge from contributory, collabor-
research institutions were also highly in- ative, co-created, and collegial projects,
volved in citizen science and formed the primarily for educational purposes. Two
largest group of lead organizations for the projects included schools in decision-
projects, and were present in projects of making, one of which was a co-created
all governance models. As a stakeholder project, and the other a collaborative proj-
group, academic and research organiza- ect. Industry and business involvement
tions were also instrumental in decision- was absent in the large majority projects.
making and support, and made use of the However, they were more deeply involved
project data or knowledge. in some of the co-created projects, where
they assisted decision-making, provided
Government agencies and departments support, and used the data and knowl-
were mentioned as important stake- edge. These were also identified in the
holders in all governance models, with contractual project (as a data user) and
the goal of influencing policy outcomes. contributory projects (for support and
Compared to other stakeholders, the data use), yet were absent from the col-
presence of this third group in the citizen laborative projects interviewed.
science projects was moderate. Gov-
ernment involvement in citizen science Potential for engaging new
was strongest in contributory projects, stakeholders
where it was found to fund, lead projects,
Potential stakeholders who were not cur-
make decisions, and/or support and use
rently involved in projects or using citi-
the data/knowledge. Government sup-
zen science data were mentioned by all
port, decision-making and/or data or
governance levels except the contractual
knowledge use was also present in col-
project. The list of stakeholders for pos-
laborative and co-created projects, yet
sible future collaborations included Local,
appeared in the contractual and collegial
state, or federal government agencies or
projects only as a potential data user.
departments; Business or industry; Civil
Participants represented the volunteers15
society organizations, informal groups
who contributed to the projects across
and community members; and, members
all governance models. While these par-
of other Academic and Research Orga-
ticipants were present in all projects, the

18
INTERNATIONAL CITIZEN SCIENCE STAKEHOLDER ANALYSIS ON DATA INTEROPERABILITY | RESEARCH SERIES

nizations. The projects indicated they Internal and external


would like these stakeholders to provide stakeholders
support and/or use the data and knowl-
Stakeholders could be distinguished as
edge generated by the projects. Barriers
being internal and external to a given
to involvement included a lack of stake-
citizen science project based on whether
holders awareness of the projects (and in
interviewees considered stakeholders
some cases, a lack of the project teams
as part of the immediate project team
awareness of potential stakeholders), dif-
or as independent contributors or us-
ficulty in accessing or knowing how to
ers of results. Consequently, SciStarter
access potential stakeholders, and time
and the Atlas of Living Australia (actual
and resource constraints for doing so, as
stakeholders), the US health agency and
discussed in depth below.
the Australian scientist (potential stake-
holders), need to be considered external
stakeholders, since they were neither

Table 2: Matrix of stakeholder groups


Examples of Internal External
stakeholder Stakeholders form part of the Stakeholders are not considered
immediate project team and part of the immediate project team
types
research process(es) and research process(es), yet have a
identified by vested interest in the project
interviewees

Actual Partners/collaborators setting SciStarter using CS project


Stakeholders that the research question in a metadata.
are currently or collegial project. Atlas of Living Australia making
have been involved University researchers leading CS data accessible.
with a CS project and analyzing data from their Volunteers in a contributory
contributory CS project. project analyzing data.
Government department Local business informing
selecting location for CS customers on CS activity.
activity. Schools using CS activities in
NGO hosting a community formal education.
lab.
Neighbor committee or
community members dissemi-
nating project findings.
High-school students
doing CS as part of their
coursework.

Potential Federal and state agencies US health agency using CS data;


Stakeholders that funding project and using CS Australian scientist and other
have not been data. researchers using CS data.
involved with a CS Special interest group
project, yet could participating in knowledge
be in the future generation.

19
COMMONS LAB | STAKEHOLDER ANALYSIS

mentioned as part of a citizen science gather quantitative natural science data.


project by any of the other interviewees Even for projects in which quantitative
nor do they operate their own projects. data generation is the central aim, learn-
All six stakeholder groups listed contain ing along with co-creation methods and
internal and external stakeholders (Table other tangible outputs, such as program-
2). The distribution of internal and exter- ming code or physical objects, represent
nal stakeholders varied according to the valuable resources for potential reuse.
governance model. For instance, in co-
created and collegial projects, the proj- Across projects and geographies, shar-
ect team was usually much larger than in ing data, knowledge and procedures was
other governance models, and included generally perceived as important. The
various stakeholders in making decisions. value of data and knowledge to science,
These stakeholder groups were seen as the environment, volunteers, collabora-
external parties in contributory projects, tors, and society at large was underlined.
among them project participants and lo- Specific reasons for sharing included
cal authorities. As elaborated below, this enabling comparisons of citizen science
distinction is important for understanding data to official records; making the use
data sharing practices and promoting in- of data possible for everyone; improving
teroperability. data quality; facilitating cross-border and
long-term studies; fulfilling a societal re-
sponsibility as recipient of public funding;
DATA SHARING, increasing accessibility to low cost moni-
ACCESSIBILITY, USE & toring tools; enabling joint action; and,
REUSE raising awareness about important issues.
Projects collect and share heteroge-
neous types of data and knowledge. Sharing and accessibility of
These ranged from quantitative data data and knowledge
(e.g. on air quality, light pollution, biodi-
While all projects used various channels
versity), to qualitative information (e.g. on
to share their raw data, results, and/or
perceptions of air pollution and adapta-
procedures, the level of accessibility var-
tion strategies or consumption behavior,
ied. Some projects allowed unhindered
aerial mapping of land, human rights and
access to raw data sets, for example as
political conflict), to knowledge (e.g. on
files could be downloaded directly from
lab experiments, local cultural traditions
project websites or were shared through
or social transformation). Other relevant
discipline-based repositories and ar-
outputs generated by projects included
chives. Others only granted restricted
data on project participants, metadata
access to raw data, for instance, sharing
on citizen science projects, maps, pro-
them upon request or through online in-
gramming code, designs, objects, lab
terfaces that support analysis and inter-
experiments, institutions, education, and
pretation, sometimes following embar-
skills. In all cases more than one output
goes. Instead of or in addition to direct
was generated. This suggests that dis-
access to raw data, some projects shared
cussions on interoperability should not
processed data, study results, or proce-
be limited in focus to projects seeking to
dures through publication in academic

20
INTERNATIONAL CITIZEN SCIENCE STAKEHOLDER ANALYSIS ON DATA INTEROPERABILITY | RESEARCH SERIES

journals (not necessarily open access), the data was only shared and used by
public reports, online graphics, reposito- the community that generated the data, in
ries, or social media. In one case, due to hard copy format.
potential conflicts and local constraints,

The reasons interviewees offered for not making project data more
accessible included:

Technical difficulty and lack of technical capacity in the project team.

Sensitivity of subjective knowledge and personal data at the center of the


project.

Privacy considerations that might hamper participants willingness to contribute.

Difficulty finding journals that publish datasets.

Embargo periods ensuring privileged access to the research group for con-
ducting. analyses and publishing results before others can do so.

Perceived lack of interest in qualitative raw data.

Concern about impacts on high-value or threatened species.

Potential for political conflict.

Perceived lack of interest beyond actual internal stakeholders.

Language barriers.

21
COMMONS LAB | STAKEHOLDER ANALYSIS

While some barriers to data openness was built around the use of open source
and accessibility should be addressed, tools. The tools were openly available for
others highlight the fact that (partial) non- everybody to choose which to apply to their
openness is functional, rather than dys- own projects. In turn, tool users were en-
functional, for the operation of citizen sci- couraged to share their experiences within
ence projects and thus should not simply the community to improve the methodolo-
be treated as barriers to access. The gies, adding rigor and expanding the ap-
reasons for limiting accessibility mentioned plicability of the tools. Thus, accessibility,
above are rooted in the complex interplay use and reuse of knowledge and meth-
of disciplinary traditions, methodologi- odologies are ingrained in the function-
cal questions, project aims, and the de- ing of the projects and larger community.
sign and implementation of the project,
all of which merit further investigation. Use and reuse of data and
knowledge
Across all projects, results were usually
Examining how stakeholders are involved in
shared among internal stakeholders, and
project governance is also relevant for bet-
rarely with external stakeholders. A compari-
ter understanding the use of citizen science
son of project governance models illustrates
data as well as of knowledge and proce-
three aspects of how different conceptions
dures. In most cases, data use by external
of internal and external stakeholders mani-
stakeholders was limited, non-existent or
fest in nuanced sharing practices. First, raw
unknown. While some projects said there
data sharing was generally practiced within
were no barriers to their data, several be-
the project team, and less so with external
lieved that potential users were unaware
parties. Co-created projects usually gave
of the data or the project itself. Several co-
participants access to raw data (as well as
created projects stated they had already
to procedures and results) while only some
involved all relevant stakeholders as project
contributory projects did so. Contributory
partners (e.g. by co-designing study ques-
projects, in turn, were more likely to share
tions or choosing activity locations), whom
data with external stakeholders for ana-
they considered having an interest in using
lytical or environmental management pur-
the data, such as local or national authori-
poses. Second, for contributory research-
ties. Although these co-created projects
er-driven projects, sharing raw data was
give various stakeholders access to results,
usually understood as making project data
future (re)use by potential new internal or
available to the research community, while
external stakeholders is rather neglected
sharing processed data or results, especial-
in these cases. Some projects that empha-
ly in the form of graphics, maps, or report
sized sharing knowledge and procedures
cards was understood as a form of making
reported their methodologies had often
their data accessible to the public and/or
been used and requested by external par-
contributing participants. Third, the collegial
ties seeking to use the project or replicate
project provided yet another perspective of
co-creation methodologies and experimen-
how sharing knowledge - in this case tools
tal designs they developed. This highlights
and methodologies - is linked to the social
the importance of sharing project results
structure of citizen science activities. The
beyond simply data.
community from which the project stems

22
INTERNATIONAL CITIZEN SCIENCE STAKEHOLDER ANALYSIS ON DATA INTEROPERABILITY | RESEARCH SERIES

To understand the motivations of poten- any, they were undertaking to ensure proj-
tial external stakeholders for using citizen ect data or other information could be eas-
science data, interviews were conduct- ily shared.
ed with an employee of the federal health
agency in the U.S. and an environmen- What is currently being done
tal scientist in Australia. Neither inter- by projects and rationale
viewee was using citizen science data,
Interoperability as defined here was not
citing the main reasons as uncertainty
considered by most interviewees and if
about data quality assurance and qual-
so, only infrequently in the beginning of
ity control measures, and a lack of data
projects. At the same time, exemplary
standardization practices for environ-
work on making project data interoper-
mental sampling. The Australian scientist
able was conducted in some of the
also mentioned a lack of awareness of
surveyed projects. Interviewees from
citizen science data sets amongst sci-
eBird (U.S.), BioBlitz Barcelona (Eu-
entists generally and, more specifically,
rope), and Atlas of Living Australia said
her lack of knowledge about the location
they use established scientific, industry
of data sets, ownership and publication
and/or regulatory data standards, like
or access rights. Given these concerns
Darwin Core. For example, eBird data,
the U.S. health agency employee said
with a location and date, can go to a
they prefer to stick to conventional data
national data repository like the U.S.
collected following U.S. Environmental
Geological Survey (USGS) or National
Protection Agency (U.S. EPA) standard
Atmospheric and Space Administration
methods that have been published in
(NASA), pull the data and relate their
scientific literature. The Australian sci-
data to the location and information
entist said the adoption of standards
they have. Globe at Night, eBird, and
in citizen science would help alleviate
USA National Phenology Network are
some concerns about the influence of
already merging topic-based datasets
externalities, the procedures used, and
and have developed or are developing
allow for multi-region comparisons, all
interoperability infrastructures to do
of which is important for the type of re-
this. This approach is effective because
search this scientist undertakes.
these projects collect the same data
that can fall under a similar field, using
DATA STANDARDIZATION the same terminology. Limitations of this
AND INTEROPERABILITY approach include when data of diverse
types or from diverse sources need to
In the interviews, interoperability was de-
be combined, as highlighted when an
fined as the ability of different information
interviewee representing Globe at Night
technology systems and software appli-
discussed the National Park Service
cations to communicate, exchange data,
implementing a system to measure light
and use the information that has been
using an entirely different standard of
exchanged, which rests on the adoption
measurement, which creates difficulties
of data and metadata standards. Further
when attempting to work with the Na-
questions explored what interoperability
tional Park Service or combine data.
meant for projects, and what practices, if

23
COMMONS LAB | STAKEHOLDER ANALYSIS

Overall understanding ecological contributory projects were


and perceptions of using Darwin Core, an established bio-
interoperability diversity informatics data standard. Con-
In general, interoperability was poorly tributory projects were also working on
understood by most interviewees. Nev- the interoperability measures described
ertheless, across projectsregardless above. In addition, Hagit Keysar stated
of governance model, geography, or that her collegial projects benefit greatly
area of studycombining data sets and from the Public Lab website where staff,
using joint standards for certain types organizers and users post best prac-
of data and/or application cases was tices and standards on how to use the
generally perceived as valuable. Inter- technologies, and that this is important
estingly, most projects had not consid- for scaling up projects and making the
ered what interoperability means specif- projects and methods meaningful to
ically within the citizen science context. other stakeholders. It was stressed that
They reported using existing standards the Public Lab methods are informed by
for publishing data in disciplinary da- the users who contribute online or offline
tabases, government standards, or and that this iterative process is about
sharing documentation through online creating and facilitating ways to conduct
content platforms such as Wikipedia, collaborative work.
YouTube as methods they used to make
data interoperable with others. Others On the other hand, interoperability beyond
simply stated they use scientific meth- the core project team is not a priority for
ods of data collection and reporting some citizen science projects. One co-
that are accepted within their disciplin- created project stated it was more impor-
ary field. For projects that dealt with tant to ensure that local community mem-
qualitative data, or are more focused bers received the data in a format they
on knowledge sharing, data analysis, or could use than adhere to strict data stan-
devising lab experiments, it was unclear dards. Although the interviewee recog-
what data standardization could mean nized the benefits of combining data sets
for their fields of activity and how they through standardization, it is not a primary
could and should relate to the ongoing concern given their specific goals. How-
interoperability initiatives. Consequent- ever, a different co-created project stated
ly, although interviewees were aware that interoperability was not a concern,
of the value of sharing, knowledge on yet as the interview proceeded, the inter-
how to make project data and results viewee realized how much it would have
interoperable was often lacking. helped their own work if the local and
state government agencies and nonprofit
When comparing citizen science gover- organizations had standardized their data.
nance models, interviewees differed in
their knowledge about and prioritization Concerns regarding
of data standardization and interopera- interoperability
bility. In some cases, interoperability was The interviews revealed six main concerns
well understood. Various large-scale about interoperability in citizen science:

24
INTERNATIONAL CITIZEN SCIENCE STAKEHOLDER ANALYSIS ON DATA INTEROPERABILITY | RESEARCH SERIES

Photo Credit: Horseshoe Crab Citizen Science Project Cedar Key.

Florida Sea Grant

It's spring and that means horseshoe crabs are nesting in high numbers along many of
Florida's beaches. A new citizen science program in the Nature Coast is helping state
managers collect important data about horseshoe crab populations. This program is
a collaboration between Florida Sea Grant, UF IFAS Nature Coast Biological Station,
FWC, and UF's Biology Department. Sea Grant Agents like @scbarry are training citizens
to count nesting horseshoe crabs and apply tags to the crabs as part of a long-term
population study and you can help! Keep your eye out for tagged horseshoe crabs the
next time you are walking the beach or near the water.

Florida Sea Grant GIS Workshop May 2017

25
COMMONS LAB | STAKEHOLDER ANALYSIS

Limited applicability due to natural also worried how efforts to standardize


science bias. Several interviewees may create significant work for project
were critical of the fact that social sci- teams, specifically for their computer
ence standards remain largely absent information and technology specialists.
from discussions, highlighting that the The costs involved in implementing such
interface of any general citizen science changes may present a great challenge
data standard would need to be acces- for a considerable portion of the citi-
sible to a wide range of projects and zen science community since many of
research methodologies. One project the interviewees said they were oper-
interviewee argued for a more compre- ating on relatively small budgets (and
hensive conception of interoperability in some cases, in-kind support) for IT
pointing out that the one put forward in services. These budgetary constraints
the stakeholder analysis was focused will likely limit the uptake of interoper-
on the natural sciences in general and ability measures.
specifically on observational data, and
not applicable to all forms of knowl- Doubts about adaptability to local
edge. Further, for the projects work- circumstances. A co-created project
ing with qualitative data it was unclear expressed concern on how to develop
what such a standard could look like. standards and interoperability with
Another issue highlighted by interview- locally-based projects. While they
ees included the lack of clarity on how recognized there would not be an off-
to treat data gathered on participants the-shelf version that suited all projects,
including demographic information and they suggested a template could be
participant evaluations. provided. However, even if the template
was provided, this interviewee said they
Costs of adopting standardization would have to work with their community
protocols and modifications to ex- to build the infrastructure and refine it,
isting infrastructure. Projects were concluding that doing this is not easy
willing to adopt standards in principle as at a local scale.
long as they could be incorporated in a
way that did not fundamentally change Resistance to curtailing participa-
how the project currently operates, or tion and passing of burden. Some
change the usefulness of the data al- interviewees expressed concern about
ready collected. There was a great deal decreasing participation in projects aris-
of uncertainty regarding what data stan- ing from a requirement for participants to
dards for citizen science might look like, set up and/or follow a set of standards.
and how they might dictate necessary Interviewees said they would consider
changes to already established projects adopting a general standard for citizen
and their information and technology (IT) science data when certain precondi-
systems. Interviewees said for new stan- tions were met, including not limiting
dards to be adopted, they would need participation by requiring participants
to be easy to incorporate into existing to conduct significantly more work, or
data collection processes, and not ren- by other barriers to participation (such
der existing data useless. Interviewees as reduced privacy).

26
INTERNATIONAL CITIZEN SCIENCE STAKEHOLDER ANALYSIS ON DATA INTEROPERABILITY | RESEARCH SERIES

Fear of losing of relevant informa- standardization. However, a second


tion and decreasing data quality. One interviewee suggested Monarch Watch
interviewee with extensive experience in as a successful example where existing
data standardization stressed that un- project databases were merged to cre-
intended consequences can arise from ate a more robust dataset and perform
standardization; for example, the data rigorous analyses.
may become too general, richness is lost,
and metadata may become broader. This Difficulties in agreeing on concrete
interviewee said every project is collect- metadata terms. Where scientific re-
ing data differently based on the area of search is highly specialized, there was
research or discipline. They emphasized some question about the value of adopt-
that all researchers need to agree on the ing broad-reaching standards (e.g. set of
metadata fields that need to be included properties, elements, fields, columns, or
across projects, and only data that is attributes), which may not be appropriate
managed and curated with the same for certain projects. One interviewee
quality and rigor as their own projects can stated that the main metadata fields
be integrated. This interviewee thought needed to allow for data relatedness16
it was critical for projects to maintain are: who collected the data, where the
high standards of data management and data was collected, when the data was
quality and to be able to control for and collected, and how the data was col-
accurately describe all the bias in the lected. Currently, most projects do not
dataset this is more important than collect the how information.

27
EXECUTIVE SUMMARY

28
DISCUSSION 4

From the exploration of stakeholders, citizen science associations across the


data accessibility and use practices in cit- globe. As the momentum of citizen sci-
izen science projects, several topics have ence practice continues to grow, these
emerged that are relevant to standardiza- associations are well-placed to encour-
tion and interoperability discussions. age the community to think about and
implement interoperability mechanisms,
including data and metadata standards.
INTEROPERABILITY
IS ONLY SLOWLY The fact that interoperability measures
BECOMING A TOPIC are currently only implemented by the
OF CONCERN IN THE contributory and collaborative projects
CITIZEN SCIENCE surveyed in this study suggests a vast
COMMUNITY potential for reaching projects with
other governance models. In addition,
This became evident from the lack of
projects from the social sciences and
clarity about what data accessibility
humanities using qualitative method-
means and what activities are related
ologies were highly uncertain about the
to interoperability, as well as the con-
applicability and impact of standards
siderable variation in the understanding
to their field. This highlights challenges
of who owns the data generated in citi-
for standardization across scientific
zen science projects (ranging from the
disciplines. For qualitative data, there
community to only the project team).
needs to be more exploratory work
Clearly, there is a need for communi-
done on what standardization could
cation and training material on mak-
mean, why it would be useful, and what
ing citizen science data, methods and
it would look like. More research is also
knowledge open and interoperable as
needed to understand differences of
well as sharing of good practice. While
knowledge production across scien-
there is no overarching body respon-
tific disciplines. In addition, current ini-
sible for providing this service to proj-
tiatives on interoperability were largely
ects, this is perhaps an area that could
unknown to the interviewees, as illus-
be more comprehensively addressed by
trated by the fact that associations like

29
COMMONS LAB | STAKEHOLDER ANALYSIS

ECSA or CSA (who are supporting those dardized, not just with citizen science, but
initiatives) were not mentioned as stake- also between federal agencies, such as
holders by interviewees. These initiatives the EPA, USGS, and NASA, which often
need to become more transparent and in- have their own standards and procedures
clusive to engage more projects and sub- for collecting, storing and sharing data.
communities in citizen science. In particu-
lar, it would be useful to specify ways in
which interested projects could engage
A BROADER CONCEPT
in interoperability initiatives and reach out OF INTEROPERABILITY
to a broader community. This issue may IS NEEDED TO WORK
be best addressed through the support of ACROSS DISCIPLINES
citizen science associations themselves, AND PROJECT TYPES
as further discussed below.
Several interviewees criticized the con-
cept of interoperability employed in this
HETEROGENEITY OF DATA study for being too narrow in its focus on
SHARING PRACTICES information technologies, natural sciences
and quantitative data, which are typical
AND ADOPTION OF
characteristics of mainstream CS projects,
STANDARDS FOR and often contributory or collaborative
INTEROPERABILITY governance models. Such a perspective
Considering the breadth of citizen sci- misses the other various concrete out-
ence project designs and research puts generated in CS projects, such as
methodologies, the clearest benefits of tools, experimental design, and program-
interoperability appear to persist among ming code, as well as the impacts of some
projects collecting similar types of data citizen science projects, especially learn-
and working with the same media or spe- ing and transformative action. The later
cies. For biodiversity monitoring, a body are more common in projects that enable
of standards (such as Darwin Core) im- greater involvement of the participants in
proves interoperability and promotes con- decision-making and project design, for
nection to environmental health efforts, example co-created or collegial projects.
yet differences in the ways in which di- Pluralityincluding through regional,
verse types of data are collected, stored disciplinary, terminological, and method-
and shared means broad-scale adoption ological considerationsis necessary and
of interoperability will not be simple. Yet, enriching for science and consequently it
this is an issue that requires urgent atten- should not be reduced. This means that
tion, particularly since there has been lim- standards for citizen science data need to
ited uptake within CS practice, because enable, not infringe upon different forms
one of the great benefits of interoperabili- of plurality. In order to account for further
ty is the potential to contribute to avoiding developments in the field and to allow for
environmental disasters. Incidents like the various practitioner communities to con-
Flint, Michigan (U.S.) lead catastrophe tribute, standards should continue to have
might be avoided in the future if water formal review and revision processes in
quality data collection efforts were stan- place and in their architecture, as well as

30
INTERNATIONAL CITIZEN SCIENCE STAKEHOLDER ANALYSIS ON DATA INTEROPERABILITY | RESEARCH SERIES

To understand the motivations of potential


external stakeholders for using citizen science
data, interviews were conducted with an
employee of the federal health agency in the
US and an environmental scientist in Australia.
Neither interviewee was using citizen science
data, citing the main reasons as uncertainty
about data quality assurance and quality control
measures, and a lack of data standardization
practices for environmental sampling.

linkages between different sets and types ADVANCING


of standards. In addition to existing stan-
INTEROPERABILITY BOTH
dards in academic research and industry,
standards and workflows originating within
FACILITATES AND RESTS
the citizen science community itself, such ON INVOLVEMENT OF
as the approach to open source tools and STAKEHOLDERS
information embraced by Public Lab, need The study highlighted the various types of
to be taken into account and built upon. stakeholders, both internal and external to
Accessibility to tools and methodologies CS projects,17 which have interests in the
that are being standardized by the users production, management and/or use of
themselves is critical for facilitating com- the project data. Considerable restrictions
munity or civic science activities. This is on the use and sharing of data amongst
essential for groups that do not have the project stakeholders were detected,18
existing infrastructure, whether tools or which contrasts with an expectation of
programs, to advocate or investigate com- some that CS data should be open and
munity-identified issues. Bottom-up initia- easily accessible. The implementation of
tives and project types should be recog- interoperability measures will be essential
nized by other stakeholders and strongly in resolving this mismatch between ex-
considered in interoperability efforts. A pectations of some in the CS community
more inclusive concept of interoperability and the reality of CS practice.
is needed that is suitable for various forms
of knowledge including tribal ecological With a few exceptions, many of the inter-
knowledge and subjective and experiential viewees in this study think about interop-
knowledge. Interoperability should also al- erability as an issue of data usability rather
low for inter-generational exchange.

31
COMMONS LAB | STAKEHOLDER ANALYSIS

than an issue related to the engagement ects, and therefore needs to be consid-
of internal and external stakeholders. At ered for both directions of exchange.
the same time, potential stakeholders
such as the US health agency representa- It is clear that the implementation of in-
tive and the Australian scientist described teroperability measures has the potential
earlierreport difficulty in accessing the to extend sharing and reuse of data and
data, knowing that it exists in the first knowledge beyond internal project stake-
place, and whether data meets quality as- holders, where it is currently concentrated,
surance/quality control standards. Better towards more external and potential stake-
interoperability can help mitigate these holders (Figure 2). In turn, identification of,
concerns. While for some projects, data and consultation with, stakeholders at the
sharing and access beyond the core proj- planning stages of CS projects will help
ect team will not be important or practical, support interoperability. The same is true
data interoperability nevertheless remains for additional work on citizen science data
a major challenge for the progression and standards, which would benefit from the
impact of citizen science as a diverse in- involvement of both more diverse proj-
put into knowledge production and deci- ects and external stakeholders. Greater
sion-making. Interoperability also has the and more diverse stakeholder involvement
potential to make external stakeholders would enable broader and more effective
data and information needs more easily use of citizen science data, methodologies
understood and accessible to CS proj- and knowledge in the future.

32
INTERNATIONAL CITIZEN SCIENCE STAKEHOLDER ANALYSIS ON DATA INTEROPERABILITY | RESEARCH SERIES

Figure 2. Potential of interoperability measures for extending sharing of results with


stakeholders of CS projects. Most sharing of data and other outputs currently happens
among internal stakeholders of CS projects and less with external ones (blue circle).
Implementing interoperability measures (arrows) has the potential to facilitate more
sharing with external stakeholder and re-use by potential stakeholders.

Actual Actual
Internal External
Stakeholders Stakeholders

Potential
Internal
Stakeholders Potential
External
Stakeholders

LIMITATIONS OF THE facilitators or constraint of interoperabil-


ity related to government and research
STUDY
systems were not considered. By limiting
In this exploratory study, a total of sixteen the projects to the U.S., Australia and Eu-
interviewees representing sixteen citizen rope, this study excludes vast parts of the
science projects were recruited through a globe with long-established participatory
purposive sampling technique. This rela- research traditions as well as growing
tively small sample may not represent the citizen science activities. Language, other
full landscape of citizen science projects ways of knowing, and cultural models
and perspectives. Further, the two ad- such as traditional ecological knowledge,
ditional intervieweesone government were not considered when selecting proj-
employee, and one scientistare not ects. These are opportunities for future
sufficiently representative of potential ex- research.
ternal stakeholders. Due to the scope of
this research, international differences in

33
EXECUTIVE SUMMARY

34
OVERALL RECOMMENDATIONS
TO THE FIELD AND,FUTURE 5
RESEARCH

Based upon the work described here, stakeholders of CS projects. Trust and
recommendations are proposed to dialogue must be built with internal and
guide future citizen science interop- external stakeholders and their experi-
erability efforts. These recommenda- ences and necessities need to be taken
tions have been crafted and tailored for into account. Potential stakeholders at
citizen science projects, overarching various levels (local, national, interna-
interoperability initiatives, and citizen tional), especially regulatory agencies
science associations and networks. and government, should be included
into interoperability initiatives to express
Citizen Science Projects: Regard- their needs and contribute expertise.
less of governance model, it is impor- When developing common vocabu-
tant to identify actual and potential lary and interoperability protocols, it is
stakeholders (including partners to imperative to build in mechanisms for
support the project and/or benefit from review and extension for new issues
using the outputs), and have data and and communities in the future to make
knowledge management and sharing standards facilitate, not curtail, the in-
protocols in place to facilitate informa- novative potential of CS as a develop-
tion use and reuse. It is also critical for ing approach. Standards should also be
projects to maintain high standards for flexible and adaptable to various fields
data management and data quality, and of CS research, e.g. implemented as
to control for and accurately describe a modular framework like the Creative
biases in datasets. Projects should Commons (CC) licenses, which allow
consider seeking out information on ex- users to select the best choice from a
isting data standards, along with similar range of options. Existing community
projects for their experiences, and get standards should be leveraged where
support where needed. possible. Work needs to progress to-
wards a concept of interoperability that
Interoperability Initiatives: Current goes beyond data, natural sciences and
standardization and interoperability ini- information technologies and embraces
tiatives should become more transpar- plurality by being open and useful for
ent and open to internal and external various forms of knowledge, qualitative

35
COMMONS LAB | STAKEHOLDER ANALYSIS

methodologies, regional specificities, and embedded in the culture and expecta-


across different languages and genera- tions of citizen science communities. This
tions. will require conversations about the im-
portance of these measures, and the im-
Citizen Science Associations and pact they can have on wider issues, such
Networks: More capacity building is as leveraging information for knowledge
needed for citizen science projects to generation and decision-making, across
inform others in the citizen science com- all sectors and types of citizen science
munity about interoperability issues and activities, in addition to conversations
the benefits of greater interoperability in held within specialized data and meta-
citizen science. There are numerous ways data working groups.
in which associations and other networks
can facilitate greater adoption of data
standards and additional work on interop- FUTURE RESEARCH AND
erability in the future. These methods in- PRACTICAL NEEDS
clude hosting workshops and presenta- Future research is needed on the role of
tions, facilitating access to data experts, different stakeholders in citizen science,
and providing recommendations and both in relation to data standardization
guidance for best practices. In addition, efforts and in relation to other research
the associations may be able to assist and development activities. Other impor-
projects in identifying other similar proj- tant external stakeholders may include
ects that are already collecting standard- citizen science data repositories, project
ized, interoperable data, and facilitate in- databases and practitioner associations.
ter-project learning. Project identification Although this study included two citizen
systems, such as SciStarter and the Atlas science databases/project repositories,
of Living Australias Project Finder, may neither of these citizen science project
also be able to assist in matching projects types were mentioned as stakeholders by
according to data requirements. Finally, the other interviewees implementing proj-
associations could provide support for ects. As these actors lead citizen science
projects to conduct their own stakeholder data interoperability initiatives, we recom-
analyses, for example by providing advice mend their inclusion in future interoper-
such as toolkits on key aspects of proj- ability stakeholder analyses. To further
ect design. This will help citizen science complement and successfully expand
projects research diverse stakeholders this existing study, more research into the
needs, and ultimately maximize their im- perspectives of a broader set of stake-
pact. holders is warranted. The distinction be-
tween internal and external stakeholders
However, the provision of information on is offered as a heuristic for linking proj-
its own will not be enough to bring about ect governance models with data sharing
the considerable change in citizen sci- and interoperability practices, and should
ence practice that is required for broad- be explored further for empirical richness
scale adoption of data standards to pro- and conceptual depth.
mote interoperability. Practices, values,
and norms will need to be more deeply

36
INTERNATIONAL CITIZEN SCIENCE STAKEHOLDER ANALYSIS ON DATA INTEROPERABILITY | RESEARCH SERIES

Many of the citizen science projects provided through another channel, such
contributing to this research shared sig- as hackathon-style events to remediate
nificant practical limitations in terms of exiting data or database records, is also
access to resources such as funding needed. It is crucial that these resources
and technological expertise that would support the full range of citizen science
be required to advance the use of data projects, including small, growing, and
standards.19 Resources provided by the grassroots initiatives, in addition to the
associations, as suggested above, can larger and more established projects that
help mitigate knowledge gaps. How- enjoy privileged access to national and in-
ever, funding or technological expertise ternational level funding schemes.

There was a great deal of uncertainty regarding


what data standards for citizen science might
look like, and how they might dictate necessary
changes to already established projects
and their information and technology (IT)
systems. Interviewees said for new standards
to be adopted, they would need to be easy
to incorporate into existing data collection
processes, and not render existing data useless.

37
EXECUTIVE SUMMARY

38
ENDNOTES

1 V. Martin completed this study at methods for natural resource manage-


Southern Cross University, Lismore, ment.Journal of environmental
Australia. management90, no. 5 (2009):
1933-1949.
2 Cooper, Caren, Janis Dickinson, Tina
Phillips, and Rick Bonney. Citizen 7 Miles, Matthew, Huberman, Michael
science as a tool for conservation in and Johnny Saldana, Qualitative Data
residential ecosystems.Ecology and Analysis, Thousand Oaks, CA: SAGE
Society12, no. 2 (2007). Publishing, 1994.

3 Kullenberg, Christopher, and Dick 8 One interviewee reported on experi-


Kasperowski. What is citizen ences from two projects of same
science?A scientometric meta- governance model and discipline.
analysis.PloS one11, no. 1 (2016): They are treated as one project in the
e0147152. analysis.

4 Citizenscience.org. PPSR_ 9 Shirk, Jennifer, Heidi Ballard, Candie


CORE Metadata Standard. Wilderman, Tina Phillips, Andrea Wig-
October 9, 2015 <http:// gins, Rebecca Jordan, Ellen McCallie
citizenscience.org/2015/10/09/ et al. Public participation in scientific
ppsr_core-metadata-standard>. research: a framework for deliberate
design.Ecology and Society17, no.
5 Shirk, Jennifer, Heidi Ballard, Candie 2 (2012).
Wilderman, Tina Phillips, Andrea Wig-
gins, Rebecca Jordan, Ellen McCallie 10 Follett, Ria, and Vladimir Strezov.
et al. Public participation in scientific An analysis of citizen science based
research: a framework for deliberate research: usage and publication pat-
design.Ecology and Society17, no. terns.PloS one10, no. 11 (2015):
2 (2012). e0143687.

6 Reed, Mark S., Anil Graves, Nor- 11 Shirk, Jennifer, Heidi Ballard, Candie
man Dandy, Helena Posthumus, Wilderman, Tina Phillips, Andrea Wig-
Klaus Hubacek, Joe Morris, Christina gins, Rebecca Jordan, Ellen McCallie
Prell, Claire H. Quinn, and Lindsay et al. Public participation in scientific
C. Stringer. Whos in and why? research: a framework for deliberate
A typology of stakeholder analysis

39
COMMONS LAB | STAKEHOLDER ANALYSIS

design.Ecology and Society17, no. 2


(2012).

12 Joint Research Centre (JRC) of the


European Commission. Citizen Sci-
ence: data and service infrastructure
meeting. January 26-27, 2016 <https://
ec.europa.eu/jrc/en/event/workshop/
citizen-science-workshop>.

13 Bryman, Alan.Social Research Meth-


ods (4th Edition). Oxford University
Press, 2012.

14 For instance see Durham, Emma, Helen


Baker, Matt Smith, Elizabeth Moore, and
Vicky Morgan. The BiodivERsA stake-
holder engagement handbook.Biodi-
vERsA, Paris(2014): 108.

15 In the case of the collegial project,


participants were specifically called
collaborators rather than volunteers.

16 See What is currently being done by


projects and rationale section above.

17 See also Tiago, Patricia. Social


Context of Citizen Science Projects.
In: Ceccaroni, Luigi and Jaume Piera
(editors). Analyzing the Role of Citizen
Science in Modern Research. IGI
Global (2016): 168-191.

18 Groom, Quentin, Lauren Weatherdon,


and Ilse R. Geijzendorffer. Is citizen
science an open science in the case
of biodiversity observations?.Journal
of Applied Ecology54, no. 2 (2017):
612-617.

19 See also Wiggins, Andrea. Free as in


puppies: Compensating for ICT con-
straints in citizen science. Proceedings
of CSCW 2013. ACM Press (2013):
1469-1480.

40
41
One Woodrow Wilson Plaza
1300 Pennsylvania Avenue, N.W.
Washington, DC, USA 20004-3027
202-691-4000
www.wilsoncenter.org

The Commons Lab of STIP seeks to


mobilize public participation and innovation
in science, technology and policy.
http://bit.ly/CommonsLab

The Commons Lab is supported by


the Alfred P. Sloan Foundation.

You might also like