Community Quality-of-Life Indicators: Best Cases VIII [1st ed.] 9783030481810, 9783030481827

Table of contents :
Front Matter ....Pages i-vi
Creating and Sustaining Community Indicators Projects: From Engagement to Results (Chantal Stevens, Lyle D. Wray)....Pages 1-14
Goldilocks Data-Connecting Community Indicators to Program Evaluation and Everything in Between (Frank Ridzi)....Pages 15-35
Strategies for Expanding Indicator Profiles to Small Rural Geographies (Jacob Wascalus, Ellen Wolter)....Pages 37-46
Measuring the Dream for an Equitable and Sustainable Future (Katie O’Connell, Andrea Young, Nisha D. Botchwey)....Pages 47-64
Meaningful, Manageable, and Moveable: Lessons Learned from Building a Local Poverty Index (Jamison Crawford, Frank Ridzi)....Pages 65-88
The Development of DISC (Decision Integration for Strong Communities): An Agile Software Application of Sustainability Indicators for Small and Rural Communities (Kevin Summers, Viccy Salazar, Dave Olszyk, Linda Harwell, Allen Brookes)....Pages 89-113
Indicators Supporting Public Health, Partnership, Liveability and Integrated Planning Practice: The Case Study of the Cardinia Shire Growth Area in Melbourne, Australia (Melanie Davern, Petrina Dodds Buckley, Pieta Bucello)....Pages 115-135
Five Conditions Conducive to Sustainability Plans and Measurements (Jim Powell)....Pages 137-162
Leveraging Data for Meaningful Improvements: How Credible Data Enables Partnership Alignment to Achieve Well-Being at the Population Level (Susan Brutschy, Keisha Frost, Michelle Luedtke, Donna Maurillo)....Pages 163-183
Data Parties: Giving the Community Tools to Use East Metro Pulse Survey Data (Emma Connell, Sheila Bell, Nicole MartinRogers, Nadege Souvenir)....Pages 185-198
Data-Driven Decision Making and Community Indicators: Towards an Integration of DDDM in Community Development (David Abraham)....Pages 199-210
Back Matter ....Pages 211-214

Citation preview

Community Quality-of-Life and Well-Being

Frank Ridzi Chantal Stevens Melanie Davern   Editors

Community Quality-ofLife Indicators Best Cases VIII

Community Quality-of-Life and Well-Being Series Editor Rhonda Phillips, Purdue University, West Lafayette, IN, USA

The Community Quality of Life and Well-being book series is a collection of volumes related to community level research, providing community planners and quality of life researchers involved in community and regional well-being innovative research and application. Formerly entitled, Community Quality of Life Indicators: Best Practices, the series reflects a broad scope of well-being. Next to best practices of community quality-of-life indicators projects the series welcomes a variety of research and practice topics as related to overall community well-being and quality of life dimensions, whether relating to policy, application, research, and/or practice. Research on issues such as societal happiness, quality of life domains in the policy construct, measuring and gauging progress, dimensions of planning and community development, and related topics are anticipated. This series is published by Springer in partnership with the International Society for Quality-of-Life Studies, a global society with the purpose of promoting and encouraging research and collaboration in quality of life and well-being theory and applications.

More information about this series at http://www.springer.com/series/13761

Frank Ridzi Chantal Stevens Melanie Davern •

•

Editors

Community Quality-of-Life Indicators Best Cases VIII

123

Editors Frank Ridzi LeMoyne College and Central New York Community Foundation Syracuse, NY, USA

Chantal Stevens Community Indicators Consortium Issaquah, WA, USA

Melanie Davern RMIT University Melbourne, VIC, Australia

ISSN 2520-1093 ISSN 2520-1107 (electronic) Community Quality-of-Life and Well-Being ISBN 978-3-030-48181-0 ISBN 978-3-030-48182-7 (eBook) https://doi.org/10.1007/978-3-030-48182-7 © Springer Nature Switzerland AG 2020 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Contents

1

2

3

Creating and Sustaining Community Indicators Projects: From Engagement to Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chantal Stevens and Lyle D. Wray

1

Goldilocks Data-Connecting Community Indicators to Program Evaluation and Everything in Between . . . . . . . . . . . . . . . . . . . . . . Frank Ridzi

15

Strategies for Expanding Indicator Profiles to Small Rural Geographies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jacob Wascalus and Ellen Wolter

37 47

4

Measuring the Dream for an Equitable and Sustainable Future . . . Katie O’Connell, Andrea Young, and Nisha D. Botchwey

5

Meaningful, Manageable, and Moveable: Lessons Learned from Building a Local Poverty Index . . . . . . . . . . . . . . . . . . . . . . . Jamison Crawford and Frank Ridzi

65

The Development of DISC (Decision Integration for Strong Communities): An Agile Software Application of Sustainability Indicators for Small and Rural Communities . . . . . . . . . . . . . . . . . Kevin Summers, Viccy Salazar, Dave Olszyk, Linda Harwell, and Allen Brookes

89

6

7

Indicators Supporting Public Health, Partnership, Liveability and Integrated Planning Practice: The Case Study of the Cardinia Shire Growth Area in Melbourne, Australia . . . . . 115 Melanie Davern, Petrina Dodds Buckley, and Pieta Bucello

8

Five Conditions Conducive to Sustainability Plans and Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 Jim Powell

v

vi

Contents

9

Leveraging Data for Meaningful Improvements: How Credible Data Enables Partnership Alignment to Achieve Well-Being at the Population Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 Susan Brutschy, Keisha Frost, Michelle Luedtke, and Donna Maurillo

10 Data Parties: Giving the Community Tools to Use East Metro Pulse Survey Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 Emma Connell, Sheila Bell, Nicole MartinRogers, and Nadege Souvenir 11 Data-Driven Decision Making and Community Indicators: Towards an Integration of DDDM in Community Development . . . 199 David Abraham Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211

Chapter 1

Creating and Sustaining Community Indicators Projects: From Engagement to Results Chantal Stevens and Lyle D. Wray

Abstract Sustaining indicator projects over time is an important challenge to address since taking effective action and tracking success on a given issue may take some time. Project sustainability is not about self-preservation but maintaining the capacity to serve and inspire the community for as long as improvements are needed and to continue to monitor over time. Care must be taken at each step of the development and maintenance of a community indicators project to insure its ability to stand the test of time. Successful community indicators projects share similarities: support from the initiating organization leadership and a host of partners; early and continuous engagement of the community and partners; selection of a solid framework and effective indicators; participation of stakeholders, decision-makers and subject matter experts; a willingness to conduct periodic evaluations and to innovate in response to community interests; good leadership and communications skills. This section includes multiple examples of practices and tools or exemplars that illustrate the relevance and strength of today’s community indicators universe.

1.1 Introduction The mission of the Community Indicators Consortium (CIC) is for community indicators to be used by all communities to facilitate sustainable improvements in their quality of life. Since 2005, CIC has offered resources and tools to help communities and practitioners advance the practice and effective use of community indicators. CIC maintains a popular database of over 300 projects located all over the world and has observed the rise and fall of dozens of projects each year.

C. Stevens (B) · L. D. Wray Community Indicators Consortium, P.O. Box 260, Issaquah, WA 98029, USA e-mail: [email protected] L. D. Wray e-mail: [email protected] © Springer Nature Switzerland AG 2020 F. Ridzi et al. (eds.), Community Quality-of-Life Indicators, Community Quality-of-Life and Well-Being, https://doi.org/10.1007/978-3-030-48182-7_1

1

2

C. Stevens and L. D. Wray

What we learned from our observations is that it is easier to create a community indicators project than to maintain it over the long term. As is true for other voluntary and community efforts, many community indicator projects are all too often short-lived. New reports or dashboards appear on the scene with a burst of staff and volunteer enthusiasm, dedication, and skill, and then fade away. Even projects that have survived for a good while struggle to remain relevant. Sustaining indicator projects over time is an important challenge to address since, in most cases, taking effective action and tracking success on a given issue may take some time. Maintenance over decades is often essential if the project is to ignite, support and evaluate progress on actions designed to improve community conditions. If a project is not around long enough for results to be seen, the benefit of much of the entire effort may be lost.

1.2 Understanding Community Indicators Community indicator projects use data, often displayed over time or comparing different locations, to tell the story of complex systems and to guide priority and agenda setting for groups involved in improving community-level conditions across the full spectrum of challenges affecting the community. Communities, whatever their scale, are complex systems made up of many components, and various levels. To understand, predict, and improve a system, community indicators need the participation and support of the community they aspire to describe. Indicators should also be logically or scientifically defensible, which speaks to the need to include various experts in the selection and interpretation of indicators. Finally, involving stakeholders with authority or resources, and the will and ability to affect the chosen changes, in the process of identifying priorities and selecting the correct measures will help to ensure that community indicator projects are used to guide and support strategies or action. Success for a community indicator project can be defined as the ability to improve outcomes, as measured by those very indicators that are expected to spur and guide the improvements. But changes in those outcomes may take years or may be masked by other trends. Meanwhile, community indicators bring a variety of less quantifiable, more subtle benefits such as: telling a compelling story about lesser known areas of the community, helping to understand complex issues, creating a common language for action, creating awareness for inequities within a community, bringing the community physically together and creating bonds and networks, informing policy-makers, or training future leaders among its volunteers.

1 Creating and Sustaining Community Indicators …

3

1.3 Exemplary Practices In 2018, CIC developed a Community Indicators Project Development Guide with the support of the William K. Kellogg Foundation. As the basis for its research, CIC interviewed the managers of ten successful1 community indicator project, conducted a review of 30 webinars that focused on the operations and success of currently active community indicators projects, and reviewed the literature on community indicators. What follows is a summary of good practices that emerged from that work.

1.3.1 Getting a Solid Start An important early step for any community indicators project is for its sponsoring organization, its various partners and its main funders to develop a common understanding of expectations, deliverables, resources and time commitment needed from the community and from the organization to support the project. For ACT Rochester, the first thing to do was to decide what the indicators project were going to do and who its key audience was (Johnson 2018). The Richmond Regional Indicators Project wanted to ensure that there was a need for their indicator project. An assessment showed that while there were a few organizations offering topic-specific data and other groups examining geography-specific data, there really wasn’t any entity offering the breadth and depth that they hoped to offer and no other organization could tie metrics to analyze and then to create opportunity for action (Harris 2018). Early and clear identification of the purpose and resources needed for a project can avoid later pitfalls. It is important at the inception of the project to consider the strength of the commitment of the home organization and the project’s main partners and their appetite for long-term involvement. Here again, the Richmond Regional Indicators Project made sure sustainability was part of the conversation early on. They worked with philanthropic and academic partners to leverage funds, other staff resources and data expertise. Once they established a need and sustainable funding, they then were able to move forward (Harris 2018). An appropriate organizational host can provide a solid foundation for an indicators project and avoid the enormous effort of starting a nonprofit organization from scratch. Community foundations, for example, often have a mission that closely aligns with that of a community indicator project and often can serve as an appropriate host. Similarly, some academic research institutes might be a good fit for a project. 1 Success here was determined subjectively by the team based on output factors such as the project’s

longevity, resources associated with the project, overall reach, level of activity among their peers and/or visibility in the literature rather than on outcomes, although most of those projects were also the recipients of CIC Impact Awards.

4

C. Stevens and L. D. Wray

Building a broad coalition from the outset may mitigate potential drawbacks associated with a project’s original home. Such a coalition is likely to offer the strongest support both initially and over time and can be a key factor in long term sustainability of a community indicator system (Barrington-Leigh and Escande 2018). Focusing on influencers and decision-makers from the beginning will help build the community support to move from problem recognition to action (Pettit 2018).

1.3.2 Engaging the Community A critical part of the initial process is to identify who will be engaged in a project, at what times, to what extent, and in what roles. The ability to engage the community and stakeholders is critical to the success of an indicator project and getting input and buy in early in the process is vital (Pettit 2018). Early consultation with the community will assure that the purpose of the project is well aligned with community desires and expectations. Diversity and inclusiveness are imperatives throughout the process. Who is around the table when priorities are set; who are the partners in implementation and who gets to decide whether a successful outcome has been achieved in a community are all considerations that will define whether the community views the product of this efforts as reflecting their culture and identify or the priorities of the researchers. Community members want results they care about, not just data, so their priorities should drive what data to collect, report, and use (Epstein et al. 2016). For Minnesota Compass, all users and stakeholders are partners in crafting something that’s usable. “If we operate in a vacuum as researchers, giving what information we think is best, then we’re going to fail in terms of sustainability, relevance and usefulness” (Liuzzi 2018). Assessing the organization’s standing in and connections to the community and other organizations within the community ensures that it has enough ability to identify and move the needle on issues that matter. If these relationships are not sufficiently strong, then the organization can consider taking on another role such as data partner and identify a partner organization with better standing in the community as its community face. Regular engagement and reporting updates and findings will keep the community feeling connected and involved in shaping its own future. Identifying and involving stakeholders from the outset can be key to the success of an initiative. Such engagement is needed in the visioning process, on steering committees, as subject matter experts, and in indicator selection groups. Subject matter experts are particularly important during the indicator selection process where they can contribute scientific or technical background to ensure that the indicator is logically connected to the goal or priority and is scientifically and technically sound. Their scientifically or technically informed views complement the practical, on-the-ground life experiences of community members and the advocacy from various stakeholders.

1 Creating and Sustaining Community Indicators …

5

1.3.3 Selecting Meaningful and Relevant Indicator at the Appropriate Scale Indicators are usually organized under a framework. Frameworks may be existing2 or newly developed for a project. An existing framework can be used as a starting point for a community discussion or as a way to check for gaps in the community vision. A new framework may be created to adapt to the specific circumstances or goals for a particular community. A framework is usually made up of domains, which are broad content areas that will serve as the support structure for the indicators. Domains should relate directly to themes that emerged as part of the original conversation about indicators. Using goals as domains names, e.g., “clean and sufficient water for all”, “healthy children”, or “affordable quality housing”, highlights these areas and demonstrates a commitment to realizing these aspirations. The selection of several indicators under each domain is another opportunity for a cooperative process. Compass uses Topic Advisory Groups, or TAGs, and starts by identifying a convener or co-conveners; for example, a project leader and a key community leader are an ideal pairing. Invitation lists of community members representing different sectors, ages, ethnicities, etc. are compiled. At the first of the two TAG meetings, participants are asked to weigh in on broad questions, e.g., What matters most to you? What do you need to know to make changes? Following that meeting, the project staff reviews notes from the discussion, looking into data sources for potential indicators, and creating a “why/why not” document that goes through every idea from the TAG meeting and provides a rationale for why an indicator should or should not be used. Then, in the second TAG meeting, a proposal for three to four indicators per domain is shared, and the TAG brainstorms how to connect the community to the indicators (Liuzzi 2018). Truckee Meadows Tomorrow (TMT) began with 9 major conceptual areas. A task force met with different groups of community members, reaching over 2000 residents. To get the word out about online surveys and events, they partnered with various organizations who had well-read listservs or large e-mail lists. TMT provided citizens with possible indicators and had citizens indicate which indicators they feel best addressed the topic. TMT came up with a community engagement process to get feedback on what quality of life meant to different people. From a list that had been made by community members, a broader audience was asked to vote—using Monopoly money—on what issues were most important. Every person got $100 of Monopoly money to allot how they wanted (Hruby 2018). Criteria can help sort out those indicators that are most appropriate for inclusion and discard other indicators that are not as good a fit. The criteria can address relevance, strength, and availability of data. For each proposed indicator, consider the following questions: • What are we trying to measure and why? 2 See

for example the United Nations Sustainable Development Goals, the Australian National Development Index (ANDI), the Canadian Index of Wellbeing (CIW), Healthy People 2020.

6

C. Stevens and L. D. Wray

• What does research show about the link between the indicator and the domain or goal? • Does this indicator measure an input, output, or outcome? • Are past, current and future data available for this indicator at the desired scale? • How often could this indicator be updated? • Can anything be done to move the needle on this indicator (is it actionable)? Actionable indicators, or leading indicators, are measures that can be moved by future actions, as opposed to lagging indicators that show what has already happened. It is important to consider the time frame at which to measure an action. Numbers on adult literacy, for example, may take a generation to improve, but tracking readiness for school in young children can inspire focused and immediate action. When considering the appropriate scale for the data, it’s important to consider whether reporting only one set of data at the province or state, county, or watershed scale provides enough information. Will the larger scale mask disparities or impacts among different sub areas or populations? Typically, the smallest geography available is the most useful in diagnosing issues to address, although availability of data at the level of specificity needed by the community may be the limiting factor. As a rapidly developing and growing suburb dealing with a wide range of social and health issues needed indicators, the community of Cardinia Shire (Victoria) needed neighborhood level livability indicators that would not mask hidden health disparities within small areas across municipalities. As described in this volume, the neighborhood level indicators were particularly useful to Victorian public health officials planning legislation and also led to the formation of a partnership of local agencies (Davern et al. 2020). Also in this volume, Ridzi (2020) offers a thorough discussion on the usefulness of measures at different scales, from regional to address-level, building on the work of the Central New York Foundation and CNYVitals. He argues that, by paying close attention to each actor’s target outcomes (and the data layer they use to measure them), it is possible to see how these levels of data help to coordinate actors that naturally inhabit different tiers of action. Finally, indicators should be looked at through different lenses that reflect the social and cultural communities. Disaggregating data by sex, income levels, rural/urban residence, or ethnic, racial, or cultural group affiliation can reveal important differences, especially for communities that aim to address equity or social justice. The Measure the Dream index is developed to uncover racial and ethnic differences to ensure equitable outcomes for all, recognizing that working towards equality can support a community’s cohesion, resilient, and sustainability (O’Connell et al. 2020, in this volume). Also relevant to deciding at what scale to report data is knowing at what scales decisions or policies made for that area; what scale will resonate with community members and help draw them into taking action; and, whether data and resources are available to acquire data at smaller scale. In addition to the raw data obtained, each indicator needs background information (metadata) comprising a few paragraphs that explains why the indicator is important

1 Creating and Sustaining Community Indicators …

7

and how it relates to the vision or goals. This description should also include details about the source and timing of the data and any information related to its strengths or limitations. This information is necessary for transparency, use, and building an organizational memory of decisions and knowledge that can outlast any particular person in the event of personnel changes in the indicators project. If the data does not exist to develop an indicator that was strongly supported by the community, proxy measures that resonate with the community can be identified. Sustainable Calgary was asked to measure cultural diversity. After convening a panel of experts, United Way, university researchers, and foundation, they identified “Diversity among positions of power and influence” as a proxy measure for cultural diversity and Sustainable Calgary was able to collect data and report on this measure. When the community said they cared about public events and parties but measures to track occurrence of those events was not available, the Baltimore Neighborhood Indicators Alliance (BNIA) figured out that every time a block party happened, a permit from the city was needed, so they worked with the city to track the number of permits it granted. It’s an important indicator of activity in the public space (Iyer 2018). Keeping indicators fresh though periodical evaluations, review and updates of the list of indicators is key to project sustainability and, most importantly, to their relevance to the community. Some projects may go as far as reorganizing their indicators under a new framework that is based on more recent science, such as the Social Determinants of Health (Brutschy et al. 2020; Davern et al. 2020) or a new focus for the community as a result of a major community trauma (Iyer 2018) or resilience following a natural disaster (Gardere 2018). For the Jacksonville Community Council, Inc. (JCCI), one of the longest living indicator projects in the US, something had to be done when the project had grown to over 180 indicators. Through a large community initiative that involved 16,000 voices, residents prioritized issues and highlighted areas that they wanted to see progress in. Through this effort, they reduced the number of indicators that were publicly maintained to 50 indicators, knowing that those indicators were aligned with the community’s vision. This helped them emphasize the relevance of what the trend lines were showing (Cohn 2015). Projects that excel are constantly keeping their ears to the ground to understand evolving or changing priorities. They position themselves ahead of the curve regarding new problems, commit to periodic evaluations, and have multiple channels of communication with their intended audiences. They know and understand the community and the community knows and understands them.

1.3.4 Building Bridges Between Project and Users Just putting out a report or an updated dashboard is not enough. “You can come up with the best indicators for the community, but if the community doesn’t know about them or what to do with them, it’s useless” (Hruby 2018).

8

C. Stevens and L. D. Wray

Successful projects emphasize continuous and direct engagement with the community and key stakeholders. The data part is easy; thinking early about a mechanism for this information to change the way people are thinking about things is where most of the time should be spent (Pettit 2018). Getting the right information to the right person at the right time and making the data become a key part of everybody’s daily life is essential (Iyer 2018). Many projects emphasize the importance of live events and presentation as part of the reporting process. Some host their own events, while others mainly respond to requests for presentations. Many attend meetings and share the indicators during conversation. For BNIA, making data a key part of everybody’s almost daily life and helping people understand the relevance of data has been an important part of their work. Supporting the written word with personal engagement is critical: a high level of interaction between indicators practitioners and users is likely to lead to greater data use. They attend community and municipal events that provide opportunities to create awareness about the indicators and to make a case about the relevance and importance of the data (Iyer 2018). For Minnesota Compass, communications means being part of communities that, they hope, will use the indicators. They engage in all sorts of forms of outreach to make sure people know who they are–their identity, what they do and don’t provide– and rely on a strategic communications plans to target specific outcomes (Liuzzi 2018). Being intentional helps to pursue and accept opportunities. Projects may keep a record of the events they attend, splitting them out by topic area to understand which topics are resonating the most. This record can support planning for where to focus future efforts. Being ready and eager to strategically use different media to keep the project and its data in the news is critical. If a journalist or local radio DJ send out an email looking for feedback on a particular topic, ACT Rochester will work on their timeline and be ready with the data journalists are looking for Johnson (2018). To help partners and users use their data, CNY Vitals started running monthly Data Fridays, inviting anyone who want to be better at managing and using their data, identifying outcomes, building databases, etc. to join this support group (Ridzi 2018). After seeing community members struggle to use the results of their survey, East Metro Pulse organized data parties to bring users of data together to learn how to read a data book and communicate what they learned to their stakeholders (Connell et al. 2020, in this volume). Every year, BNIA hosts Baltimore Data Day, a free and open workshop to help communities expand their capacity to use technology and data to advance their goals. Structured around a series of “how to” interactive workshops, Data Day brings Community leaders, nonprofit organizations, governmental entities and civic-minded “hackers” came together to see the latest trends in communitybased data, technology and tools (BNIA, n.d.). At least 25% of attendees identify themselves as being community people (Iyer 2018).

1 Creating and Sustaining Community Indicators …

9

1.3.5 Surviving Threats to Community Indicator Projects Over time there is a natural drain of resources and energy and a buildup of fatigueinducing habits that have the power to erode the relationships established with the community, funders, and volunteers. Volunteer fatigue can affect the vitality of a community indicator project, but this can be counteracted by good practices related to the “care and feeding” of volunteers. Care must be taken to nurture and expand relationships. Each new stage in a project is an opportunity to re-engage, celebrate, deepen and strengthen the connections with existing partners, and create new ones. Maintaining a strong network of partners helps soften the blow inflicted by internal or external strife. Diversity, inclusion and inclusiveness are a lens by which to view community indicators and serve as benchmarks for overall success for community efforts. Beyond equitable participation in the indicators work, these concerns mesh with evidence based practice—what actually works to produce improved outcomes as viewed through an equity lens. Developing equity agendas for key elements of the process, and guidelines for effective practice should be front and center as the indicators project moves ahead. While regular and meaningful communications must take place with community members, stakeholders, funders and policy-makers throughout the life of the project, the website or other communications media can be used to strengthen partnerships. “Being a volunteer network, often we have to compete with other priorities of partner organizations. Since we don’t directly provide funding for our partners, we have to offer something that’s of value to each organization. We highlight partners, communicate success stories, and track partner engagement over time” (Joo 2018). Similarly, SA 2020’s website is a hub for 145 nonprofit partners with a microsite for each partner, to show who they are, but also to provide a space for them to talk about their connection to the SA2020 community vision. SA 2020 points website visitors to either volunteer, or get more information, or even donate to those organizations (Fox 2016). Projects that are overly associated with one charismatic leader, dependent on a person that holds most of the knowledge and connections, or that are the “pet project” of a funder or elected official may not survive if that person leaves the scene. Applying good leadership practices and building strong partnerships should prevent this from happening. Decreased or lost funding is usually the single most difficult problem to solve. When funding goes down, so does the capacity to support a successful project. Funding is influenced by the health and stability of the lead organization and the diversity of its revenue sources, the level to which the local community, and its funders, understand the value of indicators, and the strength of the partnerships that were established. To respond to funding cuts, the community indicator project may reduce staffing, research and outreach, which can in turn lead to decreased interest in the project on the part of funders thus creating a downward spiral. Careful

10

C. Stevens and L. D. Wray

understanding of funding considerations, potential sponsors, and resource needs is essential before starting a project. Projects facing threats, such as those described above, may cut down on outreach and engagement to double down on research and analysis, eventually becoming the proverbial “ivory tower,” that is, an entity separated from the community and practical realities. This may work for a while as there may be a specialized audience that is receptive to the well-researched product, but it will not be as effective in the community, and eventually the project will lose the connections that allows it to identify what matters to, and therefore what can be done in, a community. Using indicators to spotlight areas of community pride, as well as its deficits can spur the community to come together to celebrate. Measures that track progress help generate funding or policies. An engaged community can offset lack of dedicated funding. In Juneau and Anchorage, community-led efforts to track sustainability measures lead to implementation of initiatives that advance the sustainability of the community and (in this volume). As described in this volume, the Santa Cruz County Community Assessment Project (CAP) now in its 25th year, has stood the test of time and continues to spur action and results through committed leadership, a rigorous community engagement process, a willingness to improve and innovate and apply new models, and a distributed funding approach (Brutschy et al. 2020). Projects with longevity have developed the ability to refresh and evolve. They evaluate periodically and strategically if their indicators are effective, if the people at the table are still the right ones, if they are still using the best delivery method for the data (Pettit 2018).

1.4 Conclusion A main goal of community indicators projects is to tell a meaningful story that can lead to sustainable improvements in community conditions. Informing the community, changing minds and effecting change takes time. Telling such a story requires planning, good data, time, and resources, as well as an understanding of and commitment to community engagement. Project sustainability is not about self-preservation but maintaining the capacity to serve and inspire the community for as long as improvements are needed and to continue to monitor over time. Care must be taken at each step of the development and maintenance of a community indicators project to ensure its sustainability. A project should be initiated with special attention to its ability to stand the test of time. Hallmarks of a successful community indicators project includes: support from the initiating organization leadership and a host of partners; early and continuous engagement of the community and partners; selection of a solid framework and effective indicators; participation

1 Creating and Sustaining Community Indicators …

11

of stakeholders, decision-makers and subject matter experts; a willingness to conduct periodic evaluations and to innovate in response to community interests; good leadership and communications skills. The rapid growth of the community indicator field and the contributions to community wellbeing across the globe provide a hopeful backdrop to the remainder of the book that details many of the “how to” aspects of the field as well as lessons learned born out of trials and applications. Chapters 2 and 3 scope the geographic levels at which to present data. Ridzi (2020) engages in a discussion of the nested scales at which data can be used and the impact data have at each of those scales while Wascalus and Wolter (2020) examine the challenges of adapting data for small area geographies to rural areas. Indices take a large amount of data to help uncover and reveal complex, “wicked” problems at the chosen scale. In Chaps. 4 and 5, Connell et al. (2020) tackles racial inequities by turning Dr. Martin Luther King Jr. ‘s call for racial equality and economic justice into an indicator framework to assess America’s realization of prosperity and equality. After examining the appropriate scales at which interventions can move the needle, Crawford and Ridzi (2020) describe how a poverty index came to be to address regional disparities and what infrastructure is needed for monitoring efforts to alleviate one of the worst rates of poverty in the US. Similarly in Chap. 6, Summers et al. (2020) presents the Decision Integration for Strong Communities (DISC) application, a “dashboard” of community characteristics to help communities assess how resilient they are and find information to encourage smart growth planning. In case studies from Australia, Canada and the United States (Chaps. 7 and 8), we are treated to the rationale for, as well as the development and implementation of, diverse community indicators projects. Davern et al. (2020) offer an example of indicator application in community and public health planning within a local government in Cardinia, a suburb of Melbourne. Powell (2019) contrasted the efforts of three Northern communities in the US and Canada and the conditions for success to establish indicators to track sustainability planning. Brutschy et al. (2020) in Chap. 9 explores the conditions and attributes needed to successfully collect and leverage community data for positive impact, using a couple of data-supported initiatives as examples while Connell et al. (2020) in Chap. 10 focuses on data as a tool to engage with community members and data practitioners and expand the use of existing data sets. In the final Chap. 11, from outside the field of community indicators, Abraham (2019) makes the argument that community indicators will improve the planning field’s need for stronger reliance on both evidence and community participation. Individually, these chapters support many of the practices outlined in this introductory chapter. Taken together they create a tapestry of practices, tools or exemplars, that represent the relevance and strength of today’s community indicators universe.

12

C. Stevens and L. D. Wray

References Abraham. (2020). Data-Driven Decision Making and Community Indicators: Towards an integration of DDDM in Community Development. In F. Ridzi, C. Stevens, & M. Davern (Eds.), Community quality-of-life indicators: Best cases VIII (p. 199–210). New York: Springer. Barrington-Leigh, C., & Escande, A. (2018). Measuring progress and well-being: A comparative review of indicators. Social Indicators Research, 135(3), 893–925. BNIA. (n.d.). Baltimore data day. Accessed July 25, 2019. https://bniajfi.org/data_day/. Brutschy, S., Luedtke, M., & Frost, K. (2020). Leveraging data for meaningful improvements: How credible data enables partnership alignment to achieve well-being at the population level. In F. Ridzi, C. Stevens, & M. Davern (Eds.), Community quality-of-life indicators: Best cases VIII (p. 163–184). New York: Springer. Cohn, S. (2015). Better know a CI Project Webinar: Jacksonville’s community indicators. February 27. Accessed February 03, 2018. http://communityindicators.net/knowledge/better-know-acommunityindicators-project-webinar-recordings/. Connell, E., Bell, S., Martin Rogers, N., & Souvenir, N. (2020). Data parties: Giving the community tools to use east metro pulse survey data. In F. Ridzi, C. Stevens, & M. Davern (Eds.), Community quality-of-life indicators: Best cases VIII (p. 185–198). New York: Springer. Crawford, J. & Ridzi, F. (2020). Meaningful, manageable, and moveable: Lessons learned from building a local poverty index. In F. Ridzi, C. Stevens, & M. Davern (Eds.), Community qualityof-life indicators: Best cases VIII (p. 65–88). New York: Springer. Davern, M., Dodds Buckley, P., & Bucello, P. (2020). Indicators supporting public health, partnership, liveability and integrated planning practice: The case study of the Cardinia Shire growth area in Melbourne, Australia. In: F. Ridzi, C. Stevens, & M. Davern (Eds.), Community quality-of-life indicators: Best cases VIII (p. 115-136). Springer International. Epstein, P. D., Coates, P. M., Wray, L. D., & Swain, D. (2016). Results that matter: Improving communities by engaging citizens, measuring performance, and getting things done (p. 105). New York: Wiley. Fox, M. (2016). Better know a community indicators project Webinar: San Antonio SA2020. Community Indicators Consortium. June 04. Accessed February 03, 2018. http://communityindicators. net/knowledge/better-know-a-community-indicatorsproject-webinar-recordings/64. Gardere, L., interview by Stevens, C. (2018). Follow up questions to June 15 New Orleans Prosperity Index Webinar (July 05). Harris, J. (2018). Capital region collaborative’s community indicators project better know a community indicators project Webinar. Community Indicators Consortium. March 30. Accessed 07 02, 2019. https://communityindicators.net/knowledge/better-know-a-community-indicators-projectwebinar-recordings/#8802a91c906999ec0. Hruby, K., interview by de Blois, M. (2018). CIC community indicators curriculum interview— Truckee Meadows tomorrow (February 15). Iyer, S., interview by de Blois, M. (2018). CIC community indicators curriculum interview—Baltimore Neighborhood Indicators Alliance (BNIA) (January 21). Johnson, A., interview by Stevens, C. (2018). CIC community indicators curriculum interview—ACT Rochester (January 20). Joo, S., interview by de Blois, M. (2018). CIC community indicators curriculum interview—Magnolia place community (LA) (February 09). Liuzzi, A., interview by Stevens, C. (2018). CIC community indicators curriculum interview— Minnesota Compass (January 24). O’Connell, K., Young, A., & Botchwey, N. D. (2020). Measuring the dream for an equitable and sustainable future. In F. Ridzi, C. Stevens, & M. Davern (Eds.), Community quality-of-life indicators: Best cases VIII (p. 47–64). New York: Springer. Pettit, K., interview by de Blois, M. (2018). CIC community indicators curriculum interview— National Neighborhood Indicators Partnership (NNIP) (February 03).

1 Creating and Sustaining Community Indicators …

13

Powell, J. (2020). A tale of three cities: Five conditions conducive to sustainability plans and measurements. In F. Ridzi, C. Stevens, & M. Davern (Eds.), Community quality-of-life indicators: Best cases VIII (p. 137–162). New York: Springer. Ridzi, F., interview by de Blois, M. (2018). CIC community indicators curriculum interview— Central New York Vitals (February 02). Ridzi, F. (2020). Goldilocks data-connecting community indicators to program evaluation and everything in between. In F. Ridzi, C. Stevens, & M. Davern (Eds.), Community quality-of-life indicators: Best cases VIII (p. 15–35). New York: Springer. Summers, K., Salazar, V., Olszyk, D., Harwell, L., & Brookes, A. (2020). The development of DISC (decision integration for strong communities): An agile software application of sustainability indicators for small and rural communities. In F. Ridzi, C. Stevens, & M. Davern (Eds.), Community quality-of-life indicators: Best cases VIII (p. 89–114). New York: Springer. Wascalus, J., & Wolter, E. (2020). Strategies for expanding indicator profiles to small rural geographies. In F. Ridzi, C. Stevens, & M. Davern (Eds.), Community quality-of-life indicators: Best cases VIII (p. 37–46). New York: Springer.

Chantal Stevens is the executive director of the Community Indicators Consortium, an open learning network and global community of practice for the field of community indicators. Her interests and expertise in sustainability, public engagement, and organizational development were honed over a 30-year career as the executive director of Sustainable Seattle, a pioneer in the development of community-grounded indicators, and People for Salmon, a state-wide public engagement initiative, as performance management analyst and legislative auditor with King County and as environmental manager with the Muckleshoot Indian Tribe. As a speaker, trainer, and consultant, Ms. Stevens has worked with organizations throughout the world on using measures to improve sustainability and wellbeing. She also held several terms on the Issaquah Planning Commission and has sat on various other commissions and boards, including as chair of the board of the largest food cooperative in the U.S. Chantal holds a master’s degree in Marine Affairs from the University of Washington. Lyle D. Wray PhD, has served as executive director of the Capitol Region Council of Governments www.crcog.org since 2004. In this role, Dr. Wray serves as chief executive for Connecticut’s largest regional planning organization serving metropolitan Hartford, Connecticut. The region includes 38 cities and towns with a population of about one million. He served as County Administrator for Dakota County, Minnesota; as Executive Director of the nonpartisan Citizens League in Minnesota; and director of Ventura County Civic Alliance that issues a regular state of the county quality of life report. He has been active in performance measurement in a number of ways including advising the Minnesota Governor on performance measurement and graduate teaching of outcomes and performance measurement. With Paul Epstein he co-authored the book “Results That Matter” on improving communities through citizen engagement and performance measurement. He is a past president of the Community Indicator Consortium. He was elected as a member

14

C. Stevens and L. D. Wray of the National Academy of Public Administration. Dr. Wray holds a Bachelor of Arts, Master of Arts and PhD in psychology from the University of Manitoba, Winnipeg, Canada. He completed the State and Local Government executive program at the Kennedy School of Government at Harvard University.

Chapter 2

Goldilocks Data-Connecting Community Indicators to Program Evaluation and Everything in Between Frank Ridzi

Abstract Community indicators have long sought to measure and inspire community level change. On a much smaller scale, program evaluation and performance management seek to measure and inspire change among program participants. While communities across the United States may have robust community indicators and performance management cultures these two efforts are often disconnected, leaving a large amount of guesswork between identifying major community needs and coordinating the many nonprofit and other community partners needed to bring about positive change. In this paper, we utilize the metaphor of GPS map zoom levels to articulate the key types of data needed to build a comprehensive data ecosystem that integrates community indicators with program level performance monitoring. We then use the case study of Syracuse, New York to elucidate how a nested logic model approach can be used to coordinate efforts that approximate the Federal Reserve Bank of San Francisco’s (Erickson 2017) vision of a “complex adaptive system” that will help organizations and sectors coordinate their work across silos to achieve shared outcomes (such as helping children be ready for kindergarten, youth graduate from high school, homeless people find homes or retrained workers hold a steady job) (Erickson 2017, p. 43).

2.1 Introduction Community indicators (CI) and quality of life indices are a growing movement that seeks to use data to bring about measurable community improvement (Holden et al. 2017; Stevens et. al. 2019, p. 1). At the other end of the spectrum program evaluation and performance management (PM) are growing in popularity, especially as funders seek evidence of impact (Ridzi 2013; Stevens et. al. 2019, p. 6). Indeed this has been a focus of the Community Indicators Consortium, which has worked to F. Ridzi (B) The Central New York Community Foundation and Le Moyne College Department of Anthropology, Criminology and Sociology, 1419 Salt Springs Rd, Syracuse, NY 13214, USA e-mail: [email protected] © Springer Nature Switzerland AG 2020 F. Ridzi et al. (eds.), Community Quality-of-Life Indicators, Community Quality-of-Life and Well-Being, https://doi.org/10.1007/978-3-030-48182-7_2

15

16

F. Ridzi

create a maturity model of CI-PM Integrations along a spectrum (Community Indicators Consortium 2010). The rise of collective impact coalitions creates a market for combining the community impact aspirations of community indicator projects with the accountability of program level outcome evaluations (Ridzi and Doughty 2017; Ridzi 2019). This is perhaps best seen in the rising popularity of the Results Based Accountability approach which seeks to coordinate program level evaluation with improvements in population-level community indicators (Friedman 2005). Though many communities have embraced a data-driven approach, the actual collection, analysis and dissemination of data in ways that drive community action has been elusive (Ridzi and Doughty 2017). The major problem that arises is that there is a large “black box” phenomenon in the space between community indicators and program evaluation. In essence, there are blind spots for community leaders who seek to make every decision a data-driven decision because they arrive at key decision points for which no data exist. As such, community indicators help to identify major trends that are problematic but, once identified, the data infrastructure that exists in many communities fails to offer concrete guidance beyond guesswork about when, how, where, and with whom to collaborate in order to offer programming or other interventions. This is similar to where Davern et al. (2017) describe indicators as being like the “tip of the iceberg” and much more detailed data are often needed to get a clear picture of the full profile of the iceberg including the parts that are hidden below the water line. As they state, “indicators provide a tip of the iceberg representation of issues that enable conversations to be started and further investigation into explanations and influencing factors. However, indicators don’t provide suggestions on interventions that could influence or improve results in the future…” (Davern et al. 2017, p. 579). Those who take action may be able to assess the success of individual programs but are left in the dark in the near term about whether the coordinated collaborative components of their interventions are having a positive impact. Indeed, organizations are often working in parallel on the same initiatives and never connect. In this paper we explore how indicator projects have a role to play in bringing these groups together. In short, they lack the leading indicators that would forecast anticipated changes in the lagging indicators that tend to exist in community indicators projects. As a result, coalitions and collective impact efforts alike can seem to be stuck in a rut, repeating the same conversations and “spinning their wheels” without clear direction as to how to move forward. To address this dilemma, in 2017, Erickson (2017) of the Federal Reserve Bank of San Francisco (FRBSF) put forth the vision of creating a “complex adaptive system” that would help organizations and sectors coordinate their work across silos to achieve shared outcomes (such as helping children be ready for kindergarten, youth graduate from high school, homeless people find homes or retrained workers hold a steady job) (p. 43). The FRBSF asserts, “We spend a lot of money in the United States (as do governments and charities around the world) to solve social problems, but we don’t know exactly which half (or more) is working (p. 41).” What is needed is better coordination and better real-time data to track and improve this coordination.” Specifically, we “need to develop new information technology tools to make cross

2 Goldilocks Data-Connecting Community Indicators to Program …

17

sector and place-based interventions work more effectively (p. 51).” The FRBSF further asserts that “all communities need both place- and people-oriented strategies to improve health, increase outcomes, and reduce the incidence of poverty (p. 37).” They advocate for an “ecosystem—or market—that creates the right incentive to collect, analyze, and be guided by data. This market also incentivizes better cooperation among multiple players from multiple sectors.” This paper takes this vision one step further by creating a prototype information technology infrastructure to explore both who potential users for such a system are (such as philanthropies seeking to create Collective Impact through their grantmaking) and how they might use it. In this paper we report on lessons learned from the creation of a prototype data ecosystem that meets the key criteria established by the FRBSF. Key lessons learned include the need for data analysis among four distinct audiences or levels of data usage: funders, coalitions, program designers, and program implementers. Each level of usage requires data sourced from different locations and, as seen in the present case, can benefit greatly from a shared data infrastructure. In this example, US census tract data act as a lowest common denominator that: (i) allows effective data sharing that is capable of driving action; and (ii) maintains the confidentiality of all clients being served.

2.2 Methods For this project we created a prototype data ecosystem that meets the criteria laid out by the FRBSF. Though this project was under construction well before the FRBSF published its vision for a complex adaptive system, shared sensibilities and insights have yielded a product that nevertheless establishes proof of concept that data structures such as envisioned by the FRBSF are not only possible but also doable through the use of open source and desktop computing tools that are relatively inexpensive and widely accessible. In this case we relied on Microsoft Access, Microsoft Excel and the open source R-studio platform to build the 1.0 version and we used the Google suite of products to build a 2.0 version. The objective of the project was to minimize the number of decision points for local collective impact efforts that could not be data informed due to lack of data. The result is a single data ecosystem that incorporates multiple databases in a coordinated fashion in order to meet the needs of funders, coalitions, program designers, and program implementers. Although it combines data from a variety of sources, each are connected through a unifying backbone of a shared unit of analysis. The FRBSF rightly pointed out what many others have asserted,“ZIP Code is more important than your genetic code for influencing your health over a lifetime (Erickson 2017, p. 45).” However, in this case, using ZIP Code (created by the United States Postal Service) as the central unit of database infrastructure is far too blunt and broad. Rather, what was needed was a much more precise geographic locator that participating organizations could convert their data to for free or low-cost. This unit of analysis is the United States Census tract and

18

F. Ridzi

each data source, by being converted to this level of neighborhood aggregation, both allows data sharing and ensures the confidentiality of people whose data are being shared.

2.3 Findings: Understanding the Components of the Data Ecosystem In this section we first explore the layers or components of data that were needed to create a data ecosystem that connected community indicators with program evaluation. Ecosystem is an apropos term since it connotes a complex network of interdependent parts that relate to each other in part through the space or layers that they inhabit. We articulate the layers by utilizing the metaphor of GPS map zoom levels to identify the key types of data needed to build a comprehensive data ecosystem that integrates community indicators with program level performance monitoring. Much like Bronfenbrenner’s ecological systems model (Singh et al. 2015) and studies of data collection patterns (Ridzi 2013), these indicators formed what can be thought of as concentric layers that situate more local data at the center and more general or distant data toward the outskirts (see Fig. 2.1).1 Consistent with this approach, the Results Based Accountability framework (Friedman 2005) elucidates how multiple community programs that do impactful work with their clients (i.e. performance accountability as measured by program evaluation) will eventually add up to population level change if they are concentrated in the same geography (i.e. population level accountability as measured by community indicators). In other words, if a group of programs working in the same neighborhood have successful results, the community indicators for that neighborhood should improve. This is the idea behind many collective impact efforts as well and requires coordination across levels. The Community Indicators Consortium’s (Stevens et. al. 2019, p. 4) Community Indicators Project Development Guide describes community indicators as “serv[ing] as a map to guide priority- and agenda-setting for the work of groups involved in improving community-level conditions…” Given that a commonly employed metaphor for using data is that of driving (as in driving with data, data driven, dashboard, roadmap) (Stevens et. al. 2019, p. 76), it seemed apropos to continue with this imagery by using the metaphor of a road map to “map out” the types of data needed during various legs of a community’s journey when moving from data to action. 1 Urie Bronfenbrenner, created the Ecological Systems Theory which showed how the various layers

of society interacted with each other much like an ecological system in order to affect the lives of individuals. This included such layers (arranged as concentric circles much like the diagram we use) as the person’s immediate environment, larger institutional actors within the community such as schools and local government, and broader national forces such as national government and national norms (Singh et al. 2015).

2 Goldilocks Data-Connecting Community Indicators to Program …

19

Fig. 2.1 The levels of a data ecosystem

In this metaphor, shifting from one level or data scale to another is akin to shifting from one map to another, more zoomed in version of that map once we get closer to our destination. Technology has revolutionized this experience in recent years with the advent of mapping apps on smartphones. Today, we no longer need to acquire a series of ever more zoomed in maps to plot out one’s trip. Instead, the apps do it for us. Nevertheless, the metaphor still holds. When we first enter our destination into the GPS unit or mapping app, the initial map that appears is zoomed out quite far and provides the general sense of direction for the trip (Highway Level). As one embarks on the trip and gets closer, there is a need to zoom in to see the exact exit from which to depart the highway and the main road to turn on to (Main Street Level). Once on the main street, it becomes imperative for the mapping app to zoom in even further to see the next turn (Side Street/or Back Road Level). Finally, as you get even closer you need to get to the actual address you are seeking (Address Level). Though shifting between these scales seems effortless today, this belies the complex data infrastructure that allows your app to zoom in and replace the previous screen image with a much more detailed one. It is only through the seamless transition between

20

F. Ridzi

these scales that we are able to continuously drive from one location to another and minimize “driving blind” (as may occur in locations where satellite connection is lost and the map on our GPS app disappears or freezes). In this paper we share how driving with data is similar to our experience of driving with GPS and requires an infrastructure similar to these multiple map layers (Highway, Main Street, Side Street/Back Road and Address) to be done well. In the section below, we describe each of these types of data, when they are useful, where to find them, and how they connect to the next layer down as we zoom in. We then explore the value of coordinating across these data types when seeking to bring about community change driven by community indicators.

2.3.1 Highway Data Highway level data are the most recognizable to those familiar with community indicators projects. They are perhaps the most commonly occurring forms of data that we see on community indicator websites and community indicator reports. They are typically derived largely from federal government sources (much like highways) such as the U.S. Census Bureau, its American Community Survey and The United States Bureau of Labor Statistics. These data can exist on multiple different scales (such as state, county, city or census tract) but as you zoom into the smaller scales the reliability of these estimates gets much less precise (much like taking a screenshot of a highway map and trying to zoom in to a specific exit—while you can get the general idea of what direction to go, you do not get such details as main street and side street names and positions). Highway level data are most useful for identifying community problems that merit further follow up. They are tremendously useful in this regard because communities do not need to spend any money or extra resources to collect these data. Rather, they simply exist on federal websites and are open to the public. Historically, these data have been the only option for communities as they seek to launch and measure the effects of collective impact efforts or community coalitions (Ridzi 2017; Hatry and Morley 2008). If you think in terms of creating a logic model for a community indicators project that is seeking to move the needle, Highway data are most fit for identifying the problem or issue that needs to be solved and then, after many years, looking to see if that issue has in fact been addressed (see Table 2.1). While these data are good for overall estimates of community well-being, they tend to lack specificity and are more ballpark estimates than measures of point-in-time community status. Oftentimes these data are based on the pooling of one-, three- or five-year estimates, which means that the community is surveyed over multiple years and those data are combined together in order to get a sample large enough to measure change. The downfall to this, of course, is that we cannot compare progress year-to-year but rather over a large series of years, and even then there is a time lag (Stevens et. al. 2019, p. 5).

2 Goldilocks Data-Connecting Community Indicators to Program …

21

Table 2.1 Data level roles, uses and sources Data Level

Role in ecosystem

Optimal logic model use

Optimal user

How to get it

Cost

Highway

Identify major community social problems

Social problem and impact

Funders, policy makers and planners within different tiers of government (e.g. municipal, state, federal)

U.S. Census Data USA Opportunity 360

Typically free unless you want to customize display

Main street

Identify moveable needles

Long-term outcome

Coalitions

Local Government request

Typically free but requires social trust

Side street

Use data to coordinate partners

Intermediate (middle) outcome

Grantwriter/program designer

Coordinated data collection across organizations

Can range from free to millions annually

Address

Use data to serve residents

Short-term outcome

Frontline implementation staff

Organizational data collection in programs

Can range from free to thousands annually

Furthermore, the closer we zoom into a specific neighborhood, such as a census tract, the less reliable these estimates become. For communities that are seeking to change the problems they notice in highway level data, there is often a sense of frustration at the inability of these approximations to reliably or precisely gauge the progress that such communities make as they seek to work together for long-term improvement (Ridzi 2013). The limitations of Highway data can lead to disillusionment when collective impact participants turn to data in order to assess whether they are making a difference and to make midcourse correction to the deployment of their resources (Ridzi 2017). In order to make the shift toward a level of data that does meet this need, communities find that they must build an infrastructure that offers them Main Street data.

2.3.2 Main Street Data Main Street data, consistent with the metaphor, differ from Highway data in that they are not made available or maintained by the federal government but rather by local governments. We are often familiar with these types of data because they appear in

22

F. Ridzi

such things as local county or city government reports, police department reports, school district report cards or public health reports issued by local health departments. While these data can be helpful in community planning, our experience for this project revealed that they seldom exist in a way that is compatible with Highway data or the next level of data down, Side Street or Back Road data. This lack of compatibility is akin to when the railroad industry realized they were using different gauges for their railroad tracks and had to figure out a common gauge to use. This lack of common standards is recognizable, for instance, when looking at a U.S. Census estimate of poverty by census tract and then looking to see how children perform in school in a high poverty census tract only to find that the school district data are not publicly available on a census tract basis. Rather, these data are only reported on a school by school or district by district (rather than census tract by census tract) basis. As a result, the data comparison is flawed in that you are using two different geographies. To solve this problem, and create a community infrastructure for data sharing across scales, the Syracuse case, which will be introduced in the next section, demonstrates that one solution is simply to ask local jurisdictions to recalculate the data that they currently report and share those data aggregated by census tract. This allows for a smooth transition from publicly available Highway data to locally produced Main Street data. This, however, is more easily said than done. Where highway data are available to anyone with an internet connection, re-calibrating local Main Street data to the census tract level takes actual sweat equity. The federal government is unlikely to change and so it is up to communities to revamp their data by converting it to the same gauge as the federal government, census tracts. In some cases, this may require payment, such as some state agencies that will recalculate local unemployment or other data for a fee. In other cases, this requires a certain amount of positive peer pressure or advocacy. Since a single organization is unlikely to convince a school district or police department to change their reporting structure, what we have seen is that coalitions of multiple organizations that are all pursuing shared goals (such as improved third grade reading, reduced lead poisoning among children, or decreased unemployment rates) will have the needed clout to enter into memoranda of understanding with local jurisdictions to request that data are processed in a format that is publicly shareable and compatible with census tract analysis. Building the social capital required to convert local data in this way can be time and resource intensive but these efforts pay off when communities are now able to track their progress using measures that both move in response to their efforts and fit the scales on which their efforts are made. For instance, a coalition seeking to shift the needle on the number of children that are succeeding in school may have a difficult time focusing on an entire city but will have much greater success focusing on a specific neighborhood or census tract. In this case, with Main Street data calculated, the community can track the progress of children in that neighborhood and assess whether their efforts are making a difference on a year-to-year basis. Similarly, a group seeking to reduce lead poisoning among children would be able to identify a target neighborhood or neighborhoods with a higher than typical percentage and

2 Goldilocks Data-Connecting Community Indicators to Program …

23

engage in such things as window and door replacements and public outreach. With access to Main Street data they could then see if their efforts are working before scaling them up to other neighborhoods in the city. For example, without Main Street data, such groups would be forced to look at countywide estimates for lead poisoning which would be extremely unlikely to move given the coalition’s focus on a single census tract due to limited resources. When we think of logic models, Main Street data are ideal for long-term outcomes in that they take time to see change but they are a leading indicator that predict eventual change in the Highway data that we can use to measure eventual impact (see Table 2.1). While this ability to measure local community change has a certain advantage over Highway data, Main Street data fall short of providing clear data-driven directives to organizations in a community that is looking to collaborate to move the needle on Main Street measures. Consistent with the mapping metaphor, ending construction of a community data ecosystem with the Main Street level data would be similar to exiting a highway and onto the main street but then having your GPS unit lose its satellite link before it tells you what side street to take to reach your destination (in this metaphor key decision points for collective impact efforts can be thought of as key roadway turns in a driving journey). In order to address this data gap and avoid a potential blind spot, communities must rely on Side Street, or in more rural areas Back Road, level data.

2.3.3 Side Street (Back Road) Data Among the four levels or types of data discussed here, Side Street or Back Road data are the most rare. These are data that allow community organizations to coordinate their activities in order to move the needle on their Main Street data goals. Side Street data are often actually just an aggregation of the data that individual community organizations collect to measure their program outcomes or outputs (Address data). As we will see, Address data are much more common in communities that have a robust performance management or program evaluation culture. These evaluation activities involve organizations collecting data on their clients in the form of outputs and outcomes and measuring how their lives have improved as a result of their individual programs activities. However, this type of evaluation approach is often critiqued for being non-collaborative resulting in organizations working on the same community problems (and even initiatives) in parallel but not actually coordinating this work in a data driven way. Side Street data, because it involves sharing outputs and outcomes across organizations, is designed to address this problem. If done well (i.e. collecting data on outcomes that reach across sector silos—such as having literacy organizations collect data about lead exposure and health organizations collect data

24

F. Ridzi

about literacy), Side Street data can also address critiques of program evaluation that assert it lacks a holistic approach that acknowledges that real life issues will take much more assistance to overcome than a single program or sector can deliver. The FRBSF points to the work of S. Leonard Syme who urged us “to think of the needs of the whole person rather than siloed focuses on medical care, or housing, or jobs, or education (Erickson 2017, p. 47).” Side Street data offer a critical functionality. These data involve multiple nonprofit organizations and their partners collecting the same data points from their clients and then pooling those data together to look for opportunities for better and more coordinated deployment of resources. This can take such forms as: co-locating services; identifying rising pools of client needs; bundling services to address needs of clients across multiple organizations (and sectors) in a shared geography; or assessing the impact of collaborative efforts on a targeted neighborhood over a short period of time. In a basic form, this process can be conceived of as nonprofit match-making or dating. Looking simultaneously at Side Street data for two organizations on a given community need (such as lead exposure) can help them to clarify the most obvious opportunities for coordination. Whereas Main Street data tells us if we are succeeding in a summative way, Side Street data drives decision-making in a formative way that helps us improve our programming and coordination as we go. Using the map metaphor, Side Street data help us to do what we do more efficiently, much like side streets help us avoid congested avenues while driving and help us reach our destination faster. In fact, the construction of the Side Street portion of the data platform we built for this project utilized a concept similar to Google Traffic maps. Google Traffic uses the GPSdetermined locations transmitted by cell phone users as they drive to identify areas of traffic congestion without revealing any personal information about the drivers (since they do it in aggregate). We did much the same thing by aggregating reported outcomes in real time by census tract. The end result is quite similar; Google Traffic uses red on a map to highlight roadway segments that are congested (as reported by many drivers) and we use red on a map to highlight census tracts with poor outcomes (as reported by many clients of various organizations across different sectors such as health and human services). In a logic model, Side Street data are ideal for medium term outcomes such as looking to see if lead testing of homes has increased in neighborhoods targeted by several partner organizations. In such cases it is typically the responsibility of the collective impact coalition’s leadership to coordinate both the data aggregation and the organizational match making. This is consistent with collective impact focus on shared measurement since we can anticipate that, if these medium-term outcomes see improvement, we will eventually see an improvement in the long-term outcome of children poisoned by lead as well. In addition to measuring outcomes however, the coordinative potential of Side Street data helps local organizations, or at least their program developers to make decisions about where to offer programs, what partner agencies to recruit participants from, and whether their initial efforts are making a difference within a matter of several months.

2 Goldilocks Data-Connecting Community Indicators to Program …

25

2.3.4 Address Data Address data are essentially the same data as Side Street data but without the sharing component across organizations. For instance, routine Address level data collected by an organization might ask clients as they come in the door whether they are in need of childcare, transportation, or food. By simply asking the same question again, at a later date, organizations are able to measure their impact on these individual clients (e.g. we can measure if the need for transportation declined after receiving our services). This is standard program evaluation practice and in a logic model is useful as a short term outcome measure. These data also offer a way to continuously tweak a specific program that an organization offers to better meet the targeted needs. What sets these data apart from Side Street data is that the names, addresses and identifying contact information are collected with these data since an organization typically knows the clients that it serves and needs to be able to track their individual progress as well as follow up with them in the future to see how they are faring. In the case of Syracuse, it was possible to use the same data infrastructure to both collect Address level data and create a Side Street data system. This was completed with multiple organizations using the same survey mechanism and then having them geocode their data to census tracts. Any identifying characteristics were removed before sharing the dataset. The result was an anonymised dataset that included the census tract, the organization that surveyed the client and the client’s responses. For example, this is the equivalent of saying, “A client working with my organization is in need of childcare and transportation and they live in census tract 17.” This process created a two-tiered data set that included Address level data that organizations could use to evaluate their programs as well as a de-identified Side Street level dataset that spans across multiple organizations and could be used to coordinate their services. Though initial thinking led people to believe that they would need to share the confidential name and contact information typical to Address level data, in actuality, sharing only de-identified data across organizations proved sufficient to guide action. It is certainly possible to also take an approach in which the personal identities and contact information present in address level data are shared across organizations, but this requires a legal memorandum of understanding and consent given from clients. This can be both costly and time-consuming and is avoided by sharing de-identified data across organizations. While Address level data may be collected apart from a broader infrastructure that coordinates Side Street data, there are many reasons why a coordinated effort is preferred. In a coordinated effort multiple organizations collect the same data points and share these data in de-identified format. Partners involved are able to notice trends across organizations such as the prevalence of people who need childcare in a specific neighborhood (and who are served by specific organizations). At this point the multiple organizations that are sharing data can reach out to one another and identify this need that has appeared in the Side Street level data. However, it requires each organization to go ahead and find the contact information for the

26

F. Ridzi

relevant clients in their Address level data and send them an email, give them a phone call, or personally invite them to a focus group that will be used to design local programming. In cases such as this scenario, it is easy to see the advantage of designing Address level data systems in coordination with or as part of a larger Side Street level data infrastructure so that these two levels can easily communicate with each other. Ideally, the transition will happen with the same ease as a GPS zooming in from Side Streets to point out the actual address at which you need to stop to reach your destination. As noted above, it should not be presumed that this technical solution will solve all problems. There are organizational issues/challenges that encourage or discourage agencies to work together and share data. Most notably this approach will only work in instances where a substantial investment has been made in cultivating a culture of collaboration (rather than competition among partners). We have seen this in instances such as the U.S. Department of Housing and Urban Development’s (HUD) anti-homeless grants where community partners have learned to work together to share the data required for successful multi-partner grants. Pre-existing collective impact community coalitions (Ridzi and Doughty 2017) are also fertile ground as we find in Syracuse. Once Address and Side Street layers of data are coordinated you can further see how each layer of data ripples down to the next or up to the one above it in order to coordinate activities and services (see Table 2.1). They are aligned as with trains that run on a common gauge track.

2.4 Connecting Layers Through Nested Logic Models: The Case of Syracuse, NY Above we have focused on the different layers or types of data needed to produce a community data ecosystem that allows communities to link their community indicator efforts to local program evaluation efforts. With this new infrastructure, communities are well poised to move toward the vision of a better “hand off” articulated by Terri Bailey at the 2004 Community Indicators Consortium Conference: I think of this work as a race, a race for our communities, our nation, and more. But we in the community indicators field treat it more like a relay race. We see our job as running the first leg of the race in which we collect, produce and disseminate data and then we hand off to those who use the data to run the last leg of the race. We tell ourselves it is their job to use the data to promote equity and justice, to affect meaningful change in communities. We have convinced ourselves that our job is done once the hand off is complete. But the hand off is NOT the finish line. Anyone who has ever run a relay race will tell you that the first runner is as responsible for what happens at the finish line as the last runner. (Stevens et. al. 2019, p. 74)

2 Goldilocks Data-Connecting Community Indicators to Program …

27

2.4.1 A Local Case Study In this section we turn to a case study of Syracuse, New York to illustrate how this data infrastructure can be applied, in “relay race” fashion, to address the community indicators goal of reduced lead poisoning. Key to the FRBSF’s concept of a complex adaptive system is the role of community entrepreneurs (Erickson 2017, p. 43) who rely on data analysis to achieve community outcomes. While the FRBSF discusses various roles in terms of general contractors and subcontractors (p. 43), here we break up the key roles along the lines of funders, coalitions, program designers, and implementers. As a theoretical tool, we introduce the nested logic model because it allows us to combine the concept of different layers of data with the different actors involved in taking community action. By paying close attention to each actor’s target outcomes (and the data layer they use to measure them) we can see how these tiers of data help to coordinate actors that naturally inhabit different tiers of action within the ecosystem. For the FRBSF, the starting place for constructing a complex adaptive system that can begin to address society’s intractable social problems starts with an end goal: “we could build a complex adaptive system by inventing an end goal—a child ready to learn at kindergarten, a student graduating from high school, a formerly homeless person who is stably housed, a retrained worker who holds a steady job, etc.—that has the potential to help align the many organizations to work together more effectively across sectors and silos to achieve these outcomes without a central planner (Erickson 2017, p. 43).” In the case of Syracuse, the overarching “end goal” arose when local funders realized that lead poisoning was high among the children in their county. The Central New York Community Foundation (the Foundation) had launched a community indicators website entitled CNYvitals.org. Using publicly available data from the state they noticed that 14.7% of children in Onondaga County experienced lead poisoning (defined as an elevated blood lead level of over five mcg/dl). This was three times the rate of children in neighboring counties (which had percentages such as 4.8 and 4.5%). This information at the Highway level of data was enough for the Foundation to commit to looking further into this problem and to consider releasing a request for proposals (RFP) to organizations that would work to reduce lead exposure among children. However, given the county’s large size, the foundation realized it needed to target areas of highest need if it wanted to have an impact that was measurable. To do this, more detailed data (Main Street level) would be needed. The Foundation reached out to a coalition of housing and public health organizations that were involved in the Green and Healthy Homes movement and that the Foundation (as a member itself) had contributed to supporting in the past. This multi-sector coalition, which included nonprofits and government from the housing, health, education and human services sectors, had the social capital with the local health department (in fact it included members of the local health department) that was needed to request Main Street level data. These data revealed the annual percent of poisoned children by census

28

F. Ridzi

tract. With Main Street data, we could see that the county level of lead poisoning had been dropping in recent years to a level of close to 5%, which was more aligned with neighboring counties. However, when drilling down to specific census tracts it was noted that the 5% county poisoning rate doubled to approximately 11% when looking at the city and then in specific census tracts it doubled again to nearly 25%. This allowed the Foundation and the coalition to put together a series of partner organizations that could target those high need neighborhoods. The Foundation, with the help of the coalition, identified key nonprofit organizations from sectors such as housing and human services that could now go ahead and target these high need areas. But there continued to be a need to monitor leading indicators that would forecast an eventual decline in lead poisoning (i.e. Side Street level data was needed). As a result, the Foundation and its partners engaged in a shared Life Needs Assessment questionnaire that was administered across a network of nonprofit organizations. This questionnaire includes 19 common questions and asks clients among participating nonprofit organizations (from multiple sectors) about the needs they are experiencing in areas ranging from childcare, transportation, healthcare and housing to safety, education, legal services and addiction. The construction of this survey is by design cross-sector since the purpose is less for organizations to measure things related to the services they provide and more for organizations to learn about the hidden needs of their clients, of which they may not be aware because they are not central to the organization’s services. For instance, while a job training organization may be interested in its clients’ employment status, learning that a large group of its clients in a specific neighborhood also need childcare, transportation, healthcare, clothes and housing empowers that organization to reach out to other providers who can help its clients by providing the services it does not. In this fashion, the data sharing can serve as a cross sector matchmaking tool for organizations that are looking to respond to their clients’ holistic (i.e. cross sector) needs. To administer the survey the Foundation developed an online, real-time web application that made it easy for organizations to administer the survey (i.e., as easy as clicking on a web link unique to their organization). The application then geocoded the results to census tracts before removing all other identifying characteristics and sharing the results among the partner organizations. Using this application, the community emerged with a valuable infrastructure of Side Street data. Now, participating nonprofits that had partnered to create these data could use the data of all of their partner organizations to track where there were clients who had not had their homes or children tested for lead. This same needs assessment survey provided the Foundation with data to identify potential partner organizations that were serving clients in the high need areas so that these organizations could be reached out to and recruited as trusted messengers for lead safe messaging and recruitment to the programs that were being funded in their neighborhoods. Finally, when it came to the case managers of programs, they now had, at their fingertips, the appropriate screening tool to refer their clients in need of lead testing (i.e. Address level data) to testing facilities and, where needed, to lead abatement programming funded by the Foundation. Utilizing

2 Goldilocks Data-Connecting Community Indicators to Program …

29

this Address level data was now as easy as filtering one’s caseload for people who had not been tested and then copying their emails and pasting them into a blind carbon copy, or BCC (outreach could be done in a matter of 4 minutes). As an added benefit, this process also offered the promise of helping collective impact groups strive toward greater cross-sector collaboration. For instance, the multi-sector Literacy Coalition began to look at the results of their client surveys as a way to discern what types of non-literacy focused entities (such as health organizations) to recruit to their governing board. In the past hunches had encouraged them to look at factors such as housing as part of their holistic mission; now they had data to show that their clientele were connected by shared needs.

2.4.2 Nested Logic Models In this way Highway, Main Street, Side Street and Address level data could be seen to nest within each other and direct action in coordinated fashion such that progress in a lower level (i.e. closer to Address level data) would serve as a leading indicator for progress in an upper level. Through such a nested logic model (see Fig. 2.2), we see that Address level data can be used to reduce the number of clients that are untested for lead. Side Street data could be used to coordinate among nonprofits to increase

Fig. 2.2 A nested logic model

30

F. Ridzi

the clients with lead safe homes in targeted census tracts. This, in turn, could then lead to a decrease in lead poisoning in targeted census tracts (Main Street level data) and ultimately to an overall decrease in lead poisoning in the overall community (Highway level data). Through engaging in this process, Syracuse was able to attain much of the components of success that we also see in the vision set forth by the FRBSF. Key to the FRBSF’s vision was that a well-built complex adaptive system would, “create a market dynamic where independent actors are responding to market mechanisms (i.e., simple decision rules) to achieve a desired outcome (Erickson 2017, p. 43).” In the case of Syracuse, we see a scenario in which each of the main types of actors now had data at their fingertips which better enabled them to do what the marketplace already encourages them to do. Funders seek to demonstrate that they are responsive to community needs, which they could document with Highway level data. Coalitions seek to demonstrate the value of working together by collectively moving a needle in the community, which they could now document using Main Street level data. Nonprofit program designers seek to develop programs with documented need and demand within the community, which they could now document among the clients that they and their partners already serve at a much more precise and convincing level than previously (with Side Street level data). Finally, program implementers could easily identify who among their clients needed specific services and could streamline their outreach to nonprofit partners as well as be reassured with data that they were serving the holistic needs of their clients (with Address level data). Furthermore, these data driven collaborations were by design cross-sector and meant to be conceived and implemented by partners ranging from the health and human services to housing and education fields. This is particularly important given the rising awareness of the social determinants of health and the social determinants of educational success. Studies reveal that core program providers such as the healthcare system and the school system are only responsible for approximately 10% (Schroeder 2007) and 30% (Johns Hopkins University Urban Health Institute 2015) of overall client success, respectively. With the rising popularity of pay for success and compensation for client outcomes as opposed to services provided, this new ecosystem also helps direct service providers to follow emerging best practices that call for an increase in the “human services ecosystem’s capacity for innovation through better data sharing and analysis (Alliance for Strong Families and Communities 2017).”

2.4.3 An Ecosystem Based on Simple Decision Rules In the Syracuse example we see that it is indeed possible to create an ecosystem in which each actor was able to “rest on fairly simple decision rules” (Erickson 2017, p. 43). Furthermore, such decision rules varied (as predicted) by specific actors

2 Goldilocks Data-Connecting Community Indicators to Program …

31

according to the roles they play within the larger system (p. 43). This is made possible in part due to two innovations that did not exist in Syracuse prior to this project but that are essential to the FRBSF vision: a common language of outcomes; and real-time capability for collecting and sharing data. The FRBSF points to the work of Cassidy who writes “an important step to leveraging this growing body of evidence will be to develop a common language in describing outcomes and measuring performance (p. 52).” For the FRBSF this includes a focus on the same target outcomes and doing so across institutions (p. 52). They note that this was something that has been tried before but not yet achieved. In the present case the shared Side Street level data application was built with the purpose of addressing this specific need. It was also created with the capability of sharing data across government as well as nonprofits ranging from hospitals to museums (and everything in between). Given the common denominator of census tract, this Syracuse pilot addresses the FRBSF’s emphasis on an approach that is both multi sector and a place-based intervention (p. 45). Finally, the FRBSF points out the importance of “real-time data to measure progress” that allows for constant evolution guided by data (p. 44). While the 1.0 version built in Syracuse did not meet this criterion, real-time functionality was a key innovation of the 2.0 version.

2.5 Connecting CI-PM in a Complex Adaptive System To truly achieve a complex adaptive system that will help organizations and sectors coordinate their work requires building bridges across both horizontal and vertical silos. In terms of horizontal silos, there is a need to reach out across the boundaries of sectors. This need is perhaps most clearly expressed in our national discussions of the social determinants of health. In this discussion, health is not just about physicians but also about housing, schools, public safety and food networks. While these fields are known to be connected, their actual points of overlap become clear when we look at a specific neighborhood. By looking at neighborhoods, we can see what configuration of challenges and strengths are present. This is perhaps most strongly visible in the data that Syracuse is sharing on the side street level in the Life Needs Assessment (which is by design cross-sector in focus). With these data, a housing professional can now see the specific job training, childcare and health care needs of their clients in a specific neighborhood and potential solutions through partnering with other nonprofits now become feasible because the problems are on a manageable scale (a neighborhood), occur across a known quantity of clients (the ones in that neighborhood that have reported these issues) and can be closely monitored (through continued use of the assessment tool) to see when they have been adequately addressed. Making these types of cross-sector connections on the local level, in the neighborhoods where they can be seen, adds a horizontal integration dimension that helps to tear down the silos that exist across sectors. In the process,

32

F. Ridzi

such cross-sector connections also help workers to see their link to the greater cause beyond their portfolio of responsibilities and become a basis for advocating for a more systems-focused approach. Vertical silos occur when people who operate at different tiers or levels in the same sector do not regularly coordinate their efforts. This can happen for a variety of reasons including lack of communication but also lack of shared data that connects one tier to another. The nested logic model helps to reduce vertical silos by aligning case managers, nonprofits, coalitions and funders. In essence, identifying needs and designing responses no longer needs to be top down, but can also be bottom-up, such as when front-line staff recognize common needs among their clients that affect the larger goals of the community and propose solutions. It also provides an infrastructure that has the potential to make community indicators more accountable to local residents. By helping them express in real time their multiple needs that cross multiple sectors the data infrastructure can be responsive to their personal day-to-day needs. As we will discuss more in future pages, it can also empower them to recommend policy solutions (such as through participatory budgeting processes) because the data infrastructure has the potential to be shared with them in such formats as a phone app. This can further help to address the vertical silos that exist between providers and clients. By building bridges across horizontal and vertical silos, creating a data ecosystem to foster a complex adaptive system can also help a community improve its Community Indicator- Performance Measure integration. The CIC articulated a Community Indicators-Performance Measures (CI-PM) Integration maturity model in 2010 (Community Indicators Consortium 2010). This model described different stages building up to a mature integration of CI and PM. The most mature communities would see a scenario in which “citizen-driven CI’s determine PM impacts linked to quantifiable and measurable results” (Community Indicators Consortium 2010, p. 1). Doing this requires the breaking down of vertical silos between citizens and policy makers so that citizens have a voice that is just as important as government. It also requires breaking down horizontal silos in order to include “key community stakeholders from all sectors” (Community Indicators Consortium 2010, p. 1). To be sure, simply building a data infrastructure will not bring about this model’s most mature components of citizen driven participation and transparent results-based governance. Indeed, there is a whole range of roles that residents can play in any public participation process that require more than simply accessible data (International Association for Public Participation 2013). To truly move forward a culture of data access and participatory consumption must be cultivated. What a data system like the one described here can do is provide the technological infrastructure that empowers a community to move toward greater maturity on CI-PM integration. In this infrastructure, we can combine the participatory qualities that community indicators bring to prioritizing community needs and planning responses with the accountability that PM brings to outcome measurement and return on investment. Syracuse is not there yet, but, as we explore in the following section, a pathway forward is emerging.

2 Goldilocks Data-Connecting Community Indicators to Program …

33

2.6 Conclusion: Looking Forward—Community Indicators Meets the Internet of Things We are in the midst of what has been called the fourth industrial revolution. Whereas the first involved steam machinery (think trains), the second electricity, and the third computers and automation, the fourth promises to be one of connectivities across the physical and digital aspects of our lives (Schwab 2015). A likely part of this is what many are calling the Internet of Things (IOT). As communities become smarter, they are seeking to capitalize on this revolution through such ventures as Syracuse’s addition of free wireless internet access (wifi) throughout the city. This is a significant step forward for the city’s most disadvantaged residents who, though likely to have smart phones, are wary to waste their data plans on responding to community surveys. The addition of universal wifi thus promises to push the community indicators ecosystem to the next level by removing a key barrier to resident participation. Connecting to people has been a goal of community indicators projects (Stevens et. al. 2019, p. 21) and for many in the community indicators field, participatory planning is seen as the way of the future (see for instance public participation geographic information systems -PPGIS). As pointed out in the experience of other communities, geographies that are too large often, “don’t resonate with folks deeply enough to incite the necessary passion” (Stevens et. al. 2019, p. 85). Sustainable Seattle is a great example in which they did not see the progress they hoped for using regional indicators but did see success once they focused on neighborhoods instead (Epstein et al. 2006; Stevens et. al. 2019, p. 85). By breaking things down into census tracts or neighborhoods, and furthermore by collecting data directly from residents, we are not only creating a data ecosystem that drills down to geographic levels of community that are recognizable to people but we are also able to measure concrete needs that resonate with them. With the life needs assessment approach, the local data are no longer just abstract concepts such as percent of people in poverty, but rather needs that people can identify in their day to day living such as needing transportation, childcare, a job or health care. Furthermore, as with other community indicators projects where “early success is important to building momentum and sustainability (Stevens et. al. 2019, p. 81),” Syracuse is poised to break the chain of listening tours with no follow through. With the addition of wifi to the nascent data ecosystem described above, Syracuse is on the cusp of offering real-time feedback to residents who will be able to see how many other people have the same needs as them in the same neighborhood. The funders, coalitions, program designers and frontline staff will furthermore be able to quickly close the feedback loop by letting residents know when and how they are acting on the needs they have shared. The marketplace that the FRBSF hoped for is one that is scalable and can bring about an innovation revolution (p. 52). The data infrastructure built for this study validates the promise of this route since it was built with widely available desktop and open source or free technology. This case study further affirms that this is a chance to reshape how institutions work (Erickson 2017, p. 53) and to use data as the key currency in a new outcomes marketplace. The Internet of Things makes things more

34

F. Ridzi

responsive to people, but what we explore here is that it can also make data itself more responsive to people’s individual needs and uses. The Internet of Things furthermore creates new opportunities for the simple markets that the FRBSF has envisioned. They admit that, “we are at the very beginning of developing the connector tools and vehicles to marry buyers to sellers, but once we have more than a few pilots, the potential here to create a large and vibrant market is enormous (Erickson 2017, p. 48).” The case of Syracuse reveals that such marketplaces can be created with investments in data infrastructure that are very cheap to implement if they also have sweat equity, ingenuity and good will.

References Alliance for Strong Families and Communities. (2017). A national imperative: Joining forces to strengthen human services in America. Retrieved from https://www.alliance1.org/web/resources/ pubs/national-imperative-joining-forces-strengthen-human-services-america.aspx. Community Indicators Consortium. (2010). CI-PM Descriptive Model. Retrieved from https:// communityindicators.net/research/ci-pm-integrations/. Davern, M. T., Gunn, L., Giles-Corti, B., et al. (2017). Best practice principles for community indicator systems and a case study analysis: How community indicators victoria is creating impact and bridging policy, practice and research. Social Indicators Research, 131, 567. https:// doi.org/10.1007/s11205-016-1259-8. Epstein, P. D., Coates, P. M., Wray, L. D., & Swain, D. (2006). Results that matter: Improving communities by engaging citizens, measuring performance, and getting things done (p. 106). San Francisco, CA: Wiley. Erickson, D. (2017). The march toward outcomes-based funding. In What matters: Investing in results to build strong, vibrant communities (pp. 29–54). Copyright (C) 2017 by Federal Reserve Bank of San Francisco and Nonprofit Finance Fund. Also available at: https://investinresults.org/ chapter/march-toward-outcomes-focused-financing. Friedman, M. (2005). Trying hard is not enough: How to produce measurable improvements for customers and communities. Victoria: Trafford Publishing. Hatry, H., & Morley, E. (2008). National Institute for Literacy, Guide to Performance Management for Community Literacy Coalitions, Washington, DC 2008. https://lincs.ed.gov/publications/pdf/ NIFLCommunityLiteracyReport.pdf. Holden, M., Phillips, R., Stevens, C. (2017). Community Quality-of-Life Indicators: Best Cases VII, Springer. International Association for Public Participation (IAP2). (2013, October 21). IAP2’s public participation spectrum. Retrieved from https://c.ymcdn.com/sites/www.iap2.org/resource/resmgr/ foundations_course/IAP2_P2_Spectrum_FINAL.pdf?hhSearchTerms=%22spectrum%22. Johns Hopkins University Urban Health Institute. (2015). Social determinants of education. Retrieved from http://urbanhealth.jhu.edu/SDH_Symposium/SDE.html. Ridzi, F. (2013). Managing expectations when measuring philanthropic impact: A framework based on experience. The Foundation Review, 4(4), 98–109. Ridzi, F. (2017). Community indicators and the collective goods criterion for impact. In M. Holden, R. Phillips, & C. Stevens (Eds.). Community quality-of-life indicators: Best cases VII. Springer. Published April 13, 2017

2 Goldilocks Data-Connecting Community Indicators to Program …

35

Ridzi, F. (2019). Collective action, collective impact and community foundations: The emerging role of local institution building in an era of globalization and declining social safety nets. In M. A. Fallov & C. Blad (Eds.), Social welfare responses in a neoliberal era policies, practices, and social problems (pp. 137–159). Leiden, The Netherlands/Boston, BRILL doi: https://doi.org/10. 1163/9789004384118. Ridzi, F., & Doughty, M. (2017). Does collective impact work? What literacy coalitions tell us. Lanham Maryland: Lexington Books/Rowman & Littlefield. Schroeder, S. A. (2007, September 20). We can do better—Improving the health of the American people. N Engl J Med 357:1221–1228. https://doi.org/10.1056/nejmsa073350 https://www.nejm. org/doi/full/10.1056/NEJMsa073350 Schwab, K. (2015). The fourth industrial revolution what it means and how to respond. Foreign Affairs. Published by the Council on Foreign Relations. Retrieved from https://www. foreignaffairs.com/articles/2015-12-12/fourth-industrial-revolution also available at https:// www.weforum.org/agenda/2016/01/the-fourth-industrial-revolution-what-it-means-and-howto-respond/. Singh, S., Sylvia, M., & Ridzi, F. (2015). Exploring the literacy practices of refugee families enrolled in a book distribution program and an intergenerational family literacy program. Early Childhood Education Journal. https://doi.org/10.1007/s10643-013-0627-0. Stevens, C., deBlois, M., Hamberg, R., and Baldwin, J. (2019). Community Indicators Project Development Guide. Amazon.com

Frank Ridzi PhD, MPA, is Vice President for Community Investment at the Central New York Community Foundation, Associate Professor of Sociology at Le Moyne College and President of the Community Indicators Consortium. Frank has helped to launch and lead community initiatives in areas such as increasing community literacy, reducing lead poisoning and addressing poverty and economic inclusion. He has been involved in launching Community Indicators efforts and has conducted research and written in the areas of collective impact, sociology of work, social policy and philanthropy. His writings have appeared in such places as the Foundation Review, the Journal of Applied Social Sciences, the Journal of Organizational Change Management, and Review of Policy Research. He is a past President of the Literacy Funders Network, an affinity group of the Council on Foundations. Frank holds a Masters Degree in Public Administration and a Ph.D. in Sociology from Syracuse University’s Maxwell School. He also carries a Certificate of Advanced Study in Women’s Studies. Prior to joining the Community Foundation, he served as Director for the Center of Urban and Regional Applied Research at Le Moyne College, where he still serves as Associate Professor of Sociology.

Chapter 3

Strategies for Expanding Indicator Profiles to Small Rural Geographies Jacob Wascalus and Ellen Wolter

Abstract Indicator projects that provide data for small area geographies have largely focused on more dense, urban communities. Yet several years ago, Minnesota Compass, a social indicators project that has provided data profiles for neighborhoods in Minnesota’s two largest cities, decided to expand its products to include small area profiles for cities and towns throughout the state. To find out if our expanded profiles satisfied the data needs of our newest users, we decided to hold a series of listening sessions in greater Minnesota. We encountered two challenges, and organizations with a similar interest can benefit from learning what we did to overcome them. In this chapter, we first explain how we called upon existing partnerships in greater Minnesota to help organize the listening sessions to learn about the strengths and needs of communities across the state. We then discuss how we distilled the feedback we received to improve the profiles we had already created for communities in greater Minnesota.

3.1 Introduction Indicator projects that provide data for small area geographies have largely focused on more dense, urban communities. Minnesota Compass (mncompass.org), a social indicators project that tracks trends in quality of life, has provided data profiles for neighborhoods and communities in Minnesota’s two largest cities, Minneapolis and St. Paul, for more than a decade. Yet when communities beyond these cities expressed interest in similar resources and information, we decided to expand our products to include small area profiles for cities and towns of 1000+ population throughout the rest of the state, or greater Minnesota. In 2014, we debuted this new feature. J. Wascalus (B) · E. Wolter (B) Wilder Research, 451 Lexington Parkway North, St Paul, MN 55104, USA e-mail: [email protected] E. Wolter e-mail: [email protected] © Springer Nature Switzerland AG 2020 F. Ridzi et al. (eds.), Community Quality-of-Life Indicators, Community Quality-of-Life and Well-Being, https://doi.org/10.1007/978-3-030-48182-7_3

37

38

J. Wascalus and E. Wolter

Did our small area profiles satisfy the users who had expressed a desire for data for communities in greater Minnesota? Was the data we included sufficient to serve the needs of communities across the state? We received support in 2017 to travel the state to ask these questions through a series of listening sessions. At the same time, we sought feedback on a tool we were developing that would enable users to create their own profiles by building or drawing custom geographies. We encountered two challenges in our initiative to expand our indicator services to smaller, non-urban geographies: connecting with stakeholders and distilling the feedback we received. Organizations with a similar interest can benefit from learning what we did to overcome them. In this chapter, we will first explain how we called upon existing partnerships in greater Minnesota to help organize a series of listening sessions to learn about the strengths and needs of communities across the state. Doing so provided logistical advantages and insight into the local communities. We will then discuss how we distilled the feedback we received to improve the profiles we had already created for communities in greater Minnesota as well as how we sifted through the suggestions we collected to enhance our customizable profile tool. This process became necessary given the wide-ranging feedback we received and the limited financial resources we had to complete the work.

3.2 Gaining Insight into Diverse Areas by Taking Advantage of Existing Partnerships Minnesota is a big and diverse state, both demographically and economically. As of 2017, it had nearly 5.6 million residents, roughly 45% of whom lived outside the Minneapolis-St. Paul metropolitan region. In addition to a growing proportion of aging residents, the share of the population that lives in greater Minnesota who identify as persons of color has nearly doubled in recent years, from 6% in 2000 to more than 11% in 2017. Many of these residents are new immigrants. Greater Minnesota enjoys a strong mix of industries: natural resource extraction, processing, and shipping in the north and northeast; manufacturing and durable goods distribution in the southeast; and agricultural commodities and value-added food products in the south and southwest. Dotting these regions are more than 350 cities with 1000 or more residents. Like states across the country, some cities are growing in population, some are shrinking modestly, and some are staying the same. With so much economic variety and a population that is growing more diverse and older in various pockets of the state, how could we provide improved data and tools to better understand and support these communities? To answer this question, we organized a series of listening sessions in cities across the state to invite feedback from stakeholders about the strengths and issues facing their communities. Rather than simply guessing at where to host meetings and whom to invite, we collaborated with organizations based in different regions of the state

3 Strategies for Expanding Indicator Profiles …

39

who work with a cross-section of their communities. These organizations knew both appropriate meeting locations and the representative stakeholders we sought to invite to the listening sessions. But the benefits of working with local partners extended beyond just meeting locations and guest lists—the local partners also helped with audience reach and branding. As community stakeholders themselves, the local partners were immersed in regional issues, and we relied on them to help calibrate each meeting to attract a robust number of attendees. We also collaborated with the organizations on outreach materials and invitations, which were cobranded. To advertise the events, we coordinated with them to distribute the invitations through their organizational contact lists. Both Minnesota Compass and the local partners promoted the events on social media. In all, we hosted 12 sessions at locations that made sense based on our and our partners’ determination of need, population, and socioeconomic trends. Four of these occurred in partners’ offices; eight occurred elsewhere (Fig. 3.1). Each convening lasted two hours, and the partner organizations, as the local hosts, provided introductions to kick off the meetings. Over five months, we met with more than 200 people from 75 communities, representing the nonprofit, business, education, health care, philanthropy, and local, state, and federal government sectors. We believe the success of these meetings was directly attributable to the help of each local partner.

3.3 Distilling the Feedback We Received We organized these meetings with two purposes in mind: one, to gather insights on what attendees believed were their communities’ strengths and challenges; and, two, to learn what data or information they needed to better understand these characteristics. To achieve these purposes, we surveyed the attendees at each of the 12 meetings. A dedicated note-taker recorded comments, and we affixed large Post-It notes to the walls on which we invited people to write their thoughts. Attendees who were reluctant to speak up recorded their ideas on feedback forms placed at each table. By the time we concluded our final meeting, we had amassed a sizable volume of feedback. To distill this information, we conducted a theme analysis to find commonality among the strengths and issues facing each community. We organized the feedback data into overarching “strength” and “issue” themes for each region. We then assessed where themes overlapped regionally across the state by visualizing them in a matrix, marking whether the meeting participants of a particular region indicated that a listed strength or issue was characteristic of their community (see Table 3.1).

40

J. Wascalus and E. Wolter

Fig. 3.1 Minnesota Compass outreach map

Meeting attendees also provided ideas for the information they needed to better track and understand their communities. We used these ideas, along with the results of the theme analysis, to guide select data enhancements for our profiles. We intended for the new data enhancements to correspond to a strength or issue that crossed all regions, which we could view in our matrix. After identifying a potential data source that reflected a strength or issue, we reviewed it for availability and reliability. For example, we heard from stakeholders in every region we visited that childcare affordability and availability, as well as workforce shortages, were growing concerns, particularly to employers and new or expectant parents. These concerns are related. If the costs associated with childcare outweigh the benefits of

3 Strategies for Expanding Indicator Profiles …

41

Table 3.1 Theme analysis matrix Regions Strengths and Issues

Central

West Central

North west

Northeast

Southwest

Southern

High cost of and limited availability of childcare Difficult to collaborate with neighboring communities

Workforce shortage

Limited affordable and accessible high-speed broadband Limited economic growth potential

Affordable and available housing

working, then a parent may opt to stay home, shrinking the supply of workers. To provide more information about this topic, we examined different data options and settled on adding information about childcare costs to our profiles. The data came from a reputable source—Minnesota’s Department of Employment and Economic Development (DEED)—and met our criteria for inclusion. We also added workforce and commuting data, including median wages by industry and distance to work. Some of this data also came from DEED, as well as the Census Bureau’s Longitudinal Employer-Household Dynamics (LEHD) program. Another example involved housing data. Although every region we visited expressed concern about housing affordability and availability, meeting attendees in northern Minnesota had particular concerns about secondary homes. Thousands of lakes dot that region of the state, and a large number of housing units are seasonal or secondary homes. Some of these houses have a high market value, and this higher-priced stock, which is often vacant, pushes up land and housing values while obscuring the reality of a limited supply of available—and affordable—year-round dwellings. To provide more information about this issue, we added data about vacant seasonal housing units to many of our profiles. The data came from the Census Bureau’s American Community Survey (ACS), a reliable and regularly updated data source.

42

J. Wascalus and E. Wolter

These meetings also provided us with an opportunity to solicit feedback on a data tool we were developing that would enable users to create customized geographic profiles across the state. Called the Build Your Own profile tool, or “BYO tool,” the application was a proposed expansion of a tool we had released for the Twin Cities metropolitan area several years prior. The BYO tool would give users three ways to create profiles: by selecting established boundaries that we loaded into the tool, such as school districts or state legislative districts; by comparing, side-by-side, any two of these boundaries, such as two census tracts or a school district and a Minnesota state house district; or by drawing custom geographic boundaries that do not conform to established boundaries. Users creating custom profiles would be able to draw boundaries with a polygon tool or with a line and point tool that they could then buffer with a distance feature. Regardless of the method users employ to create profiles, the BYO tool would generate a profile that presented a range of demographic, economic, housing, and other data, drawn largely from 5-year ACS estimates, LEHD, and Decennial Census sources1 We invited feedback on the BYO tool’s usability and appearance, and we wanted to know which established political geographies we should incorporate into the tool’s design. Ultimately, we moved forward with enhancements based on two criteria: if the proposed upgrade would improve functionality and user experience, and if we could implement it within our timeline and budget.

3.4 Lessons Learned Holding stakeholder meetings greatly benefited Minnesota Compass’s indicator services to smaller, non-urban geographies. We learned about strengths and issues facing a range of communities, and we learned about the data they sought to help them understand these characteristics. The success of these meetings would not have been possible without the help of local organizations. Distilling the feedback we received through a methodical theme analysis helped us guide our data enhancements. In the end, we added data for eight strengths and issues (see Table 3.2). We also collected valuable suggestions on improving our statewide BYO geography tool, which we factored into the application’s development before launching it after the conclusion of our regional meetings (see Table 3.3 and Fig. 3.2). And we’re not done: We’re planning future improvements based on the feedback we received at the meetings (see Table 3.4). 1 For

customized profiles that do not conform to established political boundaries, the BYO tool creates a unique data profile using small geographic areas, such as census tracts and block groups, by apportioning data for custom-drawn boundaries. For example, for a customized geography that incorporates 35% of a census tract’s land area, 35% of that census tract’s attributes would be included in the new profile. (For more information about the BYO tool’s methodology, visit https://www. mncompass.org/profiles/custom/faq.)

3 Strategies for Expanding Indicator Profiles …

43

Table 3.2 Sample list of data enhancements to Minnesota Compass’s small area profiles Data enhancements

How it helps or provides context

Cost of living

The ability to afford childcare, housing, and basic needs was noted as a key issue in all greater Minnesota regions, affecting employees and employers. Minnesota’s Department of Employment and Economic Development provides cost-of-living data specific to these issues, based on different family structures.

Median wages

To better understand whether median wages are keeping up with the cost of living in Minnesota, we will display median wages for the primary industries in all Minnesota counties. This data will allow local communities to compare wages against cost-of-living requirements in counties, regions, and Minnesota.

Preschool enrollment and parents working

Data available through the American Community Survey details the number of children aged 3 and 4 who are enrolled in preschool and the number of households with two parents working full-time. Enrollment in preschool provides a proxy to evaluating the number of families who have access to and are able to afford preschool-age childcare.

Workers by industry

To provide in-depth context about local economies, the updated greater Minnesota profiles will detail the percentage of workers employed by industry statewide, and for regions, counties, and cities. Data is available from the Census Bureau’s Longitudinal Employer-Household Dynamics program.

Policymakers and practitioners across the state are already using the BYO tool. In Duluth, program administrators of the city’s Early Childhood Family Education program have used the build and draw functions of the BYO tool to identify the location and number of children birth to four years of age. They need this information both to satisfy a needs-assessment report that they are required to submit to state officials and to help them better focus their outreach efforts. The tool, explains one of the administrators, is invaluable.

3.5 Conclusion The reaction by the Duluth administrator is exactly the kind we hoped for when, several years ago, we made available hundreds of preset data profiles of smaller, non-urban geographies and began developing plans for expanding our BYO tool. The information and resources we provided then were indeed valuable, but we knew

44

J. Wascalus and E. Wolter

Table 3.3 Sample list of enhancements to Minnesota Compass’s Build Your Own profile tool Mapping tool enhancement

How it helps or provides context

Expand the custom mapping tool statewide with ability to build, draw, and compare customized data profiles for all of Minnesota

Users will be able to create customized data profiles for any area throughout the state of Minnesota. All custom mapping profiles include demographic, housing, workforce, and economic data specific to the user’s area of interest.

Enlarge map screen and the ability to zoom in and out, similar to “Google maps”

Users will have an improved ability to view the map. The screen area will be increased to accommodate better navigation of our build, compare, and draw tool and enhance user ability to customize data profiles.

Add new “Geographic View” options to the custom mapping tool

Users will be able to view Minnesota broken into different geographies: 201 state house and senate districts, 336 school districts, 8 U.S. Congressional districts, and distinct regions identified at the listening sessions.

Include an illustration of the geographic area built or drawn in the custom data table profile

For all custom area profiles that users build or draw, the data table will now include a graphic of the custom geographic area of interest.

that our users wanted more. We set out to learn what stakeholders living and working in greater Minnesota wanted, and in the end we delivered a resource that better serves communities across the state. Organizations wishing to create geographic profiles of areas they are unfamiliar with can follow similar steps to the ones we took when creating and expanding profiles for communities in greater Minnesota. In particular, organizations should: • Talk to stakeholders directly. We held meetings throughout greater Minnesota to learn from the communities that we wanted to build profiles for. Doing so helped us better understand what was important to them and improve the value and usefulness of the profiles and tools that we were developing. • Work with a local partner. The success of these meetings would not have been possible without the help of local organizations. A local partner can help take a lot of the guesswork out of stakeholder outreach and serve as a trusted, locally recognized co-convener. • Use stakeholder feedback to strategically enhance your work. We needed to develop a way to update our profiles in a way that maintained data uniformity and consistency. Distilling the feedback we received through a methodical theme analysis helped us achieve this goal and ultimately led to data enhancements that our users valued.

3 Strategies for Expanding Indicator Profiles …

Fig. 3.2 Example of using the Minnesota Compass BYO tool

45

46

J. Wascalus and E. Wolter

Table 3.4 Sample of future feature enhancements for Minnesota Compass’ Build You Own profile tool Increase users’ ability to interact with data and mapping tools on the site; potential enhancements include additional comparison options, visual charts and graphics, and mapping functions. Add the ability to download an Excel file of customized data profiles that users have built, drawn, or compared. Update the visual presentation of data through charts, graphs, and infographics. Develop ability to import shapefiles into the mapping tool to create profiles for those specific geographies.

Jacob Wascalus For more than a decade Jacob Wascalus has promoted community vitality by researching, writing, convening, and presenting on a wide variety of subjects, including housing, transportation, workforce development, and neighborhood revitalization. He currently works on Minnesota Compass, an indicators project that tracks trends in Minnesota on more than a dozen quality-of-life topics. Previously, he held positions at the Federal Reserve Bank of Minneapolis, the Center for Urban and Regional Affairs, and the Institute for Agriculture and Trade Policy, among others. Jacob earned a master’s degree in urban and regional planning from the Hubert H. Humphrey School of Public Affairs at the University of Minnesota and a bachelor’s degree in English from James Madison University in Virginia. Ellen Wolter serves as a research scientist for Minnesota Compass where she tracks indicators for the project which provides community-level trends for Minnesota residents. Ellen analyzes community-level data trends by working collaboratively with Minnesota organizations to identify data indicators that are relevant to their everyday work. Before joining Wilder Research, Ellen worked for the University of Iowa’s Center for Evaluation and Assessment and the University of Minnesota’s Office of Community Engagement for Health conducting research and evaluation related to education, health, and social services. She has worked with community-based organizations, state and local government, and higher education institutions for the past 15 years to develop indicators that support informed decisionmaking. Ellen holds dual master’s degrees in public health and public policy from the University of Washington and a bachelor’s degree in English from Grinnell College.

Chapter 4

Measuring the Dream for an Equitable and Sustainable Future Katie O’Connell, Andrea Young, and Nisha D. Botchwey

Abstract More than 50 years after Dr. Martin Luther King, Jr.’s assassination, what has become of his call for social, political, and economic equality for African Americans? How do we equip a new generation with knowledge, techniques and strategies of past activists, and assessments of the policies that were implemented and/or defeated in the struggle for justice and equality? How do we measure our progress toward a more just and equal society? The Measuring the Dream (MTD) project informs, enables, and inspires a diverse audience to understand the history of the journey toward a more equal society. To quantify national changes in equity since the 1950s, the project’s MTD Index identifies six arenas in which the struggle for equal justice has taken place and can be objectively traced. Data within each area are cataloged and then combined to calculate a MTD Index score from pre-1950–2016. No single index exists with this set of indicators that provides the scaling and longitudinal framework necessary to assess America’s strive towards prosperity and equality for all. This chapter describes the importance of a national equity measurement, the process to select indicators, the challenges of historical data, and construction of the MTD Index.

4.1 Introduction In his 1963, “I have a Dream” speech Dr. Martin Luther King Jr. issued a call for racial equality and economic justice, infusing the American Dream with new meaning and fresh purpose. Yet there are differing opinions by Blacks and Whites on the current K. O’Connell · N. D. Botchwey (B) Georgia Institute of Technology, Atlanta, GA, USA e-mail: [email protected] K. O’Connell e-mail: [email protected] A. Young American Civil Liberties Union, Atlanta, GA, USA e-mail: [email protected] © Springer Nature Switzerland AG 2020 F. Ridzi et al. (eds.), Community Quality-of-Life Indicators, Community Quality-of-Life and Well-Being, https://doi.org/10.1007/978-3-030-48182-7_4

47

48

K. O’Connell et al.

state of equality in twenty-first century America. We need an understandable measure that quickly demonstrates how measurable data on every area of American life that prompted the Civil Rights Movement compares African-American reality to reality for White Americans. The Measuring the Dream (MTD) Index fills this void with an analysis of six domains that also have strong, measurable, national-level data sets— education, criminal justice, citizenship rights, health, housing, and poverty. Each domain is comprised of five indicators, each coming from varied sources including the U.S. Centers for Disease Control and Prevention (CDC), the U.S. Census, and the U.S. Bureau of Labor Statistics (BLS). The goal of the MTD Index is to provide a comprehensive point of reference as a tool for research on race in America and to stimulate important policy and planning related discussions. In this chapter, a description of the Measuring the Dream Index is a starting place to understand changes in Black-White equality since the 1950s. The lessons from building the Index help frame the challenges to discussing community well-being in the context of equality. We will discuss ways to measure equity, the importance of a holistic view of inequality, where inequality measures should locate in the context of history, and why scale matters when talking about equality.

4.2 What Is Equality? The preamble of the Declaration of Independence states, “all men are created equal.” However, when those words were written in 1776, they did not recognize all members of society, only property-owning white males. King spoke about this dichotomy in his iconic “I Have a Dream” speech in 1963 when he called the US Constitution and Declaration of Independence “promissory notes… (that) guaranteed the unalienable rights of life, liberty and the pursuit of happiness”. He, like so many scholars, saw equality as the central tenet of the United States and the fight for civil rights was to “make real the promises of democracy” (King 1963, p. 2). Through court cases, like Shelley v. Kraemer (1948) and Brown v. Board of Education of Topeka (1954), where African Americans achieved an end to restrictive housing covenants and access to desegregated schools, King turned his attention towards the struggle for “genuine equality.” However difficult, winning these battles was easy when compared to redesigning a society with equal access to quality jobs, housing, and education. In his opinion, equal access to voting booths did nothing if people could not afford to live. He saw the national challenge of moving beyond equal rights for all to a society that valued Blacks as much as Whites. He described the distinction as an incremental improvement compared to substantive equality. Polls show that in 1970, a majority of Black and White Americans saw racial equality as a top priority for policymakers. After the turmoil of the 1960s, the nation felt unified towards a common goal. Policies like affirmative action worked to correct injustices faced by generations of African Americans. However, half a century later,

4 Measuring the Dream for an Equitable and Sustainable Future

49

racial equality appears more like a fading fantasy than an achievable future. The Pew Research Center reports that less than 50% of Americans believe a significant amount of progress has been made towards King’s dream of racial equality with twenty-seven percent of Blacks see almost little to no progress (Pew Research Center 2013). At the same time, there are diverging opinions between Black and Whites regarding the state of race and racism in America. We have become two nations and a country of strangers (Hacker 1992; Shipler 1997) especially around race. A 2016 Pew poll found that 70% of African Americans compared to 36% of Whites believed racial discrimination made it harder for African Americans to get ahead. Forty-three percent of Blacks compared to eleven percent of Whites believe the country will not make the necessary changes to give Blacks equal rights to Whites (Pew Research Center 2016). Eibach and Ehrlinger (2006) find that Whites and Blacks view the goal of equality differently. Whites tend to view how far society has come from the racial divisions of the past, since the days of Jim Crow and segregated lunch counters. For African Americans, the focus is on how far the nation still has to go until those two groups are equal. Essentially, “compared with where we were there is progress. Compared with where we should be, that progress is insufficient” (Wolfe 1998, p. 223). These differing views demonstrate two of the three standard measures of equality most often used by policy-makers and think tank—absolute and relative. Using an absolute measure of equality, we calculate how far Blacks have come from the past (Coleman 2016). For example, 10% of Blacks age 25 or older completed high school in 1940 compared to 85% in 2012 (Coleman 2016) which is a 750% increase. Across many measures, Blacks have seen significant improvements since the 1950s compared to Whites. While this measure has importance, it demonstrates improvements but does not readily identify disparities between groups. To highlight disparities, it is best to demonstrate using a relative measure of equality, which takes a Black-White ratio to illustrate the gap between the two groups. For example, the median income (in 2014 dollars) for Blacks in 1950 was $18,359 and $35,398 in 2014. As an absolute increase, that is 92%, which seems impressive especially compared to Whites who saw only a 68% increase. However, as a relative measure, Blacks in 1950 earned just more than half as much as Whites and by 2014 that only increased to 62%. By taking a relative measurement compared to absolute, it is clear there continues to be vast income inequality between Blacks and Whites. Another option we explored was the idea of setting a goal for both groups and then calculating how far each group has to reach that goal. This is the method used by the United Nations for its Millennium Development Goals. While this is useful for the UN, for the MTD Index, we found it difficult to determine the appropriate benchmarks. For example, homeownership opportunities are very different in New York City compared to suburban Atlanta. As Eibach and Ehrilinger (2006) note, setting goals can be different among individuals. We believed setting a benchmark created a value judgment.

50

K. O’Connell et al.

4.3 Equality in the Context of History Historical context is necessary for the public to understand the root cause of inequality as data alone does not sway public opinion. For example, being “presented evidence of extreme racial disparities in the criminal justice system, cause(s) the public to become more, not less, supportive of punitive criminal justice policies that produce those disparities which could reinforce their views. By placing the disparate data points in the context of historical challenges may help to change people’s minds” (Hetey and Eberhardt 2018, p. 184). When we talk about unequal outcome between indicators, we must recognize they did not happen within a vacuum. Instead, they are part of a historical legacy that continues to affect the African American community decades after courts made racially motivated policies illegal. For example, the GI Bill was one of the greatest wealth builders for Americans after World War II. It gave returning soldiers access to free money for home purchases and college education, both of which have substantial wealth building potential. However, the GI Bill was not available to all soldiers. To pass the Bill, John Rankin of Mississippi, known for his inflammatory racist language, pushed to make the handouts from the GI Bill determined by states rather than federally mandated. Ultimately, southern states denied most Black veterans access to GI benefits. For example, a 1947 survey found that “of 1700 veterans employed in the veteran’s Administration in one southern state, only seven are Negroes” despite the fact that Blacks comprised a third of all southern veterans at the time (Herbold 1994). Unequal treatment was not only a southern problem. Across the United States cities like Chicago IL, Portland OR, and Los Angeles CA participated in redlining. Redlining started in the 1930s as a practice by the Home Owners’ Loan Corporation (HOLC) to create “Residential Security” maps of U.S. cities. Neighborhoods received grades from “Best” to “Hazardous” based on a number of criteria including housing quality, amenities, and racial composition of residents. Neighborhoods were colorcoded based on their criteria with the lowest level marked in red. Those high-risk areas did not receive home loans and this kept many African Americans from purchasing houses. Since home ownership is the largest producer of generational wealth, the financial impact of redlining is still felt today. At the same time, neighborhoods that were redlined continue to be neglected long after the practice was made illegal (Mitchell and Franco 2018). While using historical data can be beneficial, it poses many problems. Longitudinal data is rarely located in a single, downloadable spreadsheet and it takes time to locate data points for each year. Some very old documents are handwritten, which can be especially time-consuming as it is necessary to manually input each data point. Another challenge of historical data is the lack of disaggregation by race as well as evolving definitions. For example, most publicly available data before 1964 is divided into White and Nonwhite, but the Voting Rights Act of 1965 expanded the need for accurate racial tabulations so most data sets post-1964 are available with Black as a separate category (National Research Council 1995). Time-series analysis

4 Measuring the Dream for an Equitable and Sustainable Future

51

of Census data by race gets further complicated with changes to the 2000 Census that allowed respondents to choose more than one race. Projects interested in panel data by race must be aware of these challenges and plan accordingly.

4.4 Data for Equality For the Measuring the Dream Project we relied primarily on public source data. We collected over 400 data points spanning a range of years from the early 1900s through the present. To reduce the data to a manageable number, we used the Thriving Cities Indicator Explorer database (explore.thrivingcities.com) as a guide (Thriving Cities Group, n.d.). The database used a panel of field experts to classify each indicator by academic strength. The five strength categories were very strong, strong, promising, slight, and inclusive. A majority of our indicators fell in the very strong and strong categories. We reduced the number to thirty-three and organized them into six domains—criminal justice, education, health, housing, poverty, and rights. To further verify our data, we presented at the Measuring the Dream Scholar’s Conference and a Technical Advisory Board, which consisted of community leaders, economic development and spatial analysis subject matter experts, and researchers from Thriving Cities and the Atlanta Metropolitan Planning Council. The full set of indicators is in Table 4.1, including definitions, years available, and data source followed by a section describing our rationale for choosing each indicator. To standardize the data, we took a ratio of the Black vs. White population. Section 4.5 “Measuring the Dream Index” gives a more detailed explanation of our calculations.

4.4.1 Criminal Justice When indicator projects look at criminal justice, they most often look at crime in the community. While this does indicate quality of life, at the same time, the over-policing of communities has led to vast disparities. Overall incarceration has increased 340,000 in the 1970s to over 2.3 million, but the disparity is more significant for communities of color. For example, 1 in 3 African American males in a lifetime can expect to be imprisoned compared with 1 in 17 White males (Bonczar 2003). In communities of color, the presence of police is higher, which often translates into higher arrest rates especially for low-level crimes like minor drug possession, even though studies find Whites use drugs at a higher rate than Blacks do. For example, 60% of police stops in Oakland were African American even though they comprise only 28% of the population. Once stopped, they are more likely to be handcuffed, searched, and arrested (Hetey et al. 2016). The impacts of incarceration on individuals include interpersonal distrust, alienation, social withdrawal, and diminished self-worth (Haney 2003) as well as voter disenfranchisement. Over

52

K. O’Connell et al.

Table 4.1 Measuring the dream index indicators Indicator

Definition

Years available

Source

Arrests

Arrests per 100,000 population

1954–2015

Bureau of Justice Statistics

Incarceration

Incarceration per 100,000 population

1950–2014

Bureau of Justice Statistics

Capital punishment

Executions per 1,000,000 population

1978–2016

Death Penalty Information Center

Violent crime rate

Victims of violent crime per 1000 population age 12 or older

1993–2016

Bureau of Justice Statistics

Probation

Probations per 1,000,000 population

1993–2016

Bureau of Justice Statistics

Suspension

Percentage of public school students in grades 6 through 12 who had ever been suspended

1993–2013

U.S. Department of Education, National Center for Education Statistics

Completed 4 years or more of high school

Percent of population that completed four years or more of high school

1950–2016

Census, Current Population Survey

Adult illiteracy

Percent of population classified as illiterate

1952–1979

Census, Current Population Survey

Preschool enrollment

Percent of population age 3–4 enrolled in school

1965–2014

Current Population Survey

4th grade reading

National assessment of educational progress score

1974–2014

U.S. Department of Education, National Center for Education Statistics

College graduate or more

Percent of population that are college graduates or more

1960–2016

Census Bureau, Census of Population and Current Population Reports

8th grade math

National assessment of educational progress score

1977–2014

U.S. Department of Education, National Center for Education Statistics

Advanced placement (AP) pass

Percent of AP test takers that pass exam

1997–2016

College Board

Criminal justice

Education

(continued)

4 Measuring the Dream for an Equitable and Sustainable Future

53

Table 4.1 (continued) Indicator

Definition

Years available

Source

Bachelor’s degree median income

Median income for full-time, year-round workers age 25–34 with bachelor’s degrees

1995–2016

Current Population Survey, National Center for Education Statistics

Life expectancy

Average period that a person may expect to live in years

1950–2014

Center for Disease Control—National Vital Statistics System

Infant mortality

Deaths per 1000 live births

1950–2014

Center for Disease Control—National Vital Statistics System

Age-adjusted death rate—diseases of heart

Deaths per 100,000 population

1950–2015

Center for Disease Control—National Vital Statistics System

No health insurance

Percent of population under age 65

1984–2016

Center for Disease Control—National Health Interview Survey

4 or more chronic conditions

Percent of respondents to the National Health Interview Survey

2002–2015

Center for Disease Control—National Health Interview Survey

Home ownership

Percent of population that own a home

1950–2014

American Housing Survey

1-year residential stability

Percent of population living in residence for 1 year or less

1960–2014

Ipums

Median Home price

Average home value in dollars

1960–2014

Census

Cost burdened households

Percent of families pay more than 30% of their income for housing

1987–2014

Census

Travel time to work

Percent of population that travels 90 min or more for work

2008–2016

Ipums

Unemployment

Percent of persons looking for employment

1954–2014

U.S. Department of Labor, Bureau of Labor Statistics

Median income

Median income in dollars

1950–2014

Current Population Survey

Health

Housing

Poverty

(continued)

54

K. O’Connell et al.

Table 4.1 (continued) Indicator

Definition

Years available

Source

Child poverty

Percent of population under age 18 living in poverty

1973–2014

Current Population Survey

Food insecurity

Percent of population that reports of reduced quality, variety, or desirability of diet

1995–2014

U.S. Department of Agriculture

Median family wealth

Median family wealth in dollars

1983–2016

Urban Institute

Voted presidential

Percent of population that participated in presidential election

1964–2016

U.S. Census Bureau, Current Population Survey

Registered

Percent of population that holds active voter registration

1966–2016

U.S. Census Bureau, Current Population Survey

Congressional representation

Percent of elected federal government representatives per total population

1950–2018

Statistics of Presidential and Congressional Election

Voted non presidential

Percent of population that participated in non-presidential election

1966–2014

U.S. Census Bureau, Current Population Survey

Rights

7% of adult African Americans cannot vote due to former convictions compared to 1.8% of non-African Americans. Four states (Florida, Kentucky, Tennessee, and Virginia) all have more than one in five African Americans unable to vote (Uggen et al. 2016). At the same time, the impact of incarceration does not touch only the incarcerated. Estimates show approximately 11% of children are at risk of having a parent incarcerated. Put another way, 50–75% of incarcerated individuals report having a minor child. The impact on children of an incarcerated parent includes depression, antisocial behavior, and changes in parent-child relationships (Martin 2017). For the Measuring the Dream Index we included incarceration and arrest rate both adjusted per 100,000 U.S. population. As already noted, incarceration has an impact on individuals but arrests, even if they do not end in incarceration, can have long-term impacts. For example, the cost of bail can be too high for individuals, and they stay in jail for extended periods, which can lead to the loss of jobs, eviction, and loss of child custody. Probation is another indicator that we included. While probation allows individuals to reenter the world and reunite with their families, it can be a thin line between freedom and incarceration. Lack of employment and stable housing can send people back to prison while many places do not allow formerly incarcerated

4 Measuring the Dream for an Equitable and Sustainable Future

55

to be residents or employees. The US banned capital punishment nationally in 1972 but reinstated it in 1976. We included this indicator as it can show sentencing disparities between Black and White defendants. We also included 6th–12th-grade school suspension as research shows disciplinary practices can affect students’ social, emotional, and academic achievement (Thriving Cities Group, n.d.). At the same time, it is recognized that suspensions are early step on the school-to-prison pipeline. Finally, we included violent crime rate, as it is a reliable indicator of community health.

4.4.2 Education School desegregation officially took place after Brown vs. Board of Education ruling in 1954. Some school systems followed the law while others, like Prince Edward County in Virginia, chose to close its public schools rather than desegregate. The Prince Edward County schools remained closed from 1958 to 1964. In 1964, the Civil Rights Act mandated the Equality of Educational Opportunity study, better known as the Coleman Report, to understand Black-White inequality in the US education system. The massive study looked at 600,000 students across 3000 schools, and it changed the educational policy narrative that defined “good schools” by its inputs (expenditures, school size, etc.) to one focused on outputs (knowledge acquisition, long-term employment, years of education, etc.). The Report lost both lay and academic readers with its dense use of statistical data and became most associated with the conclusion that student problems were first the fault of the family followed by the schools. This conclusion left many uneasy with the report so many forgot its most significant findings. For example, “the average Black 12th grader in the South had an achievement level that was comparable to the average 7th grader in the urban Northeast” (Hanushek 2016, p. 25). Since the publication of the Coleman Report, great strides have happened over the past half-century for African American students, but the gap still has not closed. Hanushek (2016) found that at the current rate of improvement, it would take two-and-a-half centuries before Black-White math gap closes and over one and a half centuries until the reading gap closes. For the MTD Index, a number of our indicators changed as societal norms changed. For example, the Census Bureau stopped collecting adult illiteracy at the national scale after 1979, so the indicator was removed and replaced by 4th-grade reading, which was assessed using the National Assessment of Educational Progress (NAEP). Studies find that 4th-grade reading is a reliable indicator for improved long-term learning outcomes as it is a child’s transition period from “learning to read to reading to learn” (Thriving Cities Group, n.d.). Completed four or more years of high school was replaced by college graduate or more as we felt the job market starting in the 1980s required a higher education than a high school diploma. Other indicators we included were 8th-grade math NAEP scores as eighth-graders with math proficiency are more likely to graduate from high school (Thriving Cities Group, n.d.). We included preschool enrollment because early childhood education has long-term math and reading skill benefits (Tucker-Drob 2012). We included median income for

56

K. O’Connell et al.

workers age 25–34 with bachelor’s degrees because we wanted to see if economic disparities disappeared between Blacks and White with the same level of education. We found that disparities continue, as Black workers between ages 25 and 34 have a median income of $41,700, well below $55,990 earned by Whites with a bachelor’s degree (Current Population Survey, National Center for Education Statistics 2016). In fact, African American workers with a masters earn $54,760, which is less than Whites with a bachelors (Digest of Education Statistics 2018, n.d.). We included Advance Placement or AP test pass rate as that indicates preparedness for college.

4.4.3 Health The health disparity between Black and White has multiple roots. Lack of access to healthy foods along with poor community design contributes to uneven health outcomes among populations (Marmot 2005). Both social determinants of health are exacerbated by structural racism as it codifies unequal distribution of goods into society’s customs, practices, and laws (Jones et al. 1991) thus challenging communities of colors from achieving health equity which is apparent as poor health outcomes are seen in minority communities independent of socioeconomic status. For the MTD Index, we looked at two standard indicators—life expectancy and infant mortality. We included the age-adjusted death rate from cardiovascular diseases. The Center for Disease Control—National Vital Statistics System reports that prevalence for Black men is lower than White men it is higher for Black women than White women. Though prevalence is lower, the death rate is higher. We also included the percent of respondents to the National Health Interview Survey that had one or more chronic conditions (hypertension, coronary heart disease, stroke, diabetes, cancer, arthritis, hepatitis, weak or failing kidneys, chronic obstructive pulmonary disease, or current asthma). Chronic conditions can increase a person’s risk of dying prematurely, being hospitalized, and have significant health care costs (Centers for Disease Control and Prevention, n.d.). Finally, we included insurance as studies find that health improves, especially for population subgroups, with the addition of health insurance (Levy and Meltzer 2008).

4.4.4 Housing The Fair Housing Act, which outlawed discriminatory housing practices, barely became law in 1968 when it passed the Senate by one vote. Senator Brooke of Massachusetts, the first African-American elected to the Senate by popular vote, cautioned the “Fair housing (Act) does not promise an end to the ghetto. It promises only to demonstrate that the ghetto is not an immutable institution in America” (Brooke 2007, p. 176). His words ring true as numerous studies show that unequal

4 Measuring the Dream for an Equitable and Sustainable Future

57

treatment continues. The subprime mortgage that led to the Great Recession (2007– 2009) impacted African American communities at a higher rate than White which in turn lead to a slower post-Recession recovery. Besides discriminatory financial practices, the practice of redlining neighborhoods had generational impacts on African American homeownership. For the MTD Index, we looked at home ownership, median home price, one-year residential stability, and cost-burdened households. We chose home ownership as it spoke to discrimination, but it also spoke to the wealth that it built from ownership. Median home prices also showed wealth building. Studies also indicate that predominantly Black neighborhoods are consistently valued less than White’s, which is a holdover from redlining. For lower-income African American communities, displacement is a common challenge hence the reason we chose one-year residential stability and cost-burdened households. These indicators speak to the challenges of gentrification. Travel time to work is an important measure as it can indicate a spatial mismatch of employment, which can cause higher childcare costs, lower job stability, and lower quality of life (National Equity Atlas).

4.4.5 Poverty In 1962, Michael Harrington published the groundbreaking book The Other America: Poverty in the United States. This book became a best seller and brought to light the struggles of the poor to a broad American audience who had increasingly become isolated from poverty and was thus blissfully unaware. Some say this book pushed both Kennedy and Johnson to fight against poverty as well as inspired students to join the fight for civil rights. Since the publication of the book, the war on poverty shifted to a war on the poor. Starting with Nixon and swelling under Reagan, the rhetoric shifted to demonize the poor (Chappell 2010, p. 201). With cuts in federal funding, the message became clear that poor people are in their situation due to their laziness and lack of motivation rather than the circumstance of which they were born. This same message continued into the twenty-first century when Speaker of the House Paul Ryan stated, “We have got this tailspin of culture, in our inner cities in particular, of men not working and just generations of men not even thinking about working or learning the value and the culture of work, and so there is a real culture problem here that has to be dealt with” (Lowery 2014). For the MTD Index, we used three standard measures of poverty—unemployment, median income, and child poverty. We included food insecurity as it measures financial stability. The USDA defines food insecurity as, at some point during the year, some of the household members were uncertain or unable to acquire enough food because they had insufficient money. The impacts of food insecurity are long. For children, it is related to cognitive problems, increased aggression and anxiety, increased hospitalizations, and a higher probability of asthma. For adults, it can include mental and physical health problems, diabetes and other chronic diseases, and depression (Gundersen 2013). Median family wealth is an important measure

58

K. O’Connell et al.

as it measures assets and can demonstrate a family’s financial security, ability to produce more personal wealth, and improve future outcomes for their children. In 2016, White family wealth was seven times greater than Black family wealth (The Urban Institute, n.d.).

4.4.6 Rights Quality of life indicator projects do not always incorporate citizen rights, and when then do, they tend to focus on voting. In 1948, the United Nations included the right to vote as a dimension of its Universal Declaration of Human Rights (United Nations 1948, p. 6) and today, 74% of US adults say it is “very important to vote in elections in order to be a good citizen” (Pew Research Center 2018, p. 93). Yet, since the Shelby County v. Holder case in 2013, which ruled two provisions of the 1965 Voting Rights unconstitutional, we have seen voting rights challenged across the US. One of the provisions no longer upheld was Sect. 4.5, which required jurisdictions who had previously enacted racial discriminatory voting practices to submit formal voting changes to the Department of Justice. Shelby v. Holder threw this out and between 2013 and the 2016 presidential election, 868 polling places were closed in those jurisdictions formerly covered by Sect. 4.5 (Simpson, n.d.). Other practices that make it hard to vote include stricter voter ID rules, reduced voting hours, and purging of voter rolls. All of these practices affect minority voters strongest. For the MTD Index, we looked at congressional representation, registered voters, presidential election voters, non-presidential election voters. Congressional representation is the only Rights indicator that spans the entire index, and it is calculated as a percent of elected federal government representatives per total population. Following the 1965 Voting Rights Act, voting data was collected so we were able to add registered to vote in 1969. It was based on the percent of voting age population holding an active voter registration.

4.5 Measuring the Dream Index For the Measuring the Dream Index, in creating a longitudinal measure of equality, we wanted something that demonstrated the continued gap between Black and White, included multiple measures to ensure a holistic understanding of issues, and wanted an easily understood number. The MTD Index followed a similar methodological approach as the Canadian Index of Wellbeing (CIW), which created an index to measure change over time across eight domains—healthy populations, democratic engagement, community vitality, environment, leisure and culture, time use, education, and living standards.

4 Measuring the Dream for an Equitable and Sustainable Future

59

Fig. 4.1 National measuring the dream index and domain scores

The CIW score was calculated in three steps: (1) for each indicator, the percent change from 1994 to each year was calculated. For negative indicators, the inverse was used. (2) The average mean was taken of the indicators for each domain. (3) The average mean of the domains were calculated to arrive at the index score for each year. The MTD Index followed this method with one notable exception—the CIW was interested in an absolute measure of equality so they used a percent change from year 1 (1994) while we wanted relative measure of equality so we calculated an annual ratio of the Black to White population. Figure 4.1 demonstrates the changes in index score from 1954 to 2014. Values range between zero and one with those closer to zero demonstrating large inequality. When the index value is one then Blacks and Whites have equal outcomes. Between 1954 and 2014, the overall MTD Index score increased from 0.47 to 0.64 with the most significant gains in the domain of Rights (0.05–0.84). The housing and education domains both saw moderate increases, with housing increasing from 0.57 to 0.75 and education from 0.68 to 0.76. It is not surprising the scores in these domains increased as policies and laws aimed at improving African American access were done in these three areas. The health domain only saw a minor increase from 0.73 to 0.77 while criminal justice and poverty decreased. Criminal justice dropped from 0.28 to 0.27 and poverty from 0.52 to 0.44. The decreased domains are not surprising, as we have seen a significant growth in US incarceration as well as higher income inequality (Table 4.2).

60

K. O’Connell et al.

Table 4.2 National measuring the dream index and domain scores Criminal justice

Education

Health

Housing

Poverty

Rights

MTD index

1954

0.28

0.68

0.73

0.57

0.52

0.05

0.47

2014

0.27

0.76

0.77

0.75

0.44

0.84

0.64

4.6 Space and Equality It is essential to think about spatial inequality because a person’s geographic location can greatly affect their quality of life (Bell and Rubin 2007). Within counties, we see vastly different numbers. For example, Fulton County, Georgia (the main county associated with Atlanta) has a median income of $61,336 (Atlanta Regional Commission 2019a, p. 3), but the disparities from north to south are vast with the northern suburb of Johns Creek at a men income $113,609 (Atlanta Regional Commission 2019b, p. 15) while the southern neighborhood Vine City is $25,413 (Atlanta Regional Commission 2017, p. 2). Similarly, across Metropolitan Statistical Areas (MSAs), we see vastly different numbers. For example, based on American Community Survey 2016 data, Bridgeport, CT saw the highest income inequality with 95th percentile household income at 485,657 and the 20th percentile at 34,258. That is a ratio of 14:2 (Berube 2018). The current iteration of the MTD Index only looks at national data, so it does not speak to the spatial differences of inequality. While we plan to extend the MTD Index into smaller geographies, this project opens an opportunity for other organizations to think about local inequality in a historical context. Many of the indicators we have discussed are available at the sub-national level although at varying spatial scales and varying levels of access. For example, criminal justice indicators are readily available at the metropolitan statistical area (MSA) level. Depending on the jurisdiction, smaller-scale data will require data requests from local jails, courthouses, and police departments. On the other hand, all of the data dealing with voting rights is available by congressional district. For organizations interested in historical spatial data, it is important to understand boundary changes over time. Census tracts are relatively stable but some are redrawn every ten years (U.S. Census Bureau, n.d.). Cities are reshaped as sections are consolidated, merged, or annexed while MSAs typically grow over time as new counties are added. To account for these changes, the Census Bureau and some jurisdictions have geographic crosswalk files designed to allow researchers to compare geographies over time.

4 Measuring the Dream for an Equitable and Sustainable Future

61

4.7 Conclusion When thinking about Quality of Life Indicators, groups must look at racial and ethnic differences to ensure equitable outcomes for all. Working towards equality is not just a moral issue, but equal society can improve community cohesion, be more resilient, and work towards greater sustainability (Wilkinson et al. 2010). At the same time, inequality can challenge the democratic principles of the United States as “democracy might fail if income discrepancies and redistributive tensions between the different social groups become too large…. Ongoing technical change and economic development that affects different groups of society in different ways, as well as increasing inequality and polarization may potentially lead to a breakdown of the democratic equilibrium and to the (re)emergence of an oligarchy or a mass dictatorship” (Jung and Sunde 2014, p. 69). The MTD Index is a set of indicators that provides the scaling and longitudinal framework necessary to assess America’s realization of prosperity and equality. This data collection and analysis can increase awareness and direct assessments for residents, community organizers, and decision-makers. While the MTD Index focuses on disparities between Black and White populations, the principles and questions outlined in this paper can also be used for other groups. We recommend that indicator projects think about looking to disaggregate data and delve deeper into the Quality of Life between racial, ethnic, and other marginalized groups. Acknowledgments We would like to thank Open Society for the grant support of the Measuring the Dream from Brown to Black Lives Matter project.

References Atlanta Regional Commission. (2017). Neighborhood statistical area L01 fact sheet. http:// documents.atlantaregional.com/NN/Profiles/AtlantaProfiles/L01.pdf. Accessed July 29, 2019. Atlanta Regional Commission. (2019a). Fulton county fact sheet. http://documents.atlantaregional. com/Profiles/County/Fulton_NN.pdf. Accessed July 29, 2019. Atlanta Regional Commission. (2019b). Demographic profile: Johns Creek. http://documents. atlantaregional.com/Profiles/City/Johns_Creek_NN.pdf. Accessed July 29, 2019. Bell, J. E., & Rubin, V. (2007). Why place matters: Building a movement for healthy communities. PolicyLink. Berube, A. (2018). City and metropolitan income inequality data reveal ups and downs through 2016. Retrieved from https://www.brookings.edu/research/city-andmetropolitanincome-inequality-data-reveal-ups-and-downs-through-2016/. Bonczar, T. P. (2003). Prevalence of imprisonment in the U.S. population, 1974--2001. Retrieved from https://www.bjs.gov/index.cfm?ty=pbdetail&iid=836. Brooke, E. (2007). Bridging the divide: My life. New Brunswick, N.J.: Rutgers University Press. Chappell, M. (2010). The war on welfare: Family, poverty, and politics in modern America. Philadelphia: University of Pennsylvania Press. Centers for Disease Control and Prevention. (n.d.). Multiple chronic conditions. https://www.cdc. gov/chronicdisease/about/multiple-chronic.htm. Accessed March 31, 2019.

62

K. O’Connell et al.

Coleman, M. G. (2016). At a loss for words: Measuring racial inequality in America. The Review of Black Political Economy, 43(2), 177–192. Digest of Education Statistics, 2018. (n.d.). https://nces.ed.gov/programs/digest/d18/tables/dt18_ 502.30.asp. Accessed March 31, 2019. Eibach, R. P., & Ehrlinger, J. (2006). “Keep your eyes on the prize”: Reference points and racial differences in assessing progress toward equality. Personality and Social Psychology Bulletin, 32(1), 66–77. https://doi.org/10.1177/0146167205279585. Gundersen, C. (2013). Food insecurity is an ongoing national concern. Advances in Nutrition, 4(1), 36–41. https://doi.org/10.3945/an.112.003244. Hacker, A. (1992). Two nations: black and white, separate, hostile, and unequal. New York City: Simon & Schuste. Haney, C. (2003). The psychological impact of incarceration: Implications for post-prison adjustment. Prisoners once removed: The impact of incarceration and reentry on children, families, and communities, 33, 66. Hanushek, E. A. (2016). What matters for student achievement. Education Next, 16(2), 1–11. Herbold, H. (1994). Never a level playing field: Blacks and the GI bill. The Journal of Blacks in Higher Education, 6, 104. https://doi.org/10.2307/2962479. Hetey, R. C., & Eberhardt, J. L. (2018). The numbers don’t speak for themselves: Racial disparities and the persistence of inequality in the criminal justice system. Current Directions in Psychological Science, 27(3), 183–187. https://doi.org/10.1177/0963721418763931. Hetey, R. C., Monin, B., Maitreyi, A., & Eberhardt, J. L. (2016). Data for change: A statistical analysis of police stops, searches, handcuffings, and arrests in Oakland, Calif., 2013–2014. https://www.issuelab.org/resource/data-for-change-a-statistical-analysis-of-policestops-searches-handcuffings-and-arrests-in-oakland-calif-2013-2014.html Accessed September 20, 2017. Jones, C. P., LaVeist, T. A., & Lillie-Blanton, M. (1991). “Race” in the epidemiologic literature: An examination of the American Journal of Epidemiology, 1921–1990. American Journal of Epidemiology, 134(10), 1079–1084. https://doi.org/10.1093/oxfordjournals.aje.a116011. Jung, F., & Sunde, U. (2014). Income, inequality, and the stability of democracy—Another look at the Lipset hypothesis. European Journal of Political Economy, 35, 52–74. https://doi.org/10. 1016/j.ejpoleco.2014.03.004. King, M. L. (1963) “I Have a Dream.” Speech presented at the March on Washington for Jobs and Freedom, Washington, D.C. https://www.archives.gov/files/press/exhibits/dream-speech.pdf. Accessed July 14, 2015. Levy, H., & Meltzer, D. (2008). The impact of health insurance on health. Annual Review of Public Health, 29(1), 399–409. https://doi.org/10.1146/annurev.publhealth.28.021406.144042. Lowery, W. (2014). Paul Ryan, poverty, dog whistles, and electoral politics. The Washington Post. https://www.washingtonpost.com/news/the-fix/wp/2014/03/18/paul-ryan-poverty-dogwhistles-and-racism/?noredirect=on&utm_term=.bef5b477a28d. Accessed July 29, 2019. Marmot, M. (2005). Social determinants of health inequalities. The lancet, 365(9464), 1099–1104. Martin, E. (2017). Hidden consequences: The impact of incarceration on dependent children. National Institute of Justice, 278. https://nij.gov/journals/278/pages/impact-of-incarceration-ondependent-children.aspx. Accessed May 30, 2018. Mitchell, B., & Franco, J. (2018). HOLD “redlining” maps: The persistent structure of segregation and economic inequality. Retrieved from www.ncrc.org. National Research Council. (1995). Modernizing the U.S. Census. Washington, DC: The National Academies Press. https://doi.org/10.17226/4805. Pew Research Center. (2013). King’s dream remains an elusive goal; Many Americans see racial disparities. https://www.pewsocialtrends.org/2013/08/22/kings-dream-remains-an-elusive-goalmany-americans-see-racial-disparities/. Accessed January 15, 2019. Pew Research Center. (2016). On views of race and inequality, blacks and whites are worlds apart. https://www.pewsocialtrends.org/2016/06/27/on-views-of-race-and-inequalityblacks-and-whites-are-worlds-apart/. Accessed January 15, 2019.

4 Measuring the Dream for an Equitable and Sustainable Future

63

Pew Research Center. (2018). The public, the political system and American democracy. https://www.people-press.org/2018/04/26/9-the-responsibilities-of-citizenship/. Accessed July 30, 2019. Shipler, D. K. (1997). A country of strangers: Blacks and whites in America. New York City: Random House. Simpson, S. (n.d.). “The Great Poll Closure” is an initiative of The Leader-ship Conference Education Fund. http://civilrightsdocs.info/pdf/reports/2016/poll-closure-report-web.pdf. Accessed February 12, 2019. The Urban Institute. (n.d.). Nine Charts about Wealth Inequality in America (updated). http://apps. urban.org/features/wealth-inequality-charts. Accessed March 31, 2019. Thriving Cities Group. (n.d.). Indicator explorer. https://www.thrivingcitiesgroup.com/indicatorexplorer. Accessed March 31, 2019. Tucker-Drob, E. M. (2012). Preschools reduce early academic-achievement gaps. Psychological Science, 23(3), 310–319. https://doi.org/10.1177/0956797611426728. Uggen, C., Larson, R., & Shannon, S. (2016). 6 million lost voters: State-level estimates of felony disenfranchisement, 2016. https://www.sentencingproject.org/publications/6-million-lostvoters-state-level-estimates-felony-disenfranchisement-2016. Accessed 15 January 2019. United Nations. (1948). The universal declaration of human rights. http://www.un.org/en/ documents/udhr/. Accessed July 30, 2019. U.S. Census Bureau. (n.d.). Census tracts. https://www2.census.gov/geo/pdfs/education/ CensusTracts.pdf. Accessed July 31, 2019. Wilkinson, R. G., Pickett, K. E., & De Vogli, R. (2010). Equality, sustainability, and quality of life. BMJ (Clinical Research Ed.), 341, c5816. https://doi.org/10.1136/bmj.c5816. Wolfe, A. (1998). One nation, after all: What middle-class Americans really think about: God, country, family, racism, welfare, immigration, homosexuality, work, the right, the left, and each other. New York: Viking.

Katie O’Connell is a certified data nerd with a focus on community empowerment through data democratization. As project manager of the Communities Who Know (formerly Westside Communities Alliance) Data Dashboard, Ms. O’Connell developed an online platform that translated complex data to diverse audiences and communicated the relevance of data as a key decision-making tool. She has worked on other digital data projects including the International Human Trafficking Institute Dashboard, and the Neighborhood Quality of Life and Health Project. As part of a collaborative team, she researched and developed the Measuring the Dream Index, which tracked changes in racial equity between 1950 and 2014 across multiple measures. Ms. O’Connell is a doctoral student in the School of City and Regional Planning at the Georgia Institute of Technology. Her current research interests include historical dominant and counter-narratives around citywide development projects, the social injustices inherent in traditional data ownership structures, and tools to bridge the gap between quantitative and qualitative data. Ms. O’Connell has Bachelor degrees in Environmental Economics, Spanish, and Anthropology from the University of Georgia and a Masters in City and Regional Planning from the Georgia Institute of Technology.

64

K. O’Connell et al. Andrea Young is the executive director of the American Civil Liberties Union of Georgia and a life-long advocate for civil and human rights. Young was an Adjunct Professor at Morehouse College and Professor of Practice at the Andrew Young School of Policy Studies at Georgia State University where she taught courses on Leadership and Social Policy. For many years, she served as executive director at the Andrew J. Young Foundation producing a nationally-syndicated series of documentary films and other programs on themes of civil and human rights. She created the Making of Modern Atlanta Project that included a book, documentary film and archive. Young is the author of “Life Lessons My Mother Taught Me”; co-author of “Andrew Young and the Making of Modern Atlanta” and collaborated with former Atlanta Mayor Andrew Young in writing, editing and researching “An Easy Burden: Civil Rights and the Transformation of America.” Young has devoted her career to promoting policies to promote civil and human rights, including: the Martin Luther King Holiday Act; South Africa sanctions legislation; reproductive freedom; and universal pre-kindergarten. She is a graduate of Swarthmore College and the Georgetown University Law Center. Dr. Nisha D. Botchwey is an Associate Professor of City and Regional Planning at the Georgia Institute of Technology and Associate Dean of Academic Programs at Georgia Tech Professional Education, and an adjunct professor in Emory University’s School of Public Health. An expert in health and the built environment as well as community engagement, she holds graduate degrees in both urban planning and public health. Dr. Botchwey directs the Healthy Places Lab at Georgia Tech that includes the work of the Built Environment and Public Health Clearinghouse, the National Physical Activity Research Center, and data dashboards. Dr. Botchwey’s research focuses on health and the built environment, health equity, community engagement, and data dashboards for evidence-based planning and practice. She is co-author of Health Impact Assessment in the USA (2014), convener of a national expert panel on interdisciplinary workforce training between the public health and community design fields, and author of numerous articles, scientific presentations and workshops. Dr. Botchwey has won distinctions including an NSF ADVANCE Woman of Excellence Faculty Award, a Hesburgh Award Teaching Fellowship from Georgia Tech, the Georgia Power Professor of Excellence Award, a Rockefeller-Penn Fellowship from the University of Pennsylvania’s School of Nursing and was a Nominated Changemaker by the Obama White House’ Council on Women and Girls.

Chapter 5

Meaningful, Manageable, and Moveable: Lessons Learned from Building a Local Poverty Index Jamison Crawford and Frank Ridzi

Abstract Community indicators projects often involve efforts to “move the needle” with respect to local challenges related to community well-being. However, in charting their progress, many of these efforts are based on estimates that are untimely and characterized by large ranges of error (particularly when focusing on small geographic areas like neighborhoods). This can cause frustration and make it difficult for community groups to see meaningful change as a result of their efforts. In 2015, Syracuse, New York, was identified as having the worst concentrated poverty among Black and Latino residents, and among the worst for White residents, in the country. To address this challenge, a poverty index was constructed using locally collected data. In this chapter we recount how the index came to be and the infrastructure needed for its success. Key advantages of the index include that it is: (1) Meaningful, constructed with indicators of community conditions that residents can see in their everyday lives; (2) Manageable, by tracking local neighborhoods that are small enough for modest interventions to make a difference; and (3) Moveable, because an intervention can be tracked with timely counts rather than delayed estimates. We conclude with possible uses and future improvements.

27 March 2019, The Central New York Community Foundation. J. Crawford (B) The Central New York Community Foundation and The Graduate School at Georgia State University, 55 Park Place NE, Atlanta, GA 30303, USA e-mail: [email protected] F. Ridzi The Central New York Community Foundation and Le Moyne College Department of Anthropology, Criminology and Sociology, 1419 Salt Springs Rd., Syracuse, NY 13214, USA e-mail: [email protected] © Springer Nature Switzerland AG 2020 F. Ridzi et al. (eds.), Community Quality-of-Life Indicators, Community Quality-of-Life and Well-Being, https://doi.org/10.1007/978-3-030-48182-7_5

65

66

J. Crawford and F. Ridzi

5.1 Introduction and Motivations In 2015, a national study on high-poverty neighborhoods showed that, of the 100 largest metropolitan areas in the United States, Syracuse, New York had the highest concentrated poverty among African American and Latino residents. It was also in the top ten metropolitan areas for concentrated poverty among White residents (Jargowsky 2015). In response, several local interventions—such as Greater Syracuse H.O.P.E. and the Alliance for Economic Inclusion (AEI)—were initiated to address this issue. These calls to action have raised important questions as to the best way to decrease concentrated poverty and measure effectiveness. In short, Syracuse faced these problems: 1. We as a community wanted to reduce poverty, but we could not measure it, only estimate it. 2. We had many different initiatives happening at the same time, all related to different dimensions of poverty, but we had no way of seeing how they connected to a common goal. During this time, the Central New York Community Foundation (CNYCF) was engaged in several data-related initiatives, including the creation of CNY Vitals Pro, an online platform that visualizes community indicators, and a biweekly data user group, Community Data. Because of this, the CNYCF was well-positioned to work with community partners to create an index that measures poverty over time and across space. The Syracuse Poverty Index was created because it offers three unique advantages: 1. The index is comprised of a series of meaningful community conditions that reflect the realities that residents see in their everyday lives and that connect to the local efforts of community organizations and collective impact coalitions. This was very important because we wanted to involve residents in interventions and tracking progress. 2. The index shows data in census tracts because they are much smaller in size when compared to the previous norm of using ZIP codes. This makes the data more relatable to local residents and manageable when seeking to make change. This is important because change at the neighborhood level often cannot be seen across larger areas like cities and counties. 3. The index is sensitive enough, because of timeliness and accuracy, to be moveable. In other words, even modest interventions can show positive change in as little as a year or even sooner.1 This was important because existing coalitions and new anti-poverty efforts sought data-driven approaches both to steer implementation and monitor success.

1 As we discuss below, some aspects of the index, such as “Unemployed” and “Welfare” are updated

monthly.

5 Meaningful, Manageable, and Moveable: Lessons …

67

To understand how the index is an improvement on current and past indicators projects, we must understand the meaning of poverty, concentrated poverty, and high-poverty neighborhoods.

5.1.1 Defining Poverty, High-Poverty Neighborhoods, and Concentrated Poverty The U.S. Census Bureau officially defined poverty in 1964 as three times the cost of a “minimal diet” (U.S. Census Bureau 2014a, b). Today, set dollar amounts per number of household members are used to define the poverty line, which changes with inflation. If a family earns less than this set dollar amount, everyone in it is considered in poverty (Fontenot 2018). For instance, using the Poverty Thresholds for 2018 by Size of Family and Number of Related Children Under 18 Years, a family of four people (with 2 children) would be considered in poverty if its members collectively earned less than $25,465. Similarly, a family of six people (with four children) would be below the poverty line if they earned less than $33,553 (U.S. Census Bureau 2019a, b, c, d, e). Defining “high-poverty neighborhoods” is important to understanding concentrated poverty. Most research on high-poverty neighborhoods use “census tracts”, which are small, geographic areas created by the U.S. Census Bureau. Official “neighborhood boundaries” don’t work well because they are defined differently in different cities. However, every city in the United States is captured by a group of census tracts, which is determined by population density. Hence, the denser the place, the smaller the area covered by the census tract. Most research defines census tracts as “highpoverty neighborhoods” when over 40% of residents in the tract also live below the poverty line (Jargowsky 2013). Because of this, we use the terms “census tract” and “neighborhood” interchangeably. Related to the above definitions, in any given area, like a county or city, concentrated poverty is the percent of the population that lives both (1) below the poverty line, and (2) in a high-poverty area (Jargowsky 2013). In Syracuse, because the biggest proportion of Black and Latino residents live both below the poverty line and in high-poverty neighborhoods, it has the worst concentrated poverty for residents of these ethnicities (Jargowsky 2015).

5.1.2 Understanding Poverty Across Space Poverty has only been studied over geographic areas since the late 1980s, when a landmark study, The Truly Disadvantaged, researched the effects of concentrated poverty. These effects, called “concentration effects”, cause intergenerational poverty, neighborhood crime, poor labor force attachment, and other challenges (Wilson 1987). Since this study, much more research on poverty has included geography in analyses.

68

J. Crawford and F. Ridzi

“Concentration effects”, the effects of concentrated poverty, also include more single-parent households, unemployment, financial hardship, wage disparities, lack of reliable transportation, and limited access to information about jobs. Schools and academic outcomes are hurt by concentration effects, like unsustainable commutes, lower expectations by teachers, fewer resources, more disruptive peers, and mocking academic success (Jargowsky 2013). Health outcomes are also hurt by concentrated poverty due to fewer parks, less greenspace and recreation, increased exposure to alcohol and tobacco, and higher rates of both communicable and preventable diseases, like diabetes, obesity, and premature births (Jargowsky 2013). Children are most affected by concentrated poverty and are even more likely to live in high-poverty neighborhoods than adults (Jargowsky 2013). Taken together, these aspects of concentrated poverty can be seen to produce a vicious cycle in which each disadvantage reinforces and magnifies another. In Syracuse, as in other locations, the hope is to reverse these trends and shift toward a virtuous cycle in which positive outcomes in these various areas foster improved outcomes among one another. Because concentrated poverty and concentration effects (whether positive or negative) are measured using census tracts, it is important that our index measures poverty with the same unit of analysis.

5.2 Creating a Local Measure of Poverty The debate over the ideal poverty measure has waged since the initial measure emerged in 1964. In 2010, a technical working group from various federal agencies created the Supplemental Poverty Measure (SPM), which estimates indicators other than gross income, like taxes, medical and work expenses, and “noncash resources” like temporary assistance and housing subsidies. Like most research on concentrated poverty, the SPM also accounts for location (Fontenot 2018). The Syracuse Poverty Index synthesizes and improves on these approaches by removing estimation and tailoring it to fit Syracuse. It is a collection of key indicators designed to measure conditions that cause and are caused by concentrated poverty. With the support of community partnerships, the index is updated using headcounts, not estimates, as soon as data are available. However, the index took time to develop. The first prototype of the index came about in an organic, community-driven way before becoming an independent tool (as we will explore in the following section). Since then, we added many indicators and experimented with different ways to make them more meaningful. Lastly, we secured a basic infrastructure to make the index sustainable by partnering with data providers and, for some indicators, making financial investments.

5 Meaningful, Manageable, and Moveable: Lessons …

69

5.2.1 Origins in the Community In 2017, several interventions to fight concentrated poverty in Syracuse were in development. Concurrently, the Central New York Community Foundation was hosting a series of data user, or “Community Data” meetings, to improve networking, data sharing, and data-driven collaboration. These efforts often involved discussion, debate, presentations, and demonstrations. Though meetings usually focused on data, attendees often had diverse backgrounds, skill sets, and levels of data literacy. At our meetings, it was not unusual for ideas to change hands between fundraising professionals, program implementers, analysts, and researchers from the social sectors. Now and then, individuals or organizations with a data challenge were referred to Community Data meetings to share ideas or request professional volunteers. Later that year, Community Data was approached by an executive committee member of the newly-launched Greater Syracuse H.O.P.E. initiative, part of the Upstate Revitalization efforts to fight concentrated poverty. H.O.P.E. had selected a set of indicators to use in their data strategy. When the committee member presented this strategy and asked for help in researching and collecting the data needed, several members of the Community Data group volunteered. At first, data collection was organic. In helping Greater Syracuse H.O.P.E., we collected data for many indicators that H.O.P.E. had culled from a series of community meetings that included both local leaders and residents. In these meetings, participants voted with sticker dots to prioritize existing data measures and suggest new ones. Importantly, these indicators were also vetted by substantive experts from public and nonprofit sector organizations in Syracuse, New York and Onondaga County. As we thought about building out the index, we also thought about which indicators would be most meaningful to the communities we serve. We considered what was needed for a successful and sustainable index. Namely, (1) How to get true headcounts, not estimates, as soon as they are available, and (2) How to make sure the data behind the index flowed reliably and consistently. Over the next month, we collected the data and documentation available for the desired H.O.P.E. indicators. The data were stored online for attendees and others to access at any time. We also learned from attendees specializing in certain areas that many indicators were not practical, available, or even measurable. Soon after, an analyst for the Syracuse City School District used the H.O.P.E. indicators to create a prototype index and introduced it at the next Community Data meeting. We soon realized that building out the index could create a measure that can be meaningful, manageable, and moveable.

5.2.2 Meaningful Components Poverty often seems abstract, misunderstood, or immeasurable and definitions of poverty vary all over the world. The Syracuse Poverty Index attempts to measure meaningful changes in poverty that affect the everyday lives of residents. In this way,

70

J. Crawford and F. Ridzi

it is much like the U.S. Census Bureau’s Supplemental Poverty Measure (SPM), as it measures poverty using more than just income (Fontenote 2018). Like the H.O.P.E. indicators, the index measures poverty with “proxy” indicators that are associated with poverty. Because our first proxy indicators for the index were chosen by the executive committee of Greater Syracuse H.O.P.E., a group of community members that understand the challenges facing Syracuse, they were communitydriven, meaningful, and designed to eliminate concentrated poverty. Over time, more indicators were carefully added for a range of areas. Some of these areas include crime, health, education, and the economy. The first indicators added were inspired by Maslow’s Hierarchy of Needs (Maslow 1943). Since concentrated poverty is most harmful to children, indicators like”Reading Below Grade Level” focus on child development (Wilson 1987; Jargowsky 2013). These “sociodemographic, health care, family, and community attributes” deeply affect children and are associated with Mental, Behavioral, and Developmental Disorders, or MBDDs (Bitsko et al. 2016). Because these indicators are community-driven, they are meaningful to measuring interventions and reflect real community issues facing residents in their everyday lives. They include: Healthy Homes The maximum percent of children with elevated blood lead levels, age 6 or younger, per census tract. Residents who live in poverty often cannot afford housing in newer or safer neighborhoods. Instead, they may be forced to live in substandard housing with environmental hazards and poisons like lead and mercury. The County Health Department releases these data by census tract in percent ranges (i.e. 0.1–4.7%, 4.8–7.3%, 7.4–12.5%, 12.6–20.0%, 20.1–30.0%, and greater than 30%). Since we do not know the actual percentage, we use the maximum of each range as the level of poisoning that may exist in a census tract (OCHD 2019). Children are at the greatest risk for lead poisoning if both living below the poverty line and in older housing (Center for Disease Control 2013)—conditions that are very common in high-poverty neighborhoods (Vivier 2010). Early childhood exposure to lead is associated with developmental delays in perceptual reasoning, memory issues, and downward mobility (Reuben 2017). “Child Lead Poisoning” is updated annually. Crime The percent of reported major crimes against the person and property (i.e. Part I crimes without larceny), for every 100 residents. Part I crimes include criminal homicide, robbery, forcible rape, aggravated assault, burglary, motor vehicle theft, and arson (Federal Bureau of Investigation 2004). Living in poverty often means that residents can only afford to live in high-crime neighborhoods (Desmond 2016). Every day, residents living in poverty face dangerous and stressful situations and must guard their lives and property. As a proxy, the index uses “Crime”. Living in high-crime neighborhoods is associated with poorer mental and physical health, depression, lower quality of life, poor academic performance, limited trust, poor selfesteem, and impaired development (Stafford 2006; Violence Policy Center 2017). Crime rates are updated every other month using the Syracuse Post Standard Crime Database (Crawford 2018a, b; Syracuse.com 2019).

5 Meaningful, Manageable, and Moveable: Lessons …

71

Education The percent of children reading below grade level, according to the 3rd grade English Language Arts (ELA) exam, per every ten children that took the exam. Education is often the best way out of poverty. By the third grade, children are expected to transition from learning how to read to reading in order to learn (Center for Public Education et al. 2015). As a proxy, the index uses “Reading Below Grade Level” to measure this. Children residing in high-poverty neighborhoods are at more risk of failure (Ludwig 2001), less educational attainment, and less adult earnings (Quillian 2017). Children who move to better areas are more likely to attend college (Chetty 2016). Most importantly, grade-level reading is a strong enough proxy to use as a “make-or-break” benchmark for predicting a child’s success in life (Annie E. Casey Foundation 2010). This indicator is updated annually by the Syracuse City School District. Employment Opportunity The percent difference below the average quarterly wage in Onondaga County. While some argue that education is the best way out of poverty, others believe a job would be better (Schramm 2013; Croom et al. 2019). However, many residents living in poverty do not have access to well-paying job opportunities. As a proxy, the index compares the average wages paid by employers in each neighborhood to the average county wage, or “Mean Wages vs. County”. Low wages give workers limited prospects for upwards, intragenerational mobility while their families have fewer prospects for intergenerational mobility (Olinsky and Post 2013). Average wages are updated quarterly by the New York State Department of Labor by special agreement, as wages are not accessible at the neighborhood level for the general public (Bureau of Labor Statistics 2019). Unemployment Unemployment insurance recipients as a percent of the working population, ages 16–64. Poverty goes hand-in-hand with poor financial stability, causing negative mental, emotional, and physical effects. As a proxy, the index scores the “Unemployed” in each neighborhood. Residents living in high-poverty neighborhoods have less labor force attachment, fewer working role models for children, and less information about jobs—in effect, causing increased likelihood of suffering from labor market insulation (Wilson 1987). Unemployment “as way of life” hurts resident employability (Elliot 1999). The unemployment rate is also updated monthly by the New York State Department of Labor by special agreement. Welfare Dependence Temporary assistance cases as a proportion of the working population, ages 16–64. Welfare is a means-tested program, meaning that individuals and families must document that both their income and assets are below a certain amount. As a result, the proportion of people in a neighborhood who qualify for welfare is an excellent measure of poverty. However, not everyone applies who is eligible, which does introduce a degree of undercount. When poverty is intergenerational, or when the children of impoverished parents also grow up in poverty, relying on welfare can make it harder to escape poverty. This is often because there is less

72

J. Crawford and F. Ridzi

access to support networks and role models to help residents build careers. Some residents may feel trapped by a dependency on welfare, known as the “welfare cliff”2 . For others, it is simply more logical to remain on welfare.3 “Welfare” is updated monthly by the County Department of Social Services. Neighborhood Vitality Property parcels with at least one vacancy as a percent of all property parcels. One important aspect of poverty is the neighborhood environment. High-poverty neighborhoods are landscapes of despair. There are many schools of thought that support this idea, including the “Broken Windows” theory (Kelling and Wilson 1982), or how residents give up on their unkempt neighborhoods, as well as “Cues to Care” theory, or how active residents can improve safety by better caring for their neighborhoods (Troy et al. 2016). Buildings in high-poverty neighborhoods depreciate and deteriorate more rapidly due to a higher number of residents in a smaller area as well as property crime. As the poverty rate increases, housing values decrease (Galster 2006). Homeowners and landlords maintain their properties to the extent that they will remain healthy assets with enough revenue streams. As poverty increases and quality of life decreases, the chance of attenuated maintenance, disinvestment, and even abandonment rises at an accelerating rate (Galster 2006). “Vacancies” are updated each quarter by the City of Syracuse’s Department of Neighborhood and Business Development.

5.2.3 Manageable The Syracuse Poverty Index is manageable because it measures change at the neighborhood level. Remember that these neighborhoods are actually “census tracts”, or small areas of land defined by the U.S. Census Bureau. Because of this, the index takes advantage of this infrastructure, but it is not limited by the challenges that come with U.S. Census Bureau data. Granular and Standardized Neighborhoods, or census tracts, are small. Social issues are often called “wicked problems”, because they are complicated, interrelated with other issues, and difficult to measure with detail (Head 2008). These problems are sometimes called “social messes”. What wicked problems and social messes have in common is the difficulty with homing in on their individual parts (Ackoff 1974). By using census tracts, which are smaller than ZIP codes and Syracuse’s official neighborhoods, we can zoom in on areas of the city that are much smaller and more 2 If

a new job or promotion ends welfare eligibility, the net loss in benefits and increase in income taxes may create effective marginal tax rates of 50–60% (Dorfman 2016; Congressional Budget Office 2015). Despite evidence of a “welfare cliff”, only a few recipients approach this drop-off (Adolphsen 2018). 3 There are multiplicative effects of “welfare culture”. For example, the high cost of childcare makes permanent dependency on welfare a more rational choice for many in poverty (Kimenyi 2008).

5 Meaningful, Manageable, and Moveable: Lessons …

73

Fig. 5.1 Size comparison of ZIP codes, TNT sectors, Neighborhoods, and Census tracts—Census tracts are much smaller and more detailed than ZIP codes or official neighborhoods. Source U.S. Census Bureau (2019a, b, c, d, e), Syracuse Open Data (2015, 2019)

detailed4 . This allows community organizations and interventions to use location intelligence for better coordination, use of resources, and even identifying areas for collaboration with other organizations. Figure 5.1 shows that census tracts are the smallest and most manageable of these units.

4 In

Syracuse, for example, city officials defined 32 neighborhoods in 2010, while Tomorrow’s Neighborhoods Today (TNT) identifies 8 sectors (Tomorrow’s Neighborhoods Today 2019). However, Syracuse has 55 census tracts, providing significantly more spatial detail to leverage location intelligence, shared measures and outcomes, and increased efficacy.

74

J. Crawford and F. Ridzi

Census tracts are not only more granular but also standardized. In the U.S., they use the same rules and have a similar population size, around 4,000 residents, all over the country (U.S. Census Bureau 2019a, b, c, d, e).5 Since they are so similar, they are easy to compare and, in addition, they are already mapped by the U.S. Census Bureau. By using the same units for measuring both success and need, the index allows organizations to use shared metrics. Under the collective impact model, the index provides a “common agenda” and “shared measurement system” for “mutuallyreinforcing activities”. Because the index uses standardized units (i.e. census tracts) and is available for all, organizations can move from “isolated impact” to “collective impact”, focusing collaborations on one neighborhood at a time (Kania and Kramer 2011).

5.2.4 Moveable: The Change You Can See The index is meaningful for residents and manageable for organizations because of the small areas it uses. However, it must also be moveable or sensitive to change (as in the common community goal of “moving the needle”). In other words, the index must be timely and accurate enough to show real change that results from community interventions. When community members learned that concentrated poverty was the worst in Syracuse among Black and Latino residents, anti-poverty organizations like H.O.P.E. and the Alliance for Economic Inclusion (AEI) were created in response to this discovery. This index was developed to assist such efforts by measuring the changes these organizations are able to bring about. Supporting Existing Interventions Many coalitions and collective impact efforts existed before learning about Syracuse’s extreme concentrated poverty. These efforts often focus on shared or overlapping goals. Because of this, the index may be used for many interventions that already exist. For example, the Green and Healthy Homes Initiative found that measuring child lead poisoning was an important community need. Anti-crime coalitions like the Trauma Response Team and the Street Addiction Institute were made to stop “Crime”. Both the Literacy Coalition of Onondaga County and Early Childhood Alliance focus on “Reading Below Grade Level”. The Work Train Collaborative, a group of funders and service providers, focuses on reducing the “Unemployed”. These interventions may use one or more indicators, or the whole index, as shared, moveable metrics. 5 Census tracts usually have a population size ranging from 1,200 to 8,000 people, with an optimum

of 4,000 people. According to the U.S. Census, a Census tract is “A small, relatively permanent statistical subdivision of a county delineated by a local committee of census data users for the purpose of presenting data. Census tract boundaries normally follow visible features, but may follow governmental unit boundaries and other non-visible features in some instances; they always nest within counties. Designed to be relatively homogeneous units with respect to population characteristics, economic status, and living conditions at the time of establishment, census tracts average about 4,000 inhabitants. They may be split by any sub-county geographic entity.” (https://factfinder. census.gov/help/en/census_tract.htm).

5 Meaningful, Manageable, and Moveable: Lessons …

75

Interrelated Indicators The diversity of social issues that led community leaders to make these organizations and coalitions also show how the proxy indicators used in our index are both diverse and interconnected. In other words, these interventions become complementary, or “mutually-reinforcing activities”, a key ingredient for collective impact (Kania and Kramer 2011). Remember that a “social mess” or “wicked problem” happens when multiple problems are interrelated (Ackoff 1974). In theory, “moving the needle” for one social issue may indirectly move the needle for another social issue. For example, less “Child Lead Poisoning” may help “Reading Below Grade Level” because lead poisoning in children is associated with worse learning and memory (Geier et al. 2017). What is more, less “Child Lead Poisoning” may help lower “Crime” since childhood lead exposure is associated with delinquency (Gump et al. 2017). Because of the relation between these indicators, we may glimpse the future of “Reading Below Grade Level” and “Crime” in our community by looking at the “Child Lead Poisoning” component of the index. Measuring Collective Impact Since each coalition focuses on one or more of the indicators in the index, the whole index may be used to measure the success of collaborative interventions holistically and in a community-wide context. Instead of reminding us of how complicated “wicked problems” can be, the index may be used as both a shared touchstone and a focal point for coordinated efforts across community organizations and coalitions. With a moveable index, we can better understand how the works of one organization can affect the works of another, and how those organizations may work even better as a team. In cases like these, the index may be used for more than simply measuring success; it can be used to strategize and coordinate interventions by many organizations to reinforce and optimize their impact. Though still nascent, the index has shown promise in aligning such disparate things as job training, banking, and adult education. For instance, it is now possible to observe whether the collective inputs in a particular census tract—like banks providing more loans to minority- and womenowned businesses, job training agencies preparing more clients, and adult education providers serving more learners—will have a cumulative impact on such things as wages, unemployment, and public assistance. These actions have historically been linked in theory. Now we will be able to see if they collectively move the index (and its component parts) in a targeted neighborhood. Before the index, it was common for collaborations and coalitions to overlap, but not reinforce, their interventions. In other words, these interventions were a kind of “isolated impact”. Because the index is meaningful and manageable, we can better understand the relationship between interventions, organizations, and the indicators that drive them. We gain clarity because the whole of the index is greater than the sum of its parts. This clarity allows us to find and develop more strategic collaborations and move from “isolated” to “collective impact”. Quick Wins and Momentum It is important that the index is moveable because of the need for “quick wins”. Using Results Based Accountability’s (Friedman 2005) differentiation between population-level indicators and performance measures helps

76

J. Crawford and F. Ridzi

to articulate how this might work. If we see multiple organizations having performance measure-level success in a given census tract and in related focus areas, we would anticipate that these changes would eventually result in a population-level change in that census tract. Though we will not be able to connect changes and interventions for certain (unless we are able to collect client-level data across organizations), we have a greater clarity about how we can geographically focus our coordinated efforts (i.e. which census tracts) and where to look for population-level changes. This clarity can expedite our chances of finding evidence of populationlevel changes and can provide the momentum that such efforts will need to sustain them. Without seeing positive change, it is common for community organizations to give up or become cynical about the ability to change. Therefore, short-term “wins” are very important in the early stages of an intervention. Because data from the U.S. Census Bureau is not timely, it is easy to lose faith in an intervention with poor or missing evidence. Because the index is meaningful, manageable, and moveable, these “wins” may appear within a year or even sooner. What is most important is that these wins are clear and unambiguous, “not just a judgment call that can be discounted by those opposing change” (Kotter 1995). This allows users of the index to make useful and informed mid-course corrections as well as identify “wins” with less ambiguity.

5.3 Discussion: Lessons Learned Though still a work in progress, the Syracuse Poverty Index has yielded a set of lessons learned. These include lessons related to data collection and infrastructure, index development, and balancing both privacy and transparency.

5.3.1 Data Collection and Necessary Infrastructure Accuracy, Reliability, and Timeliness Data from the U.S. Census Bureau has many limitations. The decennial census tries to “count everyone once”, but “once” happens every ten years (U.S. Census Bureau 2019a, b, c, d, e). Another option, the American Community Survey (ACS), provides rolling estimates over 1, 3, and 5 years. Oneyear estimates are more up-to-date but less accurate, while five-year estimates are less up-to-date but more accurate (U.S. Census Bureau 2019a, b, c, d, e).6 Fiveyear estimates are still helpful to fine-tune the index (Jargowsky 2013). However, using five-year estimates, which often have large estimation errors, is unacceptable 6 Importantly,

1-year estimates are only given for areas with more than 20,000 residents and 3-year estimates are only given for areas with over 65,000 residents. Remember, the index uses census tracts, and the average census tract has 4,300 residents, and so these data are not even available (U.S. Census Bureau 2019a, b, c, d, e).

5 Meaningful, Manageable, and Moveable: Lessons …

77

if community organizations want to be responsive to a moveable indicator. Because of this, the index must be accurate and up-to-date to be truly moveable. To address this, we found it imperative to forge the partnerships necessary to compile the index by using headcounts rather than estimates. Existing Databases and Portals Syracuse is the fortunate beneficiary of an increasing number of open data portals and other online databases which any resident can access. These include CNY Vitals, CNY Vitals Pro, Syracuse Open Data, and HealtheCNY. Data collection from these sources is straightforward because they are generally convenient, accessible, reliable, and do not require partnership with organizations. These sources often have documentation to better understand their data, which can be an advantage over other sources, including community partners that do not document data well or at all. Even with open data sources, additional work is needed, like documenting how data are collected and changed. A 2016 study of 467 nonprofit professionals found that even though 90% say their organization collects data, 49% say they “either don’t know or weren’t sure about all the ways their organization was collecting data” (Every Action and Nonprofit Hub 2016). For example, “Crime”, one indicator in our index, is updated each month by the Syracuse Police Department and collected from Syracuse.com (2019). To be used, several steps are needed to process the data and prepare them for the index, such as converting blocklevel addresses to census tracts and other tasks. We document and publish each task so they can be reproduced by anyone in the same way (Crawford 2018a, b). Community Partners Most indicators used are made up of data provided by local community partners. Compared to open data portals and online databases, community partners are often the next best source for accurate and timely data. For data sources, three key ingredients are needed: (1) Trust, (2) A data sharing agreement, and (3) Technology, data maturity, or both (Cooper and Shumate 2015; U.S. Department of Education 2016). Data for most social sector organizations can be very sensitive, especially data on the vulnerable residents they serve. Protecting the identity of residents is critical for public trust and reputation, but also has important legal constraints. For example, more steps in sharing data are required of publicly-funded schools, colleges, and universities under the Family Educational Rights and Privacy Act (FERPA) and of public health organizations under the Health Insurance Portability and Accountability Act (HIPAA). For our purposes, we found that sharing data at the census tract level was the most expedient since it both served to protect confidentiality and promised data on a granular enough level to be meaningful for planning and monitoring interventions. Financial Investment When online databases and community partners do not have the necessary data, financial investment may be needed. In the private sector, contractual models like Service Level Agreements (SLAs) are used for more detailed arrangements. These details may involve the quality, availability, and schedule of

78

J. Crawford and F. Ridzi

regular updates and are enforceable with “obligation policies” (Swarup et al. 2006). The data for our index’s “Unemployed” and “Mean Wages vs. County” indicators are not provided by the private sector. Instead, we have an SLA with the New York State Department of Labor, or a contractual agreement where we pay for these indicators to be processed.

5.3.2 Index Creation Modular Components The index is made up of many indicators and each indicator has their own score, sources, raw data, and local experts. This allows any organization to focus on one or more indicators, and only those indicators that are most important to the community and/or their mission. In this way, organizations can use this shared system of measurement for grantwriting, needs assessments, research, program design, monitoring, evaluation, and other tasks to bring about positive change with respect to a specific social issue.7 Furthermore, the index is chronologically modular. This means that organizations using the index may focus on a specific year, quarter, month, or other range of time. This allows them to study past interactions between indicators and other ways one or more neighborhoods are affected by external events that the index does not measure. This may be seen in Fig. 5.2. Lastly, the index is geographically modular. In other words, it can be separated by neighborhoods. This allows community organizations to focus on a single neighborhood or a cluster of specific ones. These organizations can then use the index to see all scores, all indicators, and all current and historical data for a specific area. This modularity is demonstrated in Fig. 5.3. In short, the index may be disassembled and reassembled like a pocket watch. Unlike a pocket watch, however, we can remove a cog or a spring to observe it, but everything keeps ticking. This modularity lets organizations drill down to a specific indicator in a specific neighborhood during a specific time and expect accurate and reliable insights. At the same time, the overall index is useful to community members and policy makers as a single score that can be easily monitored and compared over time and across geographies. An Absolute Zero Poverty can be measured in financial terms but it is much harder to measure poverty in terms of welfare, unemployment, building vacancies, gradelevel reading, and other index indicators. In earlier prototypes of this index we used relative scores that divided each of Syracuse’s 55 census tracts into quintiles. While 7 The ability to “parse”, or separate, the index by its components is especially valuable to statisticians,

analysts, and data scientists. This is because of “Simpson’s Paradox”, a phenomenon where unique trends disappear when data are aggregated or grouped (Simpson 1951). Without the modularity of the index, trends may remain hidden or indiscernible.

5 Meaningful, Manageable, and Moveable: Lessons …

79

Fig. 5.2 Demonstration of index resolution over time—Poverty as measured by the U.S. Census Bureau 5-year estimates versus poverty measured by the Poverty Index. The six months represented by the Poverty Index, right, fits into the grey area of the U.S. Census data, left (CNY Vitals Pro 2019)

this created an attractive scale, it has a fatal flaw in that the index was completely relative and pitted these neighborhoods against one another. Because the scores were based on quintiles, for one neighborhood to improve, another neighborhood had to worsen. In other words, a significant positive change in one neighborhood would mean that another neighborhood has fallen off the leaderboard. This is true even if the neighborhood that fell in “rank” did not actually become worse. What is more, this “downgraded” neighborhood may have improved, but it did not improve as much as the supplanting neighborhood. This is deceptive and demoralizing, especially for community organizations. Now, instead of ranking neighborhoods, our index uses an “absolute zero” for each indicator. This is possible by converting each indicator into a percentage by using raw counts. For example, “Vacancies” represent the percent of properties with vacant units out of all properties in a neighborhood. This allows the indicator score to reach 0% in every neighborhood in the city, which is the long term goal for both the components of the index and the index as a whole. An index which force-ranks neighborhoods will rise and fall at the expense of other neighborhoods, but an index with an absolute zero makes any positive change a rising tide that lifts all boats. Creating Scales Each indicator is unique and tells us about a different side of poverty. Because they are unique, each indicator is also measured using different kinds of units. For example, the unit for “Unemployed” is an adult resident; for “Mean Wages vs. County”, it’s dollars; for “Child Lead Poisoning”, it’s children under the age of 6; for “Vacancies”, it’s properties; for “Reading Below Grade Level”, it’s third-graders. We do not know

Fig. 5.3 Demonstration of index modularity—The poverty index may be broken down into individual components and compared across specific neighborhoods. Above, three neighborhoods are compared across six unique indicators

80 J. Crawford and F. Ridzi

5 Meaningful, Manageable, and Moveable: Lessons …

81

how to convert dollars into third-graders. Therefore, a meaningful index must find a way to measure and visualize these unique indicators in a way that uses a common scale. To overcome this challenge, we transform the scale of each indicator into percentages that represent their own index score. These percentages may be used on a common scale even with different units of measurement.8 We find these percentages by taking the raw count in a neighborhood, like unemployed adults or vacant properties, and we divide them by the most appropriate population for the indicator. For example, “Unemployed” is the percent of unemployed adults out of the total workforce population, age 16–64. “Vacancies”, on the other hand, are represented by the percent of properties with at least one vacant unit out of the total number of properties in the neighborhood. This is not always as easy as it may seem. For example, “Mean Wages vs. County” is used to measure the wages that employers pay in each neighborhood. To make this into a percentage, we calculate the percent difference between wages paid in Onondaga County and wages paid in each neighborhood. If the percent is negative, it means that average wages are better in that neighborhood than all of Onondaga County, and it does not add points to the neighborhood’s index score. If the percent is positive, it tells us to what extent wages are lower in that neighborhood, when compared to the county, and adds points to the neighborhood’s index score. In other cases, indicators may be affected by outliers, or values that are unusually high or low in some neighborhoods. This was the case with “Crime”. At first, “Crime” included all “Part I” crimes, which include many violent crimes and some nonviolent ones like larceny. Because of this, the neighborhood that has Destiny USA, Syracuse’s super-regional shopping mall, looks like a hotbed for potentially violent crime. During some months, this neighborhood had crime percentages that were more than 300%. In other words, for every one resident in the neighborhood, three crimes are committed. To make this indicator more meaningful, we divided “Crime” into two new indicators, larceny and all other crimes (i.e. “Crime”). As a result, we can tell when and where violent crimes are the worst in the city without larceny complicating our understanding. Two community partners provide data in a very different format. This is not a coincidence. The indicators, “Reading Below Grade Level” and “Child Lead Poisoning”, are protected under FERPA and HIPAA privacy laws, respectively. In order to protect the privacy of these residents, the data for these indicators are converted into percentages before we receive them. While this is less trouble for us, it is not ideal. When partners convert these data into percentages before they share them, we do not know the true numerator (the number of children affected by the issue) nor the true denominator (the total number of children in the neighborhood). The downside of this is that two different neighborhoods may both show, for example, that 90% of 8 Normalization,

or standardization, of variables with different units of measurement may be more statistically accurate by using z-scores, standard deviations, and other methods which we may explore in the future. However, the index is intended for audiences of mixed proficiencies and, therefore, interpretability is prioritized.

82

J. Crawford and F. Ridzi

third-graders are reading below grade level. However, one neighborhood may have 30 third-graders, and the other may have 100. To address this problem, groups that are seeking to work in a particular neighborhood are encouraged to reach out to the organization that provides the data in order to ascertain whether there is a large enough population to warrant taking action. Lastly, even when indicators were converted into percentages and a common scale, some indicators may overshadow others. For example, “Reading Below Grade Level” is typically between 80 and 90% for most neighborhoods in Syracuse. That is, in most neighborhoods, 80–90% of third-graders are reading below grade level. This is a very large score compared to other indicators. This is a significant problem because it is of such a magnitude that it visually dwarfs other indicators within the index and often makes them undetectable. This is a critical issue since data visualization is the main vehicle for identifying patterns and tracking progress in the index. To avoid this, we convert “Reading Below Grade Level” to a scale that is commonly used in the community by reporting it per every 10 children instead of every 100. This mirrors common community discussions in which we say 9 out of 10 children are not reading at grade level. These conversions are documented in detail using the online platform GitHub (Crawford 2018a, b). To summarize, using a common scale is advantageous when possible but a truly meaningful index must also score these indicators in a way that they do not overshadow important sides of poverty and the many missions of community organizations to combat them.

5.3.3 Transparency and Privacy Private Nonprofit and public sector organizations must protect the privacy of the residents they serve. This may be for ethical reasons, a competitive advantage, good faith and reputation, or legal requirements like FERPA and HIPAA. Because of these concerns, organizations cannot publish a map that pinpoints where their clients live. Instead, they can use geographic areas, like neighborhoods, to share data about its residents without the details of any one individual. This process is called “anonymizing” data because it allows residents to remain anonymous.9 If we cannot protect client information, we cannot share it. Therefore, it is important to use areas that are big enough to remain anonymous and small enough to be manageable.10 This project has

9 Another process called “geocoding” allows data users to convert client addresses into census tracts

and other units. Free tools from the U.S. Census Bureau allow anyone to convert addresses into census tracts quickly and easily (US Census Bureau 2019a, b, c, d, e). 10 There is a smaller geographic area called “Block Groups”, or groups of city blocks, used by the U.S. Census Bureau. These areas may contain between 600 and 3,000 residents, and each census tract contains at least one “Block Group” (U.S. Census Bureau 2019a, b, c, d, e). However, in our experience, these areas are too small too hide the identity of its residents. In such cases, the data are “suppressed” or hidden from the public.

5 Meaningful, Manageable, and Moveable: Lessons …

83

taught us that census tracts seem to be a workable solution to maintaining privacy while sharing data that is specific enough to guide action. Reproducible and Open Source Many social sector tools do not show their users what is under the hood. Instead, data pass through a “black box” and emerge as “insights”. We attempt to buck this trend by making the index reproducible. What raw data we share may be accessed by the public. In addition, our community partners and other data sources may be contacted or accessed directly by anyone. Finally, any steps in data processing are documented and may be accessed online. To facilitate this, we have created a publically available “code book” that provides indicator definitions and formulas (Crawford 2018a, b). These are key ingredients to reproducible research, a standard in scientific research that is designed to “leave finished work in a state that allow[s] colleagues to reproduce the calculation, analysis, and final figures” (Ellis and Leek 2017; Barba 2018). In this way, we welcome scrutiny and ideas, encourage others to expand on our work, and make sure the index is owned and operated by the community.

5.4 Future Improvements This is the first version of the index, or any index like it, to our knowledge. While we believe the above is a strong case for an index which is meaningful, manageable, and moveable, we plan to improve the tool in many ways and invite others to explore these possible improvements. The indicators that make up the index were both communityand coalition-driven. However, there are still indicators that would be meaningful to many organizations and important to measuring poverty in different ways. For example, we understand that upwards mobility is strongly associated with commute time, which is generally longer in segregated, high-poverty neighborhoods (Chetty et al. 2014). What is more, not all meaningful indicators are feasible for any index. As another example, the Greater Syracuse H.O.P.E. executive committee believed that adult literacy rates would be very meaningful. Unfortunately, there is no consensus on how to define literacy, nor is there a one-size-fits-all way to assess it (Miller 1988). There are other meaningful indicators that are not feasible for a neighborhood-level index. For example, the Housing and Homeless Coalition would like to publish their data, and homelessness is clearly associated with poverty. However, because these populations are transient, we cannot convert non-existent addresses to census tracts. Lastly, we are still considering the best method to create a public-facing data product. We have considered various open source and proprietary tools. Even though our data and processes are entirely public, the index needs a clean, user-friendly, and online interface to further increase its accessibility. We are hopeful that these and other improvements will ensure a meaningful, manageable, and moveable index that is community-owned, community-operated, and capable of helping our city in addressing even the most wicked of problems.

84

J. Crawford and F. Ridzi

References Ackoff, R. L. (1974). The systems revolution. Long Range Planning. Vol. 7, Issue 6. December, 1974. Retrieved on 9 March 2019 from https://www.sciencedirect.com/science/article/abs/pii/ 0024630174901277?via%3Dihub. Adolphsen, S. (2018). Three myths about the welfare cliff. The Foundation for Government Accountability. Annie E. Casey Foundation. (2010). Early warning! Why reading by the end of third grade matters. Barba, L. A. (2018). Terminologies for reproducible research. The George Washington University. Retrieved on 8 March 2019 from https://arxiv.org/pdf/1802.03311.pdf. Bitsko, R., et al. (2016). Health care, family, and community factors associated with mental, behavioral, and developmental disorders in early childhood. Center for Disease Control: Morbidity and Mortality Weekly Report (MMWR). Retrieved on 9 March 2019. Bureau of Labor Statistics. (2019). Schedule of news releases and full data availability for county employment and wages. Retrieved on 30 January 2019 from https://www.bls.gov/cew/ releasecalendar.htm. Center for Applied Internet Data Analysis (CAIDA). (2019). Promotion of data sharing. Retrieved on 8 March 2019 from https://www.caida.org/data/sharing/. Center for Disease Control. (2013). Childhood lead poisoning. Factsheets. Center for Public Education, Black Council of School Board Members, Hispanic Council of School Board Members, National Caucus of American Indian/Alaska Native School Board Members, & Council of Urban Boards of Education. (2015). Why third-grade is a pivotal year for mastering literacy. Retrieved on 11 March 2019 from http://www.centerforpubliceducation.org/system/ files/Leading%20to%20Read%20%28Full%20report%29_0.pdf. Chetty, R. et al. (2016). Effects of exposure to better neighborhoods on children. American Economic Review. Chetty, R., Hendren, N., Kline, P., & Saez, E. (2014). Where is the land of opportunity? The geography of intergenerational mobility in the United States. The Quarterly Journal of Economics, 129(I4), November 2014. CNY Vitals Pro. (2019). Census Tract 5.01: Poverty. Syracuse location profile. Retrieved on 26 March 2019 from https://pro.cnyvitals.org/profile/census-tract-5.01#introduction. Congressional Budget Office. (2015). Effective marginal tax rates for low- and moderate-income workers in 2016. Cooper, K. R., & Shumate, M. (2015). Sharing data in collective impact efforts. Stanford Social Innovation Review. Retrieved on 10 March 2019 from https://ssir.org/articles/entry/sharing_data_ in_collective_impact_efforts. Crawford, J. (2018a). Syracuse crime analysis. Retrieved on 30 January 2018 from https://github. com/jamisoncrawford/Syracuse-Crime-Analysis. Crawford, J. (2018b). Syracuse poverty index. Github. Retrieved on 17 March 2019 from https:// github.com/jamisoncrawford/index. Croom, A. M., Ciavarri, P., Dancil, D., Lane, A., Last, A., King, P., & Crawford, J. (2019). Building equity in the construction trades: A racial equity impact statement. Urban Jobs Task Force & Legal Services of Central New York. Retrieved on 20 March 2019 from https://issuu.com/lscny/ docs/reisfinal_3-8-19_compressed. Desmond, M. (2016). Evicted: Poverty and profit in the American City. New York: Crown. Dorfman, J. (2016). Welfare offers short-term help and long-term poverty, Thanks to asset tests. Forbes. Retrieved on 30 January 2019 at https://www.forbes.com/sites/jeffreydorfman/2016/10/ 13/welfare-offers-short-term-help-and-long-term-poverty/#24f6610832cd. Edelstein, S. (2017). Syracuse neighborhoods. Syracuse Open Data. Retrieved on 30 January 2019 from https://data.syrgov.net/datasets/4df344e3ad8f45a8b5625967c7b3bbe3_0. Elliot, J. R. (1999) Social isolation and labor market insulation: Network and neighborhood effects on less-educated urban workers. The Sociological Quarterly, 40(2).

5 Meaningful, Manageable, and Moveable: Lessons …

85

Ellis, S. E., & Leek, J. T. (2017). How to share data for collaboration. Peer J Preprints. Retrieved on 9 March 2019 from https://peerj.com/preprints/3139v5.pdf. Empire State Development. (2015). The upstate revitalization initiative. Retrieved on 30 January 2019 from https://esd.ny.gov/about-us/signature-projects/upstate-revitalization-initiative. Every Action, & Nonprofit Hub. (2016). The state of data in the nonprofit sector. Nonprofit Hub. Retrieved on 9 March 2019 from file:///C:/Users/jamis/Downloads/Data-Survey-Whitepaper.pdf. Federal Bureau of Investigation. (2004). Crime in the United States: 2004. Uniform Crime Reports. Fontenot, K., et al. (2018). Income and poverty in the United States: 2017. September, 2018. Friedman, M. (2005). Trying hard is not enough: How to produce measurable improvements for customers and communities. Victoria: Trafford Publishing. Galster, George, et al. (2006). The social costs of concentrated poverty: Externalities to neighboring households and property owners and the dynamics of decline. Cambridge: Joint Center for Housing Studies, Harvard University. Geier, D. A., Kern, J. K., & Geier, M. R. (2017). Blood lead levels and learning disabilities: A crosssectional study of the 2003–2004 National Health and Nutrition Examination Survey (NHANES). International Journal of Environmental Research and Public Health. Retrieved on 9 March 2019 from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5664703/. Gump, B. B., Dykas, M., MacKenzie, H. A., Dumas, A. K., Hruska, B., Ewart, C. K., Pearsons, P. J., Palmer, C. D., & Bendinskas, K. (2017). High normal lead and mercury exposures: Psychological and behavioral problems in children. Environmental Research. Retrieved on 19 March 2019 from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5562507/. Head, B. W. (2008). Wicked problems in public policy. Public Policy, 3(2). Retrieved on 9 March 2018 from https://search.informit.com.au/documentSummary;dn=662880306504754; res=IELFSC. Jargowsky, P. A. (2013). Concentration of poverty in the new millennium. The Century Foundation & Rutgers Center for Urban Research and Education. December 18, 2013. Jargowsky, P. A. (2015). The architecture of segregation. The Century Foundation. August 9, 2015. Kania, J., & Kramer, M. (2011). Collective impact. Stanford Social Innovation Review. Kelling, G., & Wilson, J. Q. (1982). Broken Windows: The police and neighborhood safety. The Atlantic. Kimenyi, M. S. (2008). Rational choice, culture of poverty, and the intergenerational transmission of welfare dependency. Southern Economic Association. Kotter, J. P. (1995). Leading change: Why transformation efforts fail. Harvard Business Review. May–June 1995 Issue. Retrieved on 11 March 2019 from https://hbr.org/1995/05/leading-changewhy-transformation-efforts-fail-2. Loewenstein. (2005). Statistical, identifiable and iconic victims and perpetrators. Stanford Law and Economics Olin Working Paper No. 30. Retrieved on 9 March 2019 from https://papers.ssrn.com/ sol3/papers.cfm?abstract_id=678281. Ludwig, J. (2001). Urban poverty and educational outcomes. Brookings-Wharton Papers on Urban Affairs. Maslow, A. (1943). A theory of human motivation. Psychological Review, 50. Miller, G. A. (1988). The challenge of universal literacy. Science, 241(4871), 1293. Olinsky, B., & Post, S. (2013). Middle-out mobility. Center for American Progress. Retrieved on 30 January 2019 from https://www.americanprogress.org/issues/economy/reports/2013/09/ 04/73285/middle-out-mobility/. Onondaga County Health Department. (2019). Retrieved on 26 March 2019 from http://www.ongov. net/health/lead/data.html. Quillian, L. (2017). Poverty, neighborhood, and school setting. Focus, 33(2) (Spring/Summer 2017). Responsible Data Forum. (2016). Sharing data: What to do with your processed data. The hand-book of the modern development specialist. Retrieved on 8 March 2019 from https://responsibledata. io/resources/handbook/chapters/chapter-02c-sharing-data.html.

86

J. Crawford and F. Ridzi

Reuben, A. (2017). Association of childhood blood lead levels with cognitive function and socioeconomic status at age 38 years with IQ change and socioeconomic mobility between childhood and adulthood. Dunedin Multidisciplinary Health and Development. Schramm, C. (2015). When economic policy isn’t focused on growth, the poor suffer mightily. Forbes. Retrieved on 18 March 2019 from https://www.forbes.com/sites/carlschramm/2013/06/ 10/when-economic-policy-isnt-focused-on-growth-the-poor-suffer-mightily/#1abf04a36987. Simpson, E. H. (1951). The interpretation of interaction in contingency tables. Journal of the Royal Statistical Society, 13(2), 238–241. Stafford, M. (2006). Association between fear of crime and mental health and physical functioning. American Public Health Association. Swarup, V., Seligman, L., & Rosenthal, A. (2006). A data sharing agreement framework. International conference on information systems security. Retrieved on 8 March 2019 from https://link. springer.com/chapter/10.1007/11961635_2. Syracuse Open Data. (2015). Syracuse TNT Areas. Syracuse Open Data. Retrieved on 30 January 2019 from http://data.syrgov.net/datasets/9daae89158654684857e227d40bec262_0. Syracuse.com. (2019). Reported crimes in syracuse and onondaga county. Retrieved on 26 March 2019 from https://www.syracuse.com/crime/page/police_reports.html. Tomorrow’s neighborhoods today. (2019). Sectors. Retrieved on 30 January 2019 from http://www. tomorrowsneighborhoodstoday.org/. Troy, A., Nunery, A., & Grove, J. M. (2016). The relationship between residential yard management and neighborhood crime: An analysis from Baltimore City and County. Landscape and Urban Planning, 147, 78–87. U.S. Census Bureau. (2014a). Poverty: the history of a measure. Measuring America. January, 2014. U.S. Census Bureau. (2014b). The history of the official poverty measure. Income & Poverty. U.S. Census Bureau. (2019a). “Poverty thresholds”. Retrieved on 5 May 2019 from https://www. census.gov/data/tables/time-series/demo/income-poverty/historical-poverty-thresholds.html. U.S. Census Bureau. (2019b). 2020 Census. Our Surveys & Programs. Retrieved on 30 January 2019 from https://www.census.gov/programs-surveys/decennial-census/2020-census.html. U.S. Census Bureau. (2019c). Geographic terms and concepts—Block groups. Retrieved on 30 January 2019 from https://www.census.gov/geo/reference/gtc/gtc_bg.html?cssp=SERP. U.S. Census Bureau. (2019d). Welcome to Geocoder. Geocoder. Retrieved on 30 January 2019 from https://geocoding.geo.census.gov/. U.S. Census Bureau. (2019e). When to use 1-year, 3-year, or 5-year estimates. Guidance for Data Users. Retrieved on 30 January 2019 from https://www.census.gov/programs-surveys/acs/ guidance/estimates.html. U.S. Department of Education. (2015a). The family educational rights and privacy act: Guidance for reasonable methods and written agreements. Privacy Technical Assistance Center. Retrieved on 10 March 2019 from https://studentprivacy.ed.gov/sites/default/files/resource_document/file/ Guidance_for_Reasonable_Methods%20final_0.pdf. U.S. Department of Education. (2015b). Written agreement checklist. Privacy Technical Assistance Center. Retrieved on 10 March 2019 from https://studentprivacy.ed.gov/sites/default/files/ resource_document/file/Guidance_for_Reasonable_Methods%20final_0.pdf. U.S. Department of Education. (2016). Data-sharing tool kit for communities: How to leverage community relationships while protecting student privacy. Privacy Technical Assistance Center. Retrieved on 10 March 2019 from https://www2.ed.gov/programs/promiseneighborhoods/ datasharingtool.pdf. U.S. Department of Health and Human Services. (2017a). HIPAA privacy rule and sharing information related to mental health. Office for Civil Rights. Retrieved on 8 March 2019 from https://www. hhs.gov/sites/default/files/hipaa-privacy-rule-and-sharing-info-related-to-mental-health.pdf. U.S. Department of Health and Human Services. (2017b). Permitted uses and disclosures: Exchange for health oversight activities. Office for Civil Rights. Retrieved on 8 March 2019 from https://www.healthit.gov/sites/default/files/phi_permitted_uses_and_ disclosures_fact_sheet_012017.pdf.

5 Meaningful, Manageable, and Moveable: Lessons …

87

Vaughan, J. (2019). Data.gov shutdown shows limits of open data. Tech Target. Retrieved on 30 January 2019 from https://searchdatamanagement.techtarget.com/news/252456184/Datagovshutdown-shows-limits-of-open-data. Violence Policy Center. (2017). The relationship between community violence and trauma. July, 2017. Vivier, P. (2010). The important health impact of where a child lives. Maternal and Child Health Journal, 15(8), 1195–202 (October 2010). Wickham, H. (2014). Tidy data. Journal of Statistical Software, 59(10). Retrieved on 9 March from https://www.jstatsoft.org/article/view/v059i10. Wilson, W. J. (1987). The truly disadvantaged. Chicago: University of Chicago Press.

Jamison Crawford Formerly based in Syracuse, New York, Jamison is an Institutional Research Associate at The Graduate School and the Center for the Advancement of Students and Alumni (CASA) at Georgia State University, Instructor at Georgia State University’s Andrew Young School, Faculty Associate at Arizona State University’s Watts College of Public Service and Community Solutions, and an independent consultant for nonprofit and philanthropic organizations. Since 2017, he has worked as a contractor for the Central New York Community Foundation and its constituents in advancing data literacy for social sector professionals, promoting organizational data maturity, and developing data analytic tools and products for local social sector organizations. He has led the organization, facilitation, and instruction of over one hundred local data user groups and applied data science sessions, and continues to manage the development of the award-winning community indicators visualization platform, CNY Vitals Pro. In 2019, he co-authored Data Science for Public Service (DS4PS), an open source textbook on applied data science for the Master of Science in Program Evaluation and Data Analytics (MS-PEDA) program at the Watts College. Jamison holds a Bachelor of Arts in Political Science from Niagara University, and a Master of Public Administration (M.P.A.) from Syracuse University’s Maxwell School of Citizenship and Public Affairs. Frank Ridzi Dr. Ridzi is Vice President for Community Investment at the Central New York Community Foundation, Associate Professor of Sociology at Le Moyne College and President of the Community Indicators Consortium. Frank has helped to launch and lead community initiatives in areas such as increasing community literacy, reducing lead poisoning and addressing poverty and economic inclusion. He has been involved in launching Community Indicators efforts and has conducted research and written in the areas of collective impact, sociology of work, social policy and philanthropy. His writings have appeared in such places as the Foundation Review, the Journal of Applied Social Sciences, the Journal of Organizational Change Management, and Review of Policy Research. He is a past President of the Literacy Funders Network, an affinity group of the Council on Foundations. Frank holds a Masters Degree

88

J. Crawford and F. Ridzi in Public Administration and a Ph.D. in Sociology from Syracuse University’s Maxwell School. He also carries a Certificate of Advanced Study in Women’s Studies. Prior to joining the Community Foundation, he served as Director for the Center of Urban and Regional Applied Research at Le Moyne College, where he still serves as Associate Professor of Sociology.

Chapter 6

The Development of DISC (Decision Integration for Strong Communities): An Agile Software Application of Sustainability Indicators for Small and Rural Communities Kevin Summers, Viccy Salazar, Dave Olszyk, Linda Harwell, and Allen Brookes Abstract Small towns and rural communities, like many larger cities, are looking for ways to strengthen their economies, improve quality of life, and build on local assets. They are seeking to create their own path to sustainability that fits their size, geography, and resources and also preserves their distinctive characteristics. To address this need, the Decision Integration for Strong Communities (DISC) application was developed to offer relevant information curated to assist smaller communities with addressing their sustainability goals. To ensure DISC is useful to a variety of smaller and rural communities, EPA project staff reached out to small numbers of interested communities and potential users at different steps of the development process. In late 2019, the DISC application was beta tested by potential users and is now downloadable upon request to collaborating communities.

Disclaimer: The views expressed in this manuscript are those of the authors and do not necessarily represent the views or policies of the U.S. Environmental Protection Agency. Any mention of trade names, products, or services does not imply an endorsement by the U.S. Government or the U.S. Environmental Protection Agency. The EPA does not endorse any commercial products, services, or enterprises. K. Summers (B) · L. Harwell US EPA, Office of Research and Development, CEMM/GEMMD, 1 Sabine Island Drive, Gulf Breeze, FL 32561, USA e-mail: [email protected] V. Salazar US EPA, Region 10, 1200 6th Avenue, Seattle, WA 98101, USA D. Olszyk · A. Brookes US EPA, Office of Research and Development, CPHEA/PESD, 200 SW 35th Street, Corvallis, OR 97333, USA © Springer Nature Switzerland AG 2020 F. Ridzi et al. (eds.), Community Quality-of-Life Indicators, Community Quality-of-Life and Well-Being, https://doi.org/10.1007/978-3-030-48182-7_6

89

90

K. Summers et al.

6.1 Introduction The intersection where the natural, built and social environments meet is where a “place” becomes a “home.” As a community changes, challenges arise that may threaten the quality of life characteristics most valued. From creating jobs to building greener infrastructure, many considerations are needed to identify local investment options that may prove most beneficial to a community over time. Assessing the status of those characteristics and the local amenities that can influence them is often the first step toward finding solutions to resolve community issues. Like many metropolitan cities, small towns, rural communities and culturally-centered neighborhoods are seeking ways to strengthen their economies, build on local assets, and improve their quality of life. However, the progress of smaller, more rural communities moving toward these goals is often hampered by a lack of accessible and integrated information. The U.S. Environmental Protection Agency (EPA) conducted a series of engagement workshops to learn more about ways to help smaller communities (~1000–50,000 population) plan for future growth. Community leaders agreed that access to appropriately scaled and easy to use data-driven tools that are responsive to the planning information needs of small towns could potentially help them identify sustainability issues, target potential improvement projects and monitor outcomes over time. Sustaining community development is beyond the capacity of any single governmental sector—whether local, regional or national or the myriad of agencies within each of these levels (Dale 2001). Municipal governments are on the front line of implementing sustainable community development, tasked with creating economic opportunities while preserving those distinct characteristics that make a community a desirable place to live. Maintaining community sustainability is a holistic endeavor and requires commitments for multiple government agencies to cooperate, align incentives and integrate policy approaches. One such approach has been the Integrated Community Sustainability Planning (Ling et al. 2009) which combines visioning, best practices, learning from other communities, incorporation of ecological limits, community engagement, and community mapping techniques. Past sustainable development concepts have focused on incorporation of ecological principles (McHarg 1969; Selman 1993; Register 2006; Wackernagel et al. 2006); participatory planning and community engagement (Healey 1997; Wates 2000; Tippet 2005); and, social impacts of urban design (Congress of the New Urbanism 1999; Groenewegen et al. 2006). Over 96% of incorporated places in the United States have populations of less than 50,000 people with 85% having populations of less than 10,000 (Duffin 2019). All of these small communities have common concerns related to economic stability (Clark et al. 2018; Capello and Nijkamp 2019), housing and public services (Bennett et al. 2015; Einstein and Glick 2017), healthcare (Douthit et al. 2015; Burstin et al. 2016), community identity (Ellis and Abdi 2017; Sisneros-Kidd et al. 2019), natural assets (Koirala et al. 2016) and natural hazards (Summers et al. 2017, 2018). Issues of growth, affordability, opportunity, preservation, restoration, cultural fulfillment

6 The Development of DISC (Decision Integration for Strong Communities) …

91

and preparedness and recovery capacity are major agenda items for all communities. However, each community has a unique order of importance of these factors that comprises the community’s value structure or identity (Smith et al. 2013; Fulford et al. 2015). Our development of a small community sustainability software application— Decision Integration for Strong Communities (DISC)—focuses on the integration of all these issues into a single software application. The DISC application integrates raw measures (e.g., median income, disease prevalence) to help community stakeholders interpret patterns observed in the data through indicators. Result ing indicators characterize qualities that represent or support community vitality, health, resilience, and well-being. Using an Agile process (Cohen and Ford 2003), the DISC software includes features that incorporate contributed data, account for local values, build investment-case scenarios, and supply resource information from multiple federal and state agencies, NGOs, and municipal sources. While the DISC application can be used by communities of all sizes, the intended audience is smaller communities (populations of ~50,000 or fewer) where information may be sparse or planning capacity is more limited.

6.2 Methods 6.2.1 Agile Process Agile software development is one of the newest approaches in the software methodology field and represents a major departure from traditional, plan-based approaches to software engineering. Agile is an umbrella term that includes multiple methods, techniques and approaches that employ short iterations and continuous customer feedback so that the software product can evolve into the customer need (Moreira 2010). What the customer sees as progress is not the standard reports and presentations but, rather, changed tangible working product functionality. Agile approaches are a major departure from traditional top down decision making seen in many, especially governmental, organizations. Agile approaches provide a means of adapting quickly to unpredictable and rapidly changing requirements. They provide options whereby the customer gets early and periodic views of potential solutions to problems. Agile approaches move projects and their teams from a hierarchical command and control world to a heterarchical self-empowering world, moving decision making to lower than typical levels. Agile approaches employ collaborative control which can be defined as having the right people making decisions, not just because a person has a title, but because that person has the best insight into the change, opportunity or problem. Part of the DISC project was to assess whether an Agile approach could be used successfully to complete an EPA software development project, in a relatively short

92

K. Summers et al.

time period (